Tech Trends Report 2024 Future Today Institute.pdf

2024
TECH TRENDS
REPORT
1 7 T H E D I T I O N

TABLE OF CONTENTS
003 LETTER FROM FTI’S CEO
004 2024 TECH TREND REPORTS
005 IMPACTS OF TRENDS ON INDUSTRY
006 EXECUTIVE SUMMARY
050 ARTIFICIAL INTELLIGENCE
166 WEB3
207 METAVERSE
257 BIOENGINEERING
328 ENERGY CLIMATE
401 MOBILITY, ROBOTICS DRONES
456 COMPUTING
517 BUILT ENVIRONMENT
572 NEWS INFORMATION
615 HEALTH CARE MEDICINE
687 FINANCIAL SERVICES INSURANCE
733 SPORTS
776 SPACE
840 HOSPITALITY
882 SUPPLY CHAIN LOGISTICS
919 ENTERTAINMENT
974 AUTHORS
975 ABOUT FUTURE TODAY INSTITUTE
976 METHODOLOGY
977 DISCLAIMER
978 USING SHARING THE TREND REPORT
FUTURE TODAY INSTITUTE • 2024 TECH TRENDS REPORT
2 © 2024 Future Today Institute. All Rights Reserved.

2024
The theme for our 2024 re-
port is Supercycle. In eco-
nomics, a “supercycle” refers
to an extended period of
booming demand, elevating
the prices of commodities
and assets to unprecedented heights. It stretches across
years, even decades, and is driven by substantial and
sustained structural changes in the economy.
We believe we have entered a technology supercycle.
This wave of innovation is so potent and pervasive that
it promises to reshape the very fabric of our existence,
from the intricacies of global supply chains to the minu-
tiae of daily habits, from the corridors of power in global
politics to the unspoken norms that govern our social
interactions.
Driving this seismic shift are the titans of technology
and three of their inventions: artificial intelligence, bio-
technology, and a burgeoning ecosystem of interconnect-
ed wearable devices for people, pets, and objects. As they
converge, these three macro tech segments will redefine
our relationship with everything, from our pharmacists to
our animals, from banks to our own bodies. Future Today
Institute’s analysis shows that every technology—AR/ VR/
XR, autonomous vehicles, low Earth orbit satellites, to
name a few—connects to the supercycle in some way.
The ramifications are stark and undeniable. As this tech
supercycle unfurls, there will be victors and vanquished,
those who seize the reins of this epochal change, and
those who are swallowed whole. For business leaders,
investors, and policymakers, understanding this tech
supercycle is paramount.
In this 17th edition of FTI’s annual Tech Trends report,
we’ve connected the supercycle to the nearly 700 trends
we’ve developed. Our research is presented across 16
technology and industry-specific reports that reveal the
current state of play and lists of influencers to watch,
along with detailed examples and recommendations de-
signed to help executives and their teams develop their
strategic positioning. The trends span evolutionary ad-
vancements in well-established technologies to ground-
breaking developments at the forefront of technological
and scientific exploration. You’ll see emerging epicenters
of innovation and risk, along with a preview into their
transformative effects across various industries.
We’ve visually represented the tech supercycle on the
report’s cover, which is an undulating image reminiscent
of a storm radar. Vertical and horizontal lines mark the
edges of each section’s cover. When all 16 section cov-
ers converge, the trends reveal a compounding effect as
reverberating aftershocks influence every other area of
technology and science, as well as all industries.
It’s the convergence that matters. In isolation, trends of-
fer limited foresight into the future. Instead, the interplay
of these trends is what reveals long-term change. For
that reason, organizations must not only remain vigilant
in monitoring these evolving trends but also in cultivat-
ing strategic foresight—the ability to anticipate future
changes and plan for various scenarios.
Our world is changing at an unprecedented rate, and this
supercycle has only just begun.
Amy Webb
Chief Executive Officer
Future Today Institute
THE YEAR AHEAD: TECH SUPERCYCLE
FUTURE TODAY INSTITUTE • 2024 TECH TRENDS REPORT

FUTURE TODAY INSTITUTE’S 2024 TECH TREND REPORT
Our 2024 edition includes nearly 700 trends, which are published individually in 16 volumes and as one comprehensive report with all trends included.
Download all sections of Future Today Institute’s 2024 Tech Trends report at http://www.futuretodayinstitute.com/trends.
2024 FUTURE TODAY INSTITUTE • 2024 TECH TRENDS REPORT

IMPACT OF TRENDS ON YOUR INDUSTRY
Long-Term Relevance
Near-Term Relevance
AI Generative AI Bioengineering
Generative
Bio
Computing
Architecture
AR/ VR/ XR
Synthetic
Media
Metaverse
Web3
Infrastructure
Mobility
Robots and
Drones
Climate and
Green Tech
Quantum
Agriculture
Automotive
Aviation and Travel
Construction, Engineering
Consumer Packaged Goods
Education
Financial Services
Government and Policy
Health Care Systems and Services
Hospitality
Media (Entertainment)
Media (News)
Pharmaceutical and Medical Products
Public and Social Sectors
Real Estate
Restaurants
Retail
Space and Aerospace Defense
Supply Chain and Logistics
Telecommunications
2024 FUTURE TODAY INSTITUTE • 2024 TECH TRENDS REPORT

EXECUTIVE
SUMMARY
1 7 T H E D I T I O N • 2 0 2 4 T E C H T R E N D S R E P O R T

EXECUTIVE SUMMARY
2024
2024
TRENDS
Trends are what we can know about today and are based on
data and evidence. This year’s trend report covers hundreds of
technology trends across various industries and categories.

EXECUTIVE SUMMARY
2024
ARTIFICIAL INTELLIGENCE
While the hype continues, now is the time to prepare.
AI is a transformative, general-purpose technology with
the potential to influence entire economies and alter
society. From multi-modal AI to self-improving agents
and wearables with on-board assistants, the ecosystem
is rapidly changing. While AI advancements promise
to reshape our world, emerging trends highlight
unprecedented risks, underscoring the need for
preparedness, governance, and alignment.
BIOENGINEERING
Brace for impact in this breakthrough year.
Recent breakthroughs have accelerated the
convergence of biology, information systems, and
advanced platforms, and attention has been focused
on generative AI. Very soon, that focus will shift to
generative biology, where AI models will lead to the
creation of novel molecules, drugs, materials, and living
organisms. While biotech promises to reshape our
world, it also requires preparedness and governance.
WEB3
Quiet developments and new regulations grow Web3.
While the industry reeled from failures and
bankruptcies over the last year, developers have
continued to develop new capabilities and features,
moving ever closer to real-life practical applications
for Web3. As regulations are on the horizon to be
implemented in 2024, this could be a year of inflection,
bringing Web3 from the experimental and theoretical to
cold hard business cases.
ENERGY CLIMATE
Emerging and tested tech comes together for progress.
As the effects of climate change become more dire,
governments are considering a more exotic set of tech-
nologies to combat the situation. Solar geoengineering,
ocean chemistry manipulation, and domed cities are
all concepts that gained significant traction in 2023.
Meanwhile, significant gains have been made in en-
abling infrastructure for renewable energy, focusing on
smart grids, energy storage, and carbon tracking.
METAVERSE NEW REALITIES
Developments indicate a possible inflection point.
Apple’s new headset leverages AI to overlay digital onto
reality, representing a pivot from fantasy metaverses
to real-life applications that may signal maturity.
Meanwhile, complementary features to enhance virtual
experiences are being developed. Though device costs
are an ongoing barrier, integrating AI and XR creates
fundamental synergies, potentially marking the
watershed moment the industry has been waiting for.
MOBILITY, ROBOTS DRONES
Challenges remain for electrification and autonomy.
Consumers are adapting to electric and semi-
autonomous vehicles and those that collect increasing
data. At the same time, battery technology is increasing,
enabling vehicles, robots, and drones to perform longer.
The rise of these machines suggests a future where they
supplement and replace human tasks, highlighting a
shift towards a more efficient, increasingly monitored
work environment.
TRENDS

EXECUTIVE SUMMARY
2024
COMPUTING
AI is transforming human-computer interaction.
AI is changing what is possible in form factors,
challenging underlying computing architecture.
Researchers are pursuing energy-efficient architectures
by reverse engineering the complex biological efficiencies
of the human brain. As AI facilitates more intuitive
communication, the technology could blend more
seamlessly into human experiences, with computing
centered directly around people rather than devices.
HEALTH CARE MEDICINE
The barrier between digital and biological is vanishing.
The merging of digital and biological worlds enables a
whole new range of treatments, the most exciting being
cells within our bodies that can produce medication
in response to external stimuli. Conversely, cyberbio-
malware creates new, existential threats to our health
that we are unaware we need to protect ourselves
against. Both developments have the potential to upend
the pharmaceutical and healthcare industry completely.
BUILT ENVIRONMENT
Automation and data collection transform practices.
In an industry used to following traditional practices,
the past years have upended decade-old norms. Such
shifts can provide critical solutions to new questions
and signify a turning point that redefines industry
standards and operational efficiency. This pivotal
moment underscores the need for strategic adaptation,
heralding a reinvention phase in response to evolving
demands and technological advancements.
FINANCIAL SERVICES INSURANCE
Modernization is slow, but consumers are ready to run.
The financial services sector, dominated by legacy
giants, is facing a critical juncture where embracing
technology like open banking, digital identity, and
blockchain is not just advantageous but imperative.
This industry must shift from reactive to proactive,
underscoring the importance of anticipation and
preparedness in navigating the future of the financial
and insurance industries.
NEWS INFORMATION
The news ecosystem hits an inflection point.
Emerging technologies like generative AI are shaping
the future of content creation, distribution, and mone-
tization. New applications of AI are reshaping the media
value change and forging new consumer behaviors for
information search and discovery. In the year ahead, the
initial frenzy of ChatGPT prompt hacking and product
launches based on large language models will fade, but
the information ecosystem will never be the same.
SPORTS
Analytics and customization are transforming sports.
Technology is enhancing the capabilities of managers,
coaches, athletes, and fans. Teams and leagues are
seeing progress in scouting, training, performance
analytics, and rehabilitation thanks to tools like mixed
reality, computer vision, and AI. Smarter stadiums offer
unique, immersive experiences, driving engagement and
revenue streams that support further improvements in a
game’s quality and spectator experiences.
TRENDS

EXECUTIVE SUMMARY
2024
SPACE
Space exploration is entering a new era.
Defined by old geopolitical rifts and new spacefaring
nations, this emerging era goes beyond past dynamics
to involve a broader constellation of smaller nations
and private enterprises, all enabled by the decreasing
cost of space access. With lowered launch costs, zero
gravity could unlock scientific discoveries previously
impossible. The dream of a multi-planetary humanity
stirs deep questions about our priorities.
SUPPLY CHAIN LOGISTICS
Real-time data and instant everything shape new terrain.
Regional instability, materials disruptions, manufac-
turing relocation, and labor tensions create hardship
for businesses and manufacturers trying to bring their
goods to consumers, who are also becoming more de-
manding. As businesses strive to meet these demands,
they will be compelled to rethink their strategies, tech-
nologies, and workflows, marking a pivotal moment in
the report on industry trends.
HOSPITALITY RESTAURANTS
Tech-driven efficiency can’t replace personalization.
Contactless payments, immersive pre-experiences,
augmentations that recognize a guest’s personalized
needs upon entering the premise, automated back-
of-house functions offer opportunities for owners and
operators to capitalize on that can provide elevated
experiences and a reduction in overhead costs—but
human workers might hold these positions, driving an
ongoing tension between efficiency and a personal touch.
ENTERTAINMENT
AI’s impact on live entertainment is just beginning.
Taylor Swift’s concert film demonstrated that secondary
live experiences resonate deeply. Add to that haptics,
holographic transmission in real-time, and climate
considerations that have consumers looking for
local options, and a whole new world of performance
experiences opens up. These developments create
completely new ways to engage with content and
opportunities for new business models.
TRENDS

EXECUTIVE SUMMARY
2024
2024
UNCERTAINTIES
Uncertainties represent what we cannot know—and identifying
them can reduce the risk of blind spots down the road. In this year’s
report, we highlight five top uncertainties that will shape 2024.

EXECUTIVE SUMMARY
2024
UNCERTAINTIES
How quickly will
AI revolutionize
business, and in
what ways?
What global
challenges will
bioengineering be
able to address?
How will countries
and businesses
commit to
decarbonization?
How will humans
prioritize their
work environments
in the future,
and what will
these preferred
workplaces look
like?
In the context
of increasing
protectionism
and geopolitical
tensions, what
is the future of
the global chip
manufacturing
industry?
JUMP TO UNCERTANTITY 1 JUMP TO UNCERTANTITY 2 JUMP TO UNCERTANTITY 3 JUMP TO UNCERTANTITY 4 JUMP TO UNCERTANTITY 5
01 02 03 04 05

01
How quickly will AI revolutionize
business, and in what ways?
EXECUTIVE SUMMARY
2024

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
How quickly will AI revolutionize business,
and in what ways?
Stratification Scales
by Function or Task
Stratification by Industry AI Scales Up for Consumers
Efficiency Optimization
in the Back Office
Bolstered Productivity
and Performance
Cautious Scaling in
Complex Industries
Agile Scaling in Less-
Regulated Industries
AI Serves as a
Creative Playground
A Revolutionary
Consumer-Facing
Assistant
GET THE DETAILS GET THE DETAILS GET THE DETAILS

SCENARIO 1 SCENARIO 2
UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Stratification Scales by Function or Task
AI scales along two distinct trajectories: by function or by task. Functional scaling involves
embedding AI into key business operations, like accounting or IT, to enhance productivity.
Conversely, task-based scaling increases efficiency for routine activities like email, leading to
workforce streamlining. Both offer unique advantages for leaders, who must choose between
broad operational transformation or targeted task refinement.
Efficiency Optimization in the Back Office
AI emerges as a linchpin, enhancing efficiency for back-office operations like
calendaring and leading to a cascade of broader organizational and cultural
shifts. But there are human capital implications: As AI takes root, the workforce
gradually contracts.
With efficiency gains, companies reconsider budgets, diverting funds from labor
to tech. This shift fosters a culture of innovation, as employees are reassigned to
more strategic, creative, or complex tasks that AI can’t easily replicate. But it also
will require companies to reevaluate talent strategies, prioritize adaptability and
technical proficiency, and expand HR departments to include upskilling programs
and change management.
Bolstered Productivity and Performance
Integrating AI into the business ecosystem results in meaningful workforce
reduction, as the tech can tackle work that previously required multiple humans.
Management adapts to overseeing hybrid teams of humans and AI, fostering
collaboration and redefining performance metrics. Money previously allocated
for salaries and benefits is redirected toward technology investment, research,
and development, potentially driving further innovation. Ultimately, AI leads
to more informed decision-making, risk assessment, and predictive analytics,
and businesses tailor their AI solutions to specific departmental needs to gain
competitive advantages.
1. HOW QUICKLY WILL AI REVOLUTIONIZE BUSINESS, AND IN WHAT WAYS?
NEXT SCENARIO | THREATS OPPORTUNITITES | STRATEGIC QUESTIONS | RETURN TO UNCERTAINTIES

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Stratification by Industry
As AI matures, its scalability pivots around industry-specific applications. The speed with
which AI scales across certain industries is a function of technological receptivity and reflects
the regulatory landscapes that govern them. Business leaders must carefully measure
tradeoffs between these two challenges, calibrating their AI integration strategies to align
with industry characteristics and regulatory confines.
Cautious Scaling in Complex Industries
Sectors like financial services and health care stand to gain improved productivity
and efficiencies as AI leaders by processing vast data streams and achieving
decision-making acumen beyond human speed. However, lagging regulatory
updates temper this progress.
Slower to replace the nuanced judgment of human experts with AI systems,
businesses take a more deliberate scaling strategy. They invest in AI RD to
create more sophisticated models built on equitable data training. Their cautious
approach extends to data privacy and security, as they implement strong
measures to safeguard sensitive information.
Agile Scaling in Less-Regulated Industries
Industries less encumbered by compliance hurdles—such as consumer
packaged goods and retail—are quick to capitalize on AI’s potential. It
becomes a dynamic catalyst for innovation, enabling rapid ideation and
deployment of new products and services.
AI’s ability to analyze vast amounts of data in real-time is leveraged to
personalize customer experiences, from tailored product recommendations
to enhanced service interactions. It also creates supply chain resilience by
predicting demand, optimizing inventory, and identifying potential disruptions.

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
AI Scales Up for Consumers
Two primary use cases of AI alter everyday life. The first is through creative and experimental applications
that introduce a sense of play and innovation to regular activities. They are reshaping entertainment, social
media, and online engagement by providing novel, personalized experiences. The second sees AI as a
practical assistant, seamlessly integrating into daily routines to manage tasks, offer recommendations,
and streamline decisions. This form of AI is becoming a dependable extension of personal efficiency. These
manifestations of AI highlight its versatility and growing significance in shaping consumer behavior.
AI Serves as a Creative Playground
For consumers, AI is more about novelty and entertainment than utility. Students
integrate generative AI into their projects, pushing the boundaries of creativity and
academic exploration. Social media enthusiasts use it to embellish their online
personas and captivate their followers.
Consumers now expect more interactive and personalized digital experiences
across all platforms, influencing a shift in how products and services are
designed and marketed. At the same time, the digital divide is more pronounced
as a broader segment of the population becomes adept at navigating and
manipulating digital environments and tools.
A Revolutionary Consumer-Facing Assistant
The AI assistant transcends its fledgling origins to become an essential element
of daily life. It simplifies complex tasks and executes functions with a level of
access and autonomy that mirrors human assistance. Consumers find tangible
ease in their routines as AI assistants manage calendars, make informed
purchases, and even anticipate needs before they arise.
The downside: The role of AI in decision-making processes raises ethical
questions, especially if these systems begin to influence areas like personal
relationships, employment opportunities, and even legal decisions.
THREATS OPPORTUNITITES | STRATEGIC QUESTIONS | RETURN TO UNCERTAINTIES

THREATS
1 Workforce disruption
As AI takes on more responsibilities, companies must
manage job losses and reskill employees to avoid
negative backlash.
2 Regulatory non-compliance
Strict governance in regulated sectors can curb AI
innovation/adoption if transparency, accountability,
and ethical standards are not upheld.
3 Data privacy issues
As AI leverages increasing amounts of consumer
data, rising public concern around data privacy may
warrant increased regulation and standards around
data use.
OPPORTUNITIES
1 Increased productivity and efficiency
AI implementation in back-office and core business
functions can result in exponential productivity gains,
superior efficiency, and cost reductions.
2 Enhanced data-driven decision-making
AI will provide business leaders with previously
inaccessible insights, predictive analytics, and risk
assessments to inform strategy and planning.
3 Competitive differentiation
Customized enterprise AI solutions present opportunities
to develop proprietary platforms, hyper-specialized
tools, and new business models that distinguish an
organization.
EXECUTIVE SUMMARY
2024 1. HOW QUICKLY WILL AI REVOLUTIONIZE BUSINESS, AND IN WHAT WAYS?
OPPORTUNITIES THREATS
STRATEGIC QUESTIONS | RETURN TO UNCERTAINTIES

EXECUTIVE SUMMARY
2024 1. HOW QUICKLY WILL AI REVOLUTIONIZE BUSINESS, AND IN WHAT WAYS?
Are you mainly focused
on leveraging AI in
your organization for
functional efficiency,
task-based productivity
gains, or both? What are
you doing to build those
strategies?
How will third-party
AI assistants change
consumer expectations
of your products and
services?
How might emerging
AI systems create a
compounding effect for
your business? Is there
a part of your business
that could be vastly
improved or optimized if a
component or compound
could be altered slightly?
How might your offerings
need to shift as consumer
and business buyer
expectations evolve?
Do you have the data
governance foundations
and AI accountability
measures to comply with
existing and emerging
regulations?
1 4
2 5
3
STRATEGIC QUESTIONS
RETURN TO UNCERTAINTIES

02
What global challenges will
bioengineering be able to address?
EXECUTIVE SUMMARY
2024

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
What global challenges will bioengineering
be able to address?
Better Living Infrastructure Resilience Securing Agricultural Infrastructure
Subscribe or Die Widening
Societal Gap
Self-Sustaining
Highways
(The End to Infrastructure Bills?)
World War Food Community-Focused
Food
The Never Retiring
Workforce

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Better Living
Bioengineering initiatives and research present a change in our approach to health care and even aging.
Researchers use generative biology to create new therapeutic compounds and test the viability of gene
editing. That work results in better drugs, personalized treatments, engineered tissues, and alternatives to
one-size-fits-some medicine. Experiments are underway to alter genetics to cure ailments and diseases, with
success already being achieved in hearing loss reversal and sickle cell treatment. Scientists leverage animal
alternatives as they work to enhance limb, organ, and joint regenerative capabilities.
Subscribe or Die
Health care shifts from reactively treating sickness and diseases to proactively
addressing potential health issues before they emerge. With the goal of making
up for any future lost “reactive” revenues, the largest health care systems use
MA to join the bioengineering landscape as facility providers and treatment
hubs. People subscribe to bio-based services, which they must use to continue
getting access to health-promoting and anti-aging solutions. Big Bio creates the
ultimate subscription package: Continue paying, or face an untimely alternative.
The Never Retiring Workforce
Economic conditions have not drastically improved for Millennials and Gen Z, who
cannot afford to retire. Thanks to preventative medicine, genetic surgeries, and
new therapies, the life expectancy of Americans spiked from 77 to 115 years. Many
older individuals feel physically and mentally capable of working indefinitely,
while younger generations face a bottleneck in career advancement.
The implications are profound, and not just in the US. HR professionals and labor
economists prompt a reevaluation of career trajectories, retirement planning, and
intergenerational equity, hoping to stave off a collapse.
2. WHAT GLOBAL CHALLENGES WILL BIOENGINEERING BE ABLE TO ADDRESS?

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Infrastructure Resilience
Our infrastructure is no longer crumbling and fragile. By creating new compounds and modifying
existing ones, bioengineering significantly enhances construction materials’ strength, durability,
and flexibility while reducing reliance on harmful chemicals and petroleum-based compounds.
Compounds that self-heal extend the lifespan of infrastructure by autonomously repairing damage.
New materials, designed with generative biology, result in “living” buildings engineered with
sustainable, organic materials.
Widening Societal Gap
As bioengineering technologies mature and scale, the demand for traditional
construction and maintenance services wanes. Construction firms that don’t
adapt— struggle, and those specializing in bioengineered solutions—thrive.
While new homes are increasingly constructed with these innovative materials,
integrating them into existing homes proves financially prohibitive for many
homeowners. A two-tier infrastructure landscape emerges, where newer
developments boast remarkable resilience and sustainability features, and older
constructions lag.
Self-Sustaining Highways (The End to Infrastructure Bills?)
Bioengineering introduces advanced materials for highways and bridges that
incorporate microorganisms, which produce substances to naturally repair
damages, thereby enhancing durability and slashing maintenance expenses.
These innovations adapt to fluctuating environmental conditions, offering better
resistance against extreme weather patterns and contributing to infrastructure
longevity. Some highway infrastructures mend themselves and reduce CO2 levels,
while bridges utilize embedded sensors for ongoing health monitoring, ensuring
safety and operational efficiency.

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Securing Agricultural Infrastructure
Bioengineering addresses the global food security challenge by revolutionizing how and where we
grow our food. Improved crop yields enable food cultivation in diverse and previously inhospitable
locations. Bioengineering also offers solutions to the most pressing agricultural challenges, such as
developing famine and disease-resistant crops, engineering drought-tolerant plants that thrive on
less water, and reducing reliance on chemical fertilizers and pesticides by creating all-natural pest
resistances. Crops now withstand shifting weather patterns, while soil health is improved.
World War Food
Wealthier nations, foreseeing the vulnerabilities of relying on traditional
agriculture, heavily invested in synthetic biology and food synthesis technologies to
produce engineered foods customized for nutritional content.
This shift has created a new divide in the global economy, with disparities in
food production capabilities. Agrarian countries, dependent on their exports of
traditional crops, face widespread unemployment and social unrest. They retaliate
through cyber sabotage and proxy conflicts, attempting to disrupt their wealthier
counterparts’ engineered food supply chains.
Community-Focused Food
Genetic modification and environmental adaptation have revolutionized
agriculture. Crops flourish in deserts and on urban rooftops, making food
production sustainable and accessible. The urban heat island effect is reduced,
and fresh produce and cultured proteins are available at local shops. Suburban and
rural communities have embraced cooperative farms growing bioengineered crops
resistant to various adversities, enhancing local biodiversity and fostering shared
responsibilities and benefits. Commercial farming has shifted toward sustainability,
with genetically engineered crops achieving faster growth and higher yields,
leading to reduced land use, shorter supply chains, and less reliance on imports.
Food is grown closer to where it is consumed, leveraging advanced technologies to
create a resilient, efficient, and eco-friendly food system.

THREATS
1 Genetic Privacy and Discrimination Risks
A rise in bioengineering raises concerns over genetic
data privacy. There’s a risk that insurers, employers, or
third parties could misuse individuals’ genetic data,
leading to new forms of discrimination.
2 Automation In Employment
Integrating bioengineering in various industries
may lead to significant job displacement. Traditional
agriculture, construction, and health care roles could be
transformed or made obsolete by automated processes.
3 Geopolitical Conflict
The strategic advantage gained through bioengineering
capabilities could become a source of global tension.
Leading nations may exert influence over others,
leading to new forms of dependency and inequalities.
OPPORTUNITIES
1 Adapting to the Never-Retiring Workforce
The increase in life expectancy will drive shifting career
dynamics, allowing businesses to develop new career
paths, flexible working conditions, and retirement plans
that accommodate an aging but active workforce.
2 Bridging the Gap in Health Access
Advancements in bioengineering offer unprecedented
opportunities to address health inequities globally through
more cost-effective production of pharmaceuticals.
3 Improved Food Access
Synthetic biology revolutionizes global food strategy by
boosting crop resilience, enhancing nutrition, producing
sustainable ingredients, developing alternative proteins,
and reducing waste, leading to more resilient food systems.
EXECUTIVE SUMMARY
2024 2. WHAT GLOBAL CHALLENGES WILL BIOENGINEERING BE ABLE TO ADDRESS?

EXECUTIVE SUMMARY
2024 2. WHAT GLOBAL CHALLENGES WILL BIOENGINEERING BE ABLE TO ADDRESS?
How can synthetic
biology be used to solve
previously unsolvable
problems in your product
RD pipeline?
How do you plan to
manage intellectual
property rights and
protect innovations in the
field of synthetic biology,
and what challenges do
you foresee in this area?
What skills and talent
do you need to leverage
synthetic biology in your
business effectively,
and how do you plan to
acquire them?
How will your customers
and the market respond
to products or services
developed through
synthetic biology?
What impacts do you
anticipate synthetic
biology will have on
your supply chain and
procurement strategies?
1 4
2 5
3
STRATEGIC QUESTIONS

03
How will countries and businesses
commit to decarbonization?
EXECUTIVE SUMMARY
2024

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
How will countries and businesses
commit to decarbonization?
Blue Economy
An Expanded Environmental
Support Ecosystem
Inclusive, Multifaceted
Environmental Protection
Sustainable Growth
and Innovation
Grid Optimization
Environmental
Efforts Stagnate
Holistic Benefits
Are Realized
Over-regulation
and Complexity
Dominate
Exploitation and
Environmental
Degradation

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Blue Economy
The blue economy, which promotes the sustainable use of ocean resources, plays an increasingly
important role in countries’ sustainability efforts by harnessing the ocean’s economic growth while
ensuring marine environmental health. This approach includes sustainable fisheries, renewable
marine energy, and eco-friendly tourism, all aimed at preserving ocean ecosystems and promoting
economic resilience.
Sustainable Growth and Innovation
Marine technology breakthroughs vastly improve sustainable fishing, ocean-based
renewable energy efficiency (like wave and tidal power), and ocean cleanup
efforts, significantly reducing marine pollution.
Concurrently, robust international agreements are instrumental in the effective
management of marine resources. These agreements introduce strict regulations
to prevent overfishing and protect marine habitats from destruction. Because of
these combined efforts, the health of the marine ecosystem improves.
Exploitation and Environmental Degradation
Inadequate global governance fuels overfishing, unregulated marine development,
and pollution, drastically diminishing ocean resources.
Wealthier nations disproportionately deplete these resources, leaving smaller
coastal communities to bear the brunt of environmental harm and resource scarcity.
Ignoring sustainable methods leads to a sharp decline in biodiversity, undermining
the blue economy’s growth, affecting worldwide food security, and disrupting
climate balance. These effects are most acutely felt by small coastal communities
reliant on the ocean for survival.
3. HOW WILL COUNTRIES AND BUSINESSES COMMIT TO DECARBONIZATION?

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
An Expanded Environmental Support Ecosystem
Countries expand their environmental efforts beyond a narrow focus on renewable energy to include
broader ecosystem support. This involves enhancing infrastructure such as power grids and EV
charging networks, implementing environmental, social, and governance (ESG) policies, and bringing
greater transparency to carbon trading schemes. The aim is to take a more holistic approach to
sustaining healthy ecosystems while still continuing the transition toward clean energy.
Grid Optimization
Energy grids have been upgraded through deployment of advanced energy
storage systems, such as large-scale batteries and pumped hydro facilities. We
now store excess renewable energy. There is widespread adoption of highly
efficient electric heat pumps for heating and cooling to reduce energy use and
curb emissions. New ESG policies include transparency for CO2 trading schemes.
Businesses supporting optimized grid initiatives benefit from reduced energy
costs thanks to improved stability and more energy-efficient technologies. These
businesses also tap into emerging renewable energy markets, creating new
revenue streams. Those leading the transition to sustainable power grids have a
first-mover advantage.
Environmental Efforts Stagnate
Persistent power outages and insufficient electric vehicle infrastructure disrupt
economic activities, causing a dip in productivity and risking job losses. Since
ESG regulations vary across regions, compliance is difficult. This leads to higher
operational expenses and hampered economic expansion. A lack of transparency
in carbon trading markets breeds investor skepticism, obstructing the development
of green finance.

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Inclusive, Multifaceted Environmental Protection
The concept of environmental protection has evolved to become more inclusive. Countries and
businesses now recognize that it encompasses efforts to mitigate climate change and initiatives for
biodiversity conservation and the promotion of humane working conditions. This broader approach
integrates ecological sustainability with social responsibility, reflecting a more holistic understanding
of our interconnected world.
Holistic Benefits Are Realized
Nations prioritize the conservation of diverse ecosystems, leading to increased
ecotourism, research opportunities, and sustainable resource management.
Responding to labor group advocacy, businesses adopt fair labor practices,
including raising wages and prioritizing safer work environments.
Companies embrace comprehensive sustainability strategies that integrate
ecological, social, and economic considerations, leading to significant reductions
in resource waste and improved financial resilience. These strategies are effective
because they strike a balance between environmental consciousness and business
viability, aligning with consumer preferences for environmentally responsible
goods and services.
Over-regulation and Complexity Dominate
Expanded environmental protection efforts bring complex and often competing
regulations to the forefront, and governments have to wrestle with noncompliance
and enforcement challenges. Businesses face exorbitant costs navigating this
convoluted regulatory landscape, stifling innovation and hindering economic
growth. The high cost of transitioning to environmentally friendly legislation poses
barriers to entry for smaller businesses, reducing market competition.
In striving to meet a multitude of environmental goals, overemphasis on specific
areas like biodiversity conservation diverts finite resources from urgent climate
mitigation needs, impeding overall sustainability progress.

THREATS
1 Compliance Complexity
Diverse and changing ESG regulations across regions
can complicate compliance efforts, increasing
operational costs and reducing competitiveness.
2 Technological Obsolescence
Rapid advancements in energy technologies could
render existing products and services obsolete,
posing a threat to businesses unable to innovate
quickly.
3 Implementation Costs
The costs associated with adopting comprehensive
sustainability strategies could be significant,
particularly for small and medium-sized enterprises
(SMEs).
OPPORTUNITIES
1 Short-Term Gains
Businesses prioritizing short-term economic gains
over sustainability may benefit from lower compliance
and operational costs in the short term but could find
themselves ill-prepared over the mid-term as new
technologies and regulations begin to scale.
2 New Revenue Streams
Companies in the energy storage, electric heat pumps,
and renewable energy sectors could tap into new revenue
streams as their technologies and services become
increasingly in demand.
3 Comprehensive Sustainability
Companies that adopt and integrate broad sustainability
strategies can appeal to a wider customer base, reduce
waste, and improve operational efficiencies.
EXECUTIVE SUMMARY
2024 3. HOW WILL COUNTRIES AND BUSINESSES COMMIT TO DECARBONIZATION?

EXECUTIVE SUMMARY
2024 3. HOW WILL COUNTRIES AND BUSINESSES COMMIT TO DECARBONIZATION?
How is your business
leveraging technological
advancements in
marine technology to
promote sustainable
growth within the blue
economy?
What measures are in
place to navigate the
complexity of expanded
environmental regulations
without stifling
innovation or economic
performance?
diversifying its operations
or supply chains to ensure
resilience and sustained
growth?
preparing to align with
new ESG policies, and
what investments are
being made to upgrade or
support the infrastructure
for renewable energy and
grid optimization?
What strategies are
in place to adapt
to and comply with
international marine
resource management
agreements, such as
in your supply chain
practices?
1 4
2 5
3
STRATEGIC QUESTIONS

04
How will humans prioritize their
work environments in the future,
and what will these preferred
workplaces look like?
EXECUTIVE SUMMARY
2024

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
How will humans prioritize their work
environments in the future, and what will these
preferred workplaces look like?
Work Is Task-Dependent Purpose-Driven Locations
Decentralized Work
Communities
Smart Workspaces
Empower Talent
AI Agents Manage
Logistics
Personalized
Coworking Ecosystems
Division in
Urban Areas
Reskilling for
Task-Based Work
Enabled by Smart
Workspaces and AI Agents

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Work Is Task-Dependent
Companies emphasize skill-focused tasks over traditional role-based assignments. This approach opens up
opportunities for both fixed-location and remote workers, breaking down previous barriers in talent pools.
Consequently, employees have the flexibility to work from home or community-based locations.
Decentralized Work Communities
The social elements and team structures of traditional offices diminish in
importance. Without defined roles anchoring them together, employees perform
their individual tasks remotely.
Workers split time between truly remote work and convening in smaller satellite
workspaces when they want a sense of community. Employees enjoy the
flexibility to simultaneously hold multiple “careers,” contributing varied tasks to an
ecosystem of employers rather than being siloed at any single organization.
Reskilling for Task-Based Work
Ensuring the right skills for each job is a priority. Sensitive to the high search
costs for new talent, companies are interested in retaining and training existing
employees to meet evolving task needs. Traditional offices are nimble learning
centers focused on immersive training for current staff.
This cultivated talent pipeline of cross-functional employees, versed in specific
on-demand abilities, replaces siloed departments. Companies equip their
workforce with precise capabilities rather than seeking elusive candidate fits.
4. HOW WILL HUMANS PRIORITIZE THEIR WORK ENVIRONMENTS IN THE FUTURE, AND WHAT WILL THESE PREFERRED WORKPLACES LOOK LIKE?

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Enabled by Smart Workspaces and AI Agents
Workflows are enabled by fluid, personalized environments that empower productivity with human
and artificial partners working in concert. People are free to perform, augmented by supportive
spaces that travel with them.
Smart Workspaces Empower Talent
Offices function as intelligent computing zones. Spatial computing and natural
language processing enable embodied interaction. Employees’ tasks and projects
surface on walls and tables as they enter a room. Workers can literally get a grasp
on their work by manipulating 3D projections by hand. Voice commands replace
typing for nimble modification.
Employees gain sensory proximity to their work. Previously abstract digital files
are now tangible in augmented environments that both feel familiar and unleash
workspace innovation.
AI Agents Manage Logistics
Specialized AI agents handle distinct tasks. A product manager AI agent could
aid market analysis, prioritize features, and develop business cases. Meanwhile, a
developer agent would focus on automating code generation and detecting bugs.
Each person’s AI agent coordinates to remove logistical burdens. Christina’s
assistant tracks that she prefers morning meetings. Jeff’s assistant knows he favors
early afternoons. By sharing insights into their humans’ constraints, the assistants
identify 11 a.m. as the optimal meeting time. Because the assistants dynamically
adjust activities based on their human’s context, they enable new flexibility: Workers
can perform efficiently wherever they happen to be.

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Purpose-Driven Locations
Workers want transparency and a seat at the table to make company decisions. Labor unions
and employees continue to push for greater autonomy and new benefit packages and locations.
Companies join together to offer a complex of potential work environments, creating stability in some
urban regions.
Personalized Coworking Ecosystems
Employees have access to coworking ecosystems that cater to their lifestyles.
For health buffs, wellness hubs provide gyms, cold plunges, and saunas. Parents
join family-friendly sites with childcare and nursing spaces. Rather than isolated
offices, companies cluster in amenity-rich locales to give workers choices.
Workers evaluate these full experience packages when considering job
opportunities. Companies differentiate with tailored sites while still coming
together in broader mixed-use developments. Competing businesses may not
share ecosystems, but complementary industries do, recognizing the potential for
expanded talent recruitment and retention.
Division in Urban Areas
The urban centers of cities are becoming patchworks of utilitarian spaces that lack
the vibrancy and interconnectedness of traditional urban life. As offices emptied
out due to employee cynicism, businesses turned their properties into lab spaces,
vertical farms, and storage. Socio-economic divides are deep, as these spaces do
not address communal needs, only corporate interests. Once vibrant downtowns
focus on efficiency and production versus community connection. Manufacturers
are using the empty spaces to fill warehouses and production sites closer to their
consumers. Malls and other shopping facilities that moved out of downtown now
compete with these manufacturing centers.

THREATS
1 Ethical Challenges with AI Integration
Integrating AI agents into the workforce introduces
cultural and ethical challenges, including dependency
on technology, potential job displacement concerns, and
ethical considerations around AI decision-making.
2 Overdependence on Technology
Heavy reliance on smart workspaces and AI for operational
efficiency could lead to overdependence on technology,
making companies vulnerable to tech failures, outages, or
obsolescence.
3 Fraud Risk
As decentralized work scales alongside increasingly
sophisticated AI, various forms of deepfakes could
present a security risk to remote working environments,
necessitating businesses to improve their cybersecurity
preparedness.
OPPORTUNITIES
1 Talent Ecosystem Development
Companies and industries can develop talent ecosystems
that foster cross-pollination of skills and ideas by creating
networks of complementary businesses and coworking
spaces that enhance talent satisfaction and retention.
2 Urban Redevelopment and Repurposing
Municipalities and businesses will need advice on
repurposing urban spaces that have lost their traditional
vibrancy due to shifts in work patterns.
3 Purpose-Driven Workspace Solutions
Companies may create purpose-driven locations and
co-working ecosystems that align with their employees’
values and lifestyle preferences. This could involve strategic
planning for the use of corporate real estate to support
community engagement, wellness, and collaboration.
EXECUTIVE SUMMARY
2024 4. HOW WILL HUMANS PRIORITIZE THEIR WORK ENVIRONMENTS IN THE FUTURE, AND WHAT WILL THESE PREFERRED WORKPLACES LOOK LIKE?

How can we develop a
continuous learning and
development ecosystem
to ensure our workforce
remains adaptable and
skilled for future tasks?
How can we use new and
emerging technologies
to foster innovation
and creativity in a
predominantly remote or
hybrid workforce?
How can we leverage
data and analytics
to improve remote
work productivity and
employee engagement
while respecting privacy
and autonomy?
What do we need to
reevaluate or adjust
in order to remain
competitive in attracting
and retaining top talent
as the workspace
evolves?
In what ways can
we leverage smart
workspaces and AI
agents to enhance
productivity without
compromising security,
privacy, and employee
well-being?
1 4
2 5
3
EXECUTIVE SUMMARY
2024 4. HOW WILL HUMANS PRIORITIZE THEIR WORK ENVIRONMENTS IN THE FUTURE, AND WHAT WILL THESE PREFERRED WORKPLACES LOOK LIKE?
STRATEGIC QUESTIONS

05
In the context of increasing
protectionism and
geopolitical tensions, what is
the future of the global chip
manufacturing industry?
EXECUTIVE SUMMARY
2024

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
In the context of increasing protectionism and
geopolitical tensions, what is the future of the
global chip manufacturing industry?
Chip Imports and Exports Are
Increasingly Restricted
Protectionist Policies
Result in a Skills Gap
The Cost of Data
Centers Rise
The US Dominates China Rallies
In-House Production
and Training Ramp Up
Open Borders for
Chip Industry Growth
Tech Companies as
New Energy Providers
The End User Pays

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Chip Imports and Exports Are Increasingly Restricted
The industry continues its pivot toward strategic autonomy. Defensively, the US is onshoring
chip production to prevent supply chain disruptions like those during the Covid semiconductor
shortage. Offensively, the US is strategically preventing exports to get a leg up in the AI race
against China. Meanwhile, independent companies, like OpenAI, want to create new types of
chips that won’t run afoul of government regulations.
The US Dominates
The US tightens enforcement of chip exports to prevent adversaries from
attaining US chips and manufacturing equipment. US allies are required to do
the same. Technologies like geotracking prevent leaks in export controls, thereby
more effectively curbing US technologies from getting to China and Russia. This
challenges those countries’ reliance on foreign chips, pressuring their big tech to
innovate with less sophisticated resources.
Since China is a major player in the consumer electronics market, the restrictions
lead to higher prices and limited availability of various consumer goods worldwide.
China Rallies
US-led export controls do not prevent China and other adversaries from acquiring
powerful chips from companies like Nvidia. Black markets emerge for such chips,
and IP espionage efforts ramp up. With these powerful chips, China’s AI rivals US
AI companies in capabilities. There is a bifurcation of “Eastern AI” and “Western AI.”
China also continues to dominate global legacy chip production. Legacy chips
underpin everything from microwaves to military weapons systems. China boxes
out foreign competitors through dumping, rendering the US dependent on China,
at least for legacy chips.
5. IN THE CONTEXT OF INCREASING PROTECTIONISM AND GEOPOLITICAL TENSIONS, WHAT IS THE FUTURE OF THE GLOBAL CHIP MANUFACTURING INDUSTRY?

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
Protectionist Policies Result in a Skills Gap
The expertise required for chip design and production, previously outsourced affordably, is now
lacking domestically. As a result, manufacturing chips domestically has become a costly endeavor.
This has sparked intense competition among companies to attract and acquire talent specialized
in chip design and manufacturing.
In-House Production and Training Ramp Up
Semiconductor companies are bringing chip production in-house and cultivating
essential skills internally. They also in-source skills training by reaching out to high
school students, offering a direct path into the burgeoning US chip industry. This
new model positions the semiconductor sector as a realm of new blue-collar
workers. By providing in-house training and certification, these companies offer
young talent a debt-free alternative to college education, enabling them to start
earning immediately while learning specialized skills.
Open Borders for Chip Industry Growth
Progress in opening new fabrication plants, funded by the US Chips Act, has
been slower than anticipated. The fabs’ hosts, including cities in Texas, Ohio, and
Florida, worry about becoming modern equivalents of Detroit—where initial high
investments won’t yield long-term benefits.
States traditionally resistant to open immigration policies are shifting their stance,
and the US is responding with expedited visas for skilled workers, particularly
from Taiwan, leading to a notable brain drain in the island nation and increased
tensions with China.

UNCERTAINTIES
EXECUTIVE SUMMARY
2024
The Cost of Data Centers Rise
As AI-driven workloads intensify, the operational costs of data centers are escalating. The
increased demand for processing power and energy to support complex AI tasks significantly
drives up the expenses associated with maintaining and running these facilities.
Tech Companies as New Energy Providers
Tech companies are venturing into producing alternative, cost-effective energy
sources like small modular reactors and geothermal power. This move is not just
about affordability; it represents a leap in overcoming long-standing negative
perceptions of nuclear energy and its historical barriers of high costs and lengthy
construction times.
The implications are profound, and not just for the energy sector: Apartment
buildings will be powered, cooled, and lit efficiently by Amazon, and entire
neighborhoods receive clean, cost-effective energy from Google or Microsoft.
The End User Pays
Data center efficiency advancements have reached a plateau, compelling the
industry to focus on expansion to meet increasing workload demands.
Providers can try to stay in densely populated areas that offer skilled labor
availability and robust fiber networks, but they’ll have to deal with increasing
restrictions amid concerns that data centers exacerbate energy shortages. Or
they could venture into less ideal locations lacking infrastructure and requiring
substantial capital investment. Either option presents an increased cost of digital
services for consumers and a significant ripple effect across the economy.

THREATS
1 Big tech gets bolder
Utilities could become disintermediated by tech
companies, who find a competitive advantage in
bundling energy with other services.
2 Regulatory disruption
Changes in trade policies and regulations in
key markets could affect access to critical chip
technologies and materials.
3 Tech transfer
Intellectual property and sensitive technology
become entangled in technology transfer
regulations, especially in countries involved in
geopolitical tensions.
OPPORTUNITIES
1 Cultivate partner and supplier relationships
Businesses will need to develop new partners,
manufacturers, and markets so they have first mover
advantage if geopolitical tensions rise unchecked,
impacting access to key components and markets.
2 Get local
Companies may Invest in local or regional production
capabilities to reduce dependence on global markets,
increasing customization capabilities and time to market.
3 Fund alternatives
Technology firms must invest more in research and
development to create more advanced or alternative chip
technologies that address the emerging challenges.
EXECUTIVE SUMMARY
2024 5. IN THE CONTEXT OF INCREASING PROTECTIONISM AND GEOPOLITICAL TENSIONS, WHAT IS THE FUTURE OF THE GLOBAL CHIP MANUFACTURING INDUSTRY?

EXECUTIVE SUMMARY
2024 5. IN THE CONTEXT OF INCREASING PROTECTIONISM AND GEOPOLITICAL TENSIONS, WHAT IS THE FUTURE OF THE GLOBAL CHIP MANUFACTURING INDUSTRY?
How will geopolitical
tensions affect the cost
of chips, and how can
you mitigate these cost
increases?
What financial cushions
can be put in place to
absorb potential shocks?
Considering the current
and potential future
geopolitical landscape,
how should you adjust
our long-term strategic
positioning to ensure
sustainability and
growth?
How do geopolitical
tensions affect
your commitment
to environmental
sustainability and ethical
sourcing, especially
concerning rare materials
used in chip production?
How must you enhance
your cybersecurity
measures to protect
against potential
threats exacerbated by
geopolitical tensions,
especially in critical
infrastructure related to
chip manufacturing and
distribution?
Do you have a robust
crisis management plan
that includes scenarios
involving escalated
geopolitical tensions
affecting chip supply?
1 4
2 5
3
STRATEGIC QUESTIONS

Tech Trends Report 2024 Future Today Institute.pdf

2024 TECH TRENDS REPORT • 17TH EDITION

TECH
TABLE OF CONTENTS
52 Your Guide to the Future of AI
53 Top Headlines
54 State of Play
56 Key Events
59 Likely Near Term Developments
60 Why Artificial Intelligence
Trends Matter to Your
Organization
62 When Will Artificial
Intelligence Trends Impact Your
Organization?
64 Opportunities and Threats
65 Investments and Actions
To Consider
66 Central Themes
69 Ones To Watch
70 Important Terms
73 Models, Techniques,
and Research
74 What is an AI model?
76 Purpose-Built Models
76 LLMs Are Getting Bigger
and More Expensive
84 Tools to Combat Broadly
Malicious AI Behavior
85 Does AI infringe on privacy?
How should we think about
customer data and AI
applications?
85 Increased Used of Ambient
Surveillance
85 Worker Surveillance
85 School Surveillance
87 Is there a feasible solution to bias?
87 Addressing Political Bias
88 Addressing Race and Gender Bias
89 What security issues should we
prepare for?
89 Cyberthreats
89 Adversarial Attacks
90 Data Poisoning: A Double-Edged
Sword
90 AI Lowers the Barrier to
Misinformation
91 Privacy Risks in Behavioral
Biometrics
92 What does AI have to do with ESG?
92 New Architectures to Make AI
Workloads More Efficient
93 A Nuclear Renaissance for AI
Workloads
93 Environmental AI
94 Policy and Regulations
95 How does geopolitics factor into
the development of AI, and is there
really a new cold war?
95 AI Nationalism
96 The AI-Driven Chip War
97 Could AI be involved in—or
cause—a hot war?
97 Autonomous Weapons Policies
97 Simulating Warfare
97 AI Used to Guide Military Strikes
98 Automated Target Recognition
98 Automating Offensive Attacks
Using AI
98 AI-Assisted Situational Awareness
98 Algorithmic Warfighting
99 Mandating Ethics Guidelines for
Tech Contractors
100 Regional Approaches
101 Countries try to regulate AI,
but plans diverge
102 How is the US specifically
regulating AI?
102 A Patchwork Approach
104 Conflicting Views About
Institutional Roles
104 Public-Private Partnerships
105 National Security
106 What is China doing?
106 China’s Expanding Market
106 China’s Big Tech
107 Deepening International Ties
109 What is Europe doing?
111 What is the Middle East doing?
113 Talent
114 Where and how do I get AI talent?
114 Demand for AI-related Skills
Increases Across Sectors
114 AI Brain Drain from Academia
115 How will AI change the nature
of work?
76 LLMS as Operating Systems
78 Should we go open-source or
proprietary?
78 Open-Source LLMs for
Commercial Use
79 Safety, Ethics Society
80 Is AI really a black box?
80 Explainable AI (XAI)
80 AI Intentionally Hiding Data
81 How do we ensure trust?
81 AI Alignment Goes Mainstream
81 Indexing Trust
82 Synthesizing Trust
83 Are there tools to make AI ethical?
83 Deepfake Detectors
83 Tools for Identifying AI-Generated
Writing
83 Tools for Detecting Copyright
Violations in AI Outputs
83 Tools for Exposing Deepfakes
84 Tools to Thwart Recognition
Systems

TECH
TABLE OF CONTENTS
115 Gains and Pains
115 Status Shifts
116 Agents Will Increasingly Perform
Tasks on Our Behalf
117 Emerging Capabilities
118 Can AI reason? And how close are
we really to AGI and ASI?
118 AI Breakthroughs in Mathematics
119 AI Persuasion
119 Prediction and Prescience Into Our
Human Lives
119 Detecting Emotion
120 Neuro-symbolic AI
121 Is the future of AI cloud, edge, or
on-device?
121 Cloud Neutrality
121 Cloud Strain From AI Boom
121 AI Breathes Life Into Legacy
Systems
122 Optimizing AI to Run at the Edge
122 Small Language Models for AI at the
Edge
131 How is AI being used in pharma?
132 Protein Folding
132 AI-First Drug Development
132 Generative Antibody Design
133 How is AI being used in health
care?
133 AI to Improve Patient Outcomes
133 AI-Assisted Diagnosis and Clinical
Decision-Making
133 Anomaly Detection in Medical
Imaging
134 AI-powered movement
134 Medical Deepfakes
134 Healthcare-Specific LLMs
135 In-Silico Trials
135 AI for Mental Health
136 How is AI being used in science?
136 AI-Driven Hypotheses
136 AI-Driven Experimentation
137 AI-Powered Analysis and
Interpretation
137 AI and the Replication Crisis
137 NLP Algorithms Detect Virus
Mutations
138 AI to Speed Up New Materials
Development
139 How is AI being used in finance?
139 Mitigating Fraud
139 Predicting Financial Risk
140 Customized Portfolios
140 Growing Concern About Centralized
Data Sets
141 How is AI being used in insurance?
141 Predicting Workplace Injuries
141 The Connected Worker
141 Improving Damage Assessment
141 Consumer-Facing Robo-Advisers
141 AI Claims Processing
142 Liability Insurance for AI
143 Creativity and Design
144 How are people using AI to be more
creative?
144 GAN-Assisted Creativity
144 Neural Rendering
145 Generating Virtual Environments
From Short Videos
145 AI Democratizes Music Production
145 Automatic Ambient Noise Dubbing
146 Generating Music From Text
147 How is AI disrupting the creative
industry?
147 AI-Assisted Invention
147 New Business Models
148 Legal Battles Between Writers
and AI
149 Scenarios
150 Scenario: The Deepfake Mafia
151 Scenario: TrailMate SLM
152 Scenario: Centralized AI Belt and
Road Infrastructure Crumbles
153 Scenario: Tabby the Tiger:
Nurturing Curiosity Through AI
Friendship
154 Scenario: What If “Thought-to-3D”
Was an AI Modality?
155 Authors
158 Selected Sources
122 On-Device AI
123 Wearable AI
124 Why should we pay attention to
emerging capabilities that aren’t
yet fully developed?
124 Vector Databases
124 Vertical Integration From Hardware
to LLMs
126 Industries
128 How is AI being used in HR?
128 Autonomous Talent Acquisition
128 Customer and Personnel
Recognition Systems
128 Benefits Selection and
Management
129 How is AI being used in
marketing?
129 AI Shifts Search
129 Dynamic Engagement Through
Deep Personalization
129 AI-Assisted Campaigns
130 Anecdotal Observations, Now
Usable Marketing Data

TECH
For two decades, our commitment at Future Today Insti-
tute to understanding and leveraging artificial intelli-
gence has been unwavering—even as general excitement
about AI has wavered considerably. We’ve watched in-
terest ebb and flow across industries, among executive
leadership and boards of directors, and with investors,
legislators, and academia.
Today, we’re at a crucial inflection point in AI’s develop-
mental journey. This moment isn’t marked by a single
technological breakthrough but rather by a development
that at first may seem less intuitive. What changed in the
past year is our perception of what AI is and how it will
change everyday life. AI models are now accessible to con-
sumers and businesses alike, so their value can be im-
mediately understood. What’s followed: investment, new
partnerships, and the grand expansion of value networks.
This rapid escalation in activity has left leaders feeling
caught off guard, prompting an urgent need for strategic
decision-making. In our conversations with clients and
partners, a common theme has emerged: Leaders, under-
standably concerned about missing out on the next wave
of innovation, need clarity about a complex area of tech-
nology that will continue to evolve for many years to come.
Recognizing this, we’ve reimagined our approach for
the AI section of our 2024 Tech Trends report. By actively
listening to leaders and experts in our network, we’ve
curated the most common questions we’re hearing from
our clients and grouped our AI trends accordingly. Our
goal is to guide leaders through a thoughtful exploration
of these questions, enabling a deeper understanding of
the implications.
While we don’t claim to hold all the answers—every orga-
nization’s journey with AI will be somewhat unique—what
we do offer is a foundation of extensive research and in-
sightful, strategic analysis. You will encounter questions
that mirror your own, as well as those you may not have
considered but will likely confront in the coming year. Our
aim is to equip you with the insights necessary to navi-
gate the upcoming impacts on your organization.
We are confident that this report will serve as an invalu-
able tool for leaders looking to identify strategic oppor-
tunities, achieve competitive advantages, and enhance
organizational resilience in the foreseeable future.
Welcome to your guide on the future of AI.
Amy Webb
Sam Jordan
Manager and Advanced Computing Practice Lead
YOUR GUIDE TO THE FUTURE OF AI

This year, the AI landscape
could undergo a
significant consolidation.
Strategic investments,
groundbreaking
innovations, and
regulatory maneuvers
will further empower a
select group of power
brokers, intensifying the
competitive dynamics and
shaping the trajectory of
global AI dominance.
TOP HEADLINES
01
02
03
04
05
06
OpenAI Seeks $7 Trillion Investment
OpenAI’s Sam Altman wants to revamp the semiconductor sector with trillions in investment, targeting
global chip capacity to boost AI growth, engaging with investors like the United Arab Emirates.
Mainstream Multimodality
For the first time, the public can interact with advanced multimodal AI models capable of understanding
and generating various media types, including images and videos. This innovation is akin to human
learning processes, enabling AI to learn from visual and auditory information, not just text—just like us.
Nvidia GPUs in High Demand for AI Training
The rush to acquire Nvidia’s powerful GPUs for AI model training has intensified, as everyone from tech giants
to startups seeks the computational horsepower these units offer for advanced machine learning tasks.
Open-Source AI Uprising
Meta sparked an open-source large language model movement by releasing the weights for LLaMA,
enabling researchers to freely build off the model, fine-tune it, and create customized versions.
US Tightens Grip on AI
The US intensifies its stance on AI by restricting access to crucial enabling technologies like semiconductors.
In a strategic move, the US also pressures allies to implement similar restrictions against China and Russia.
EU AI Act Sets New Global Standard
The EU introduces the first-ever comprehensive AI regulation and a European AI Office, aiming to ensure
safety and respect for fundamental rights, while encouraging innovation and investment in AI technologies
across Europe. Yet, concerns arise about stifling innovation and the high costs for businesses to comply.
TECH

STATE
OF PLAY
The past year marked a watershed moment for artificial intelligence. Central
to this transformation are the leaps in large language models (LLMs) and their
practical applications, which have not only advanced the frontiers of AI but have
also catalyzed a broader integration of AI technologies into everyday life. AI prom-
ises revolutionary improvements in health care and life sciences: Now that we’ve
cracked the code on protein structures, an unimaginable number of new thera-
peutics are on the horizon, along with alternatives to address climate change. In
the coming year, AI’s reach will extend to people, pets, and objects alike, paving
the way for a very near-future in which digital assistants, automated systems,
and spatial awareness are seamless, ubiquitous, and invisible. In parallel, the
advancements in robotics, both hard and soft, are pushing the boundaries of
automation and human-machine interaction.
At the same time, AI’s energy demands pose a paradox, offering climate solu-
tions but also contributing to carbon emissions, a concern in energy-con-
strained areas. The persistent talent gap in AI, particularly in data science, lim-
its its application in critical sectors like agriculture and health care. Political
engagement with AI is growing, which is good, but the end result has been a
slew of competing policies. Regulatory compliance and enforcement remains a
challenge, since depending on the country’s position, they promise to both ac-
celerate and curtail the deployment of AI systems. The unequal distribution of AI
advancements risks deepening global inequalities, with the global south facing
significant disadvantages. Amid all these developments is the persistent geopo-
litical tensions between China and the West.
For many, generative AI (genAI) is the first entry point into this new reality, which
explains the explosive growth we’ve seen in the past year. Our analysis reveals a
AI embeds into everything,
transforming how we
interface with computers
while researchers work to
make AI more capable and
efficient. This sprint toward
progress unfolds amid rising
geopolitical tensions, as
dominance of these strategic
technologies reshapes
global power dynamics.
© 2024 Future Today Institute. All Rights Reserved.
TECH ARTIFICIAL INTELLIGENCE
54

STATE
OF PLAY
surge in experimentation with genAI tools across various sectors, indicating a
transformative shift toward embracing AI’s potential to innovate and streamline
operations. This widespread interest in genAI spans a broad spectrum of stake-
holders, from business leaders to frontline workers, highlighting just how perva-
sive the first generation of tools has become.
But here’s the thing: GenAI isn’t all of AI. Often, when people talk about “AI” what
they really mean is “automation.” Artificial intelligence is an umbrella term that
encompasses many different techniques, models, and frameworks that make up
the field. AI’s aim is to create intelligent machines that can sense, reason, act,
and adapt like humans do, or in ways that go beyond our capabilities. Today, cars
can park themselves, while emerging platforms are capable of having seemingly
natural conversations. Now, AI is evolving to have beyond-human capabilities.
It has invented new drugs, predicted the real-time movement of wildfires, and
autonomously designed machine parts.
Developing AI requires extraordinary resources, which is why consolidation
among the tech giants is tightening. The biggest names in AI––OpenAI, Deep-
Mind, Anthropic––are increasingly hitched to the world’s biggest hyperscalers
and cloud providers (Microsoft, Google, Amazon). Venture capital and private eq-
uity are still flooding into startups and mature companies alike, and now, sover-
eign wealth funds have a seat at the table.
AI is magical, but it isn’t magic. As long as expectations are tempered, this
should be an era of significant innovation, experimentation, and growth, espe-
cially as AI propels growth in other areas of science and technology. We are cau-
tiously optimistic about what’s on the horizon.
55

KEY EVENTS
JANUARY 12, 2023
AI Breakthrough in Lung Cancer
Detection
A joint effort by MIT and Mass
General Hospital yields a significant
advancement in lung cancer
prognosis with the creation of a deep-
learning model that assesses lung
cancer risk from CT scans, potentially
enhancing early detection and saving
numerous lives.
JANUARY 26, 2023
High-Fidelity Music from Text
Google Research’s MusicLM
introduces a transformative approach
to generating detailed music
from text descriptions, achieving
unprecedented audio quality and text
adherence in the field.
FEBRUARY 6, 2023
Google Unveils Bard
Google introduces Bard, an innovative
AI chatbot powered by its language
model LaMDA, as a response to
ChatGPT.
FEBRUARY 7, 2023
Bing Adopts ChatGPT
Microsoft revolutionizes its Bing search
engine and Edge browser with the
integration of OpenAI technology.
FEBRUARY 21, 2023
AWS and Hugging Face Collaboration
AWS teams up with Hugging Face to
streamline AI projects on Amazon’s
cloud, simplifying the deployment of AI
applications.
FEBRUARY 21, 2023
Real Fusion’s Photographic
Breakthrough
Oxford researchers showcase Real
Fusion, a cutting-edge AI that can
reconstruct a complete 360-degree
photographic model from just one image.
FEBRUARY 24, 2023
Meta introduces LLaMa
The compact yet advanced 65-billion
parameter language model is open-
sourced and free for research and
commercial use.
MARCH 1, 2023
OpenAI Expands Developer Tools
OpenAI launches ChatGPT and
Whisper APIs, providing developers
with advanced language processing
and speech-to-text capabilities
beyond basic chat functions.
MARCH 14, 2023
Google Introduces AI in Workspace
Google’s launch of assistive AI
features in Workspace started with
AI-powered writing tools in Docs and
Gmail for trusted testers.

KEY EVENTS
MARCH 21, 2023
Adobe Unveils Firefly
This new generative AI suite is
designed to help users at all skill
levels create high-quality images and
text effects.
MARCH 28, 2023
Khan Academy Launches Khanmigo
Khan Academy launch of the
Khanmigo AI platform integrates
virtual bots as counselors, curriculum
designers, and teaching assistants.
AUGUST 15, 2023
Google Launches Search
Generative Experience
Google introduces genAI into search
queries, automatically generating
summaries.
SEPTEMBER 21, 2023
Microsoft Unveils Co-pilots
Microsoft’s AI-powered 365 Copilot
and GitHub’s CopilotX offers enhanced
assistance by integrating web context,
work data, and real-time PC activities,
prioritizing privacy and security.
SEPTEMBER 21, 2023
YouTube Debuts AI Editing App
The new app, YouTube Create, makes it
easy to trim videos, slow down the pace,
or add audio.
OCTOBER 25, 2023
Amazon Introduces AI Image
Generation
Amazon rollout of AI-powered image
generation capabilities through
Amazon Ads in beta aims to enhance
ad experiences by enabling brands to
create lifestyle imagery that boosts ad
performance.
NOVEMBER 4, 2023
Elon Musk’s xAI Debuts Grok
Inspired by the “Hitchhiker’s Guide to
the Galaxy,” Grok answers questions
with wit and provides real-time
world knowledge via the X platform,
distinguishing itself by addressing
inquiries often declined by other AIs.
NOVEMBER 14, 2023
Google Announces AI Genesis
AI Genesis features the Gemini
large language model in three
sizes: Gemini Ultra for extensive
capabilities, Gemini Pro for broad
task applications, and Gemini Nano
optimized for specific tasks and
mobile use.
NOVEMBER 17-22, 2023
OpenAI’s Turbulent Week
After a tumultuous five days marked
by his ouster and subsequent
reinstatement, Sam Altman resumes
his role as CEO of OpenAI, buoyed
by a concerted effort from allies,
employees, and investors.

KEY EVENTS
NOVEMBER 21, 2023
StabilityAI Introduces Stable Video Diffusion
StabilityAI’s inaugural foundation model for
generative video builds on the technology of its
image model, Stable Diffusion.
NOVEMBER 28, 2023
Pika Debuts AI Video Editing App
The app includes a new suite of videography
tools with a generative AI model that
edits videos in diverse styles such as “3D
animation,” “anime,” and “cinematic.“
NOVEMBER 29, 2023
DeepMind Predicts Novel Material Structures
Google DeepMind’s researchers have leveraged
AI to accurately predict the structures of
more than 2 million new materials, offering
significant implications for renewable energy
and computing sectors.
DECEMBER 5, 2023
AI Alliance for Responsible Innovation Forms
The AI Alliance for Responsible Innovation,
including IBM, Meta, and 50 other organizations,
launches as a global consortium aimed at
promoting open, safe, and responsible AI
development and adoption.
DECEMBER 6, 2023
Google’s Gemini Surpasses GPT-4
Google’s next-generation AI model
outperformance of OpenAI’s GPT-4 set a new
standard in AI capabilities.
DECEMBER 9, 2023
EU Finalizes AI Act
The European Union achieved a landmark
agreement with the Artificial Intelligence Act,
introducing binding rules and standards for
developing AI more responsibly.
DECEMBER 13, 2023
Axel Springer Partners with OpenAI
The German media titan’s partnership
lets OpenAI use Politico and Business
Insider articles for AI training, while those
news platforms get to employ ChatGPT for
summarizing news, marking a significant
yet controversial collaboration in the news
industry’s quest for innovation and survival.
DECEMBER 14, 2023
DeepMind’s FunSearch Breaks Boundaries
The tool has successfully solved complex
issues, proving AI’s ability to surpass the
limitations of its training data in large
language models (LLMs).
DECEMBER 14, 2023
DeepMind Solves the Unsolvable with AI
DeepMind’s use of an LLM to crack an
“unsolvable” math problem marks a
historic achievement, as detailed in Nature,
showcasing the model’s capacity to uncover
new, verifiable knowledge on a longstanding
scientific challenge.

LIKELY NEAR TERM DEVELOPMENTS
GENERAL
Commoditization of General Purpose Models
In the near future, expect the commoditization of
general purpose models. LLMs are becoming widely
accessible and integral to app development. As
these models become ubiquitous and cost-effec-
tive, akin to cloud services, their adoption will stan-
dardize across industries, diminishing their role as
a competitive differentiator.
Large Reasoning Model
Vertically integrated solutions will garner a higher
transactional value. Some companies will win by
providing “a refined/value-added LLM product” to
the end consumer and meeting the customer in
desired distribution channels, such as LLMs for
health care, legal, finance, and architecture.
Adoption of Natural Language Interfaces
The evolution toward natural language interfaces
will soon diminish the reliance on traditional graph-
ic user interfaces. This shift will enable more intu-
itive interactions with computers, using everyday
language. This transition may also influence device
form factors, potentially leading to an increase in
wearables and the development of AI-specific devic-
es and operating systems centered around LLMs.
AUTOMATION
AI Assistants Transform Coding Landscape
AI coding assistants, such as GitHub’s Copilot
and Meta’s Code Llama, are transforming software
development with advanced autocomplete func-
tions and innovative debugging tools, offering both
premium and free solutions to enhance coding
efficiency and creativity. Expect to see more im-
provements to these tools and more tools to launch
in this space.
AI Integration in Health Care and Life Sciences
Generative AI will lead to breakthroughs in pro-
teins, antibodies, and drugs. Specialized models
will continue to accelerate discovery in biology and
chemistry, sparking more practical applications
and boosting investment.
REGULATION AND GEOPOLITICS
US Strategy on AI and China Relations
The US is expected to intensify efforts to get allies
to limit their collaborations with China in AI de-
velopment, following President Biden’s enhanced
export restrictions on semiconductors. With the
Netherlands aligning with US requests, further
demands on allies to adopt similar stances aim to
curb China’s AI advancements.
Europe Begins Regulating AI
The European Commission will open its European
AI Office, which will oversee the development and
use of safe artificial intelligence (within Europe, at
least) and assist with the implementation of the AI
Act. The office will enforce general purpose AI rules,
monitor compliance, and attempt to become a hub
for international cooperation on AI governance.
Challenges in US Chip Manufacturing Expansion
The US moves to onshore chip fabrication will expe-
rience growing pains associated with higher labor
costs compared to Taiwan. This shift may lead to
increased expenses in constructing fabs and pro-
ducing domestically made chips, surpassing initial
estimates outlined in the CHIPS Act.
ENTERPRISE
Talent Shift in AI Industry
Expect a significant talent crunch as top innova-
tors depart major tech giants like Google, OpenAI,
and Meta to launch their own ventures, ranging
from conversational agents to AI-first biotech
firms, signaling a broad diversification and special-
ization within the AI sector.
Consolidation in 2024
Consolidation will persist this year, building on
moves like Microsoft’s 2023 increased investment
in OpenAI for Bing, aimed at capturing market share
from Google search. Similar strategies by major
tech companies are anticipated throughout 2024.
Increased Enterprise Adoption of AI
The current macroeconomic environment is driving
leaders to view AI as essential for growth, antici-
pating increased enterprise adoption despite the
potential for making some job categories obsolete.

WHY ARTIFICIAL INTELLIGENCE TRENDS MATTER TO YOUR ORGANIZATION
Future Today Institute believes AI is a force multiplier on technological progress because it is an en-
abler of other technologies and powers the evolution of business, government, and society. But new
large language model capabilities deeply concern some in professional and creative services. Models
can now reason about concepts in text, not just perform pattern matching. They display forms of com-
mon sense and analogy—tasks once seen as uniquely human. And they apply these reasoning abilities
across modalities—text, image, video, and more. Most alarming to some is that models seem to en-
gage in recursive self-improvement when given the right training. They don’t just learn a static set of
parameters. They learn how to learn better, becoming moving targets.
Since publishing our first Tech Trends report 17 years ago, we have included and expanded our cover-
age on artificial intelligence. What began as several pages of insights is now a dedicated, stand-alone
report with more than 100 trends to monitor. AI is already transforming most economic sectors, but we
anticipate deeper impacts this year across insurance, finance, entertainment, health care, biotechnol-
ogy, and cloud computing.
Global Tech
Rivalry
The race for AI supremacy
is intensifying geopolitical
tensions, notably
between the US and
China. Businesses must
navigate a landscape
where technology and
national security are
increasingly intertwined,
affecting international
supply chains, market
access, and regulatory
compliance. Companies
specializing in AI and
related technologies
might face stricter export
controls, requiring them
to adjust strategies for
product development and
global expansion.
Supply Chain
Diversification and
Onshoring
As tensions escalate,
particularly in the
semiconductor industry,
businesses will need to
diversify their supply
chains to mitigate risks.
The bifurcation in the
AI chip market might
compel companies to
innovate independently
or bring supply chains in-
house, potentially leading
to increased costs.
Business Impacts

WHY ARTIFICIAL INTELLIGENCE TRENDS MATTER TO YOUR ORGANIZATION
Strategic Talent
Acquisition
Companies must innovate
in talent acquisition and
retention strategies to
compete for scarce AI
expertise, particularly
against tech giants. This
may include offering
competitive salaries,
benefits, and unique work
environments, as well as
investing in employee
development and internal
AI training programs to
build talent in-house.
Custom, Fit-for-
Purpose LLMs
Organizations that opt
for custom AI models
over general-purpose
ones can achieve greater
alignment with specific
business objectives. This
differentiation can lead to
competitive advantages
in operational efficiency,
customer insights, and
product innovation.
Adversarial AI
Preparedness
The susceptibility of AI
systems to adversarial
attacks calls for
robust testing and
defense mechanisms.
Companies specializing
in AI security services
could see increased
demand as businesses
seek to protect their
AI investments from
manipulation and
exploitation.
Model
Commodification
Open-source language
models with commercial
licensing, such as
Databricks’ Dolly, could
disrupt the market by
offering high-quality
capabilities at a
fraction of the cost. This
commodification poses
an existential threat to
proprietary models from
big tech companies.
Defense Sector
Innovation
Updated Department
of Defense policies on
autonomous weapons
and the use of AI in
military strategies signal
growing opportunities
for businesses in
the defense sector.
Companies developing
AI technologies could
find new applications in
warfare, surveillance, and
security, but also face
ethical and regulatory
scrutiny.
Strategic International
Collaborations
Countries like China
and the UAE are
heavily investing in
becoming global AI
leaders, which presents
both opportunities
and challenges for
international business
collaborations.
Companies might need
to align with national
AI strategies to enter or
expand in these markets,
while also considering
the implications of
technology transfer and
data security regulations.

TECH
62
WHEN WILL ARTIFICIAL INTELLIGENCE DISRUPT YOUR ORGANIZATION?
AI WILL
DISRUPT
EVERY
INDUSTRY
WITHIN THE
NEXT FIVE
YEARS
Drawing a parallel to Moore’s law, which posits the doubling of transistors on
microchips roughly every two years, there’s speculation that AI’s intelligence
could follow a similar trajectory. If this is the case, several factors will drive this
exponential growth in intelligence: enhancements in data quality, increasing
computational power, and strides in algorithm efficiency, extracting more
intelligence per unit of data and compute.
However, unlike the steady hardware advancements Moore’s law describes,
AI has the potential for self-improvement. As AI begins to self-improve and
contribute to its own development, we may witness a self-reinforcing cycle
of intelligence growth. This positive feedback loop means that AI’s capacity
to learn and evolve could accelerate, leading to profound impacts across all
industries.
The inevitability of AI-driven transformation is not a matter of if but when. Our
AI report is one section of our 2024 Tech Trends report, which offers in-depth
coverage of 15 additional technology and industry sectors. Each industry
section contains timelines that outline how AI, along with other emerging
technologies, are expected to impact and influence that particular sector over
time. Refer to the “When will AI impact your organization?” page to find details
on specific timelines related to AI adoption and impact on your industry.

TECH
63
Below, we highlight high level near-term developments to keep an eye on across industries.
Scaling
Enormous amounts of training data are still required for most AI
models to learn. For example, recommender systems coupled with
generative AI could lead to deep personalization for the hospitality
and health care sectors—as long as data is made available. Histor-
ically, data is locked inside proprietary systems built by third par-
ties, and regulation often hinders access to certain forms of data.
Investment
AI has passed through cycles of enthusiasm and disillusionment,
leading to either too much or not enough capital being made
available. Investors prioritize commercialization over basic RD—
though the latter yields bigger impact and often stronger returns.
Investors’s patience will influence progress and commercialization.
Constraints on adoption
Even if a technology is maturing, constraints on its adoption can
hinder its impact on an industry. For example, a business may re-
fuse to adopt an automated system because it challenges existing
orthodoxy or an existing successful strategy. This is especially true
in health care, insurance, and financial services.
Regulations
Advances in technology typically outpace regulatory changes. This
has benefited AI, which until very recently was not targeted for
regulation. Additionally, whether local regulations are conflicting or
complementary, influences adoption in the marketplace.
Media mentions
Increased awareness and enthusiasm can influence the momen-
tum of a technology, even when there’s been no real breakthrough.
Until OpenAI’s ChatGPT breakthrough in late 2022, leaders weren’t
talking about the impact genAI might have on their business.
Media bursts related to AI will drive momentum, especially if those
stories are favorable, and more importantly, are easily understood
by the public.
Public perception
How the public understands and responds to AI advancements
will create or quell demand. This is especially true of generative AI
and education/creativity/ intellectual property/misinformation,
and the role assistive technologies will play in shaping the future
workforce.
RD developments
The pace of new research breakthroughs can’t be scheduled to
coincide with a board meeting or earnings report. Factors like fund-
ing, quality, and size of staff, and access to resources can improve
the likelihood and speed of new discoveries. We closely monitor
RD developments but treat them as wild cards.
WHEN WILL ARTIFICIAL INTELLIGENCE DISRUPT YOUR ORGANIZATION?

Threats
It’s possible for agents to learn the right skills but the wrong objectives; an AI system can be asked
to learn something that then could be used for harmful purposes. Commercial AI products could
inadvertently incentivize bad behavior.
Publicly available LLMs are often the foundation for AI startups, but some researchers and technologists
are questioning their defensibility when it comes to capturing value. The moat is in data. Techniques and
models will largely get commoditized, and served via the infrastructure layer, where real value will be
realized.
Long-term sustainability depends on network effects to gather enough user data. User-generated
data can be harnessed to differentiate systems by offering tuned models on top of foundational/
commoditized LLMs, creating a flywheel effect. Longer term, niche LLMs will be owned by a select few
players, while general-purpose LLMs become commoditized.
The challenge of balancing data collection for workflow optimization with concerns of worker
surveillance requires careful navigation by companies. AI’s use and understanding of behavioral
biometrics could be considered intrusive into deeply personal behaviors, often subconscious to the
individual, starkly confronting worker privacy expectations.
Heightened protectionism across nations could escalate the costs of producing chips and other critical
technologies, and make it more difficult to find the right talent. Companies should brace for the adverse
economic impacts of geopolitical shifts as supply chains undergo realignment.
AI models might achieve assigned goals by any means necessary, including suppressing or hiding data.
Systems are needed to identify when this happens—until then we risk using bad information to make
decisions.
High-performing models are susceptible to “jailbreaking,” where bypassing LLM limitations can lead to
manipulations, resulting in unpredictable and potentially harmful outputs. Given that businesses and
entire institutions are starting to rely on LLMs, jailbreaking represents an urgent security threat that has
yet to be addressed.
Opportunities
AI is on track to become an indispensable tool for knowledge workers. The next 18-24 months will see the
development of assistive technologies tailored to various professions, akin to GitHub’s Copilot, but designed
for financial analysts, commercial real estate developers, and lawyers.
Companies sitting on industry-specific data hold the cards to create powerful AI agents. In industries like
law, finance, and other knowledge-based sectors, proprietary data can train more capable AI agents.
Within the next 18-24 months, generative AI will integrate into many consumer apps. Where clicks and
keywords once dominated, intelligent assistants will guide users through voice and text. Personalized
support gets weaved throughout experiences, changing how people engage with information.
AI models that understand language will lead to more devices that enable people to interact with technology
through voice and conversation instead of screens. Opportunities await for companies quick to challenge
status quo screen-centric form factors.
AI is going local. Wearables and endpoints of all kinds will be embedded with AI, from pets’ collars that
report on animals’ activities, to smart home devices that understand and execute complex commands from
natural language. Large language models will migrate on-device, perhaps in lieu of a conventional operating
system.
The rising energy needs of AI could incentivize tech companies to adopt alternative, greener energy sources
like nuclear and geothermal, potentially driving a shift toward sustainable energy independently of
government mandates.
Open-source models allow businesses and developers to adapt and enhance foundational models for
specific uses, saving the cost and effort of starting from scratch or investing heavily in data and training.

Create domestic intern-
ship and apprenticeship
programs to build talent
pipelines in AI skills, where
shortages loom. Partner
with schools to develop
a homegrown workforce
proficient in these tech-
nologies vital for national
strategic interests.
Investing in data centers
powered by renewable
energy or exploring part-
nerships with alternative
energies like nuclear and
geothermal could align AI
operations with ESG goals,
reducing the carbon foot-
print of data processing
and storage, and reducing
the cost of compute.
Nvidia dominates the GPU
market, yet demand out-
paces even its cutting-edge
chips. With shortages
routine, space exists for
rivals while cloud partners
hunger for inventory.
As AI models grow in com-
plexity, investing in alter-
native computing architec-
tures like neurosymbolic
AI, processing-in-memory
technology, and special-
ized AI chips for on-device
processing could offer
significant advantages
in efficiency, speed, and
privacy.
Build atop shared foun-
dations. Open-source
models like LLaMA and
FLAN offer springboards
to launch specialized
solutions tuned to distinct
industry needs. These spe-
cialized models are more
accurate and focused to
the industry they serve
and give proper weight to
relevant parameters.
Foster development of
small language models
(SLMs). Investing in the
research and deployment
of SLMs suitable for edge
devices can open new
avenues for AI applica-
tions in environments
where cloud connectivity
is limited or nonexistent.
SLMs can significantly
expand the reach of AI
into everyday devices, en-
hancing user experience
and functionality.
INVESTMENTS AND ACTIONS TO CONSIDER
1 4
2 5
3 6

CENTRAL THEMES
66
New data sources are coming
The integration of hardware, particularly wearables, will
redefine the landscape of data collection and utilization.
Coming to market soon are an array of different wearable
devices equipped with sensors, cameras, and speak-
ers, and they represent a significant leap forward in our
ability to gather real-time, contextual data. This evolution
marks a future where the volume of data available for
analysis will expand exponentially, offering unprecedent-
ed insights into consumer behavior and environmental
interactions. The challenge for organizations won’t just
be in the collection, but in the sophisticated parsing and
interpretation of this deluge of data, requiring advanced
AI algorithms and analytical frameworks.
Race for AI hardware supremacy
The intersection of hardware development and geopolit-
ical competition is reshaping the landscape of AI ad-
vancement, with implications spanning national security,
technological sovereignty, and economic prowess. As
governments worldwide vie to establish AI supremacy
and reduce dependence on foreign technology, substan-
tial investments are pouring into domestic chip fabrica-
tion and AI research. The US and China, in particular, are
locked in a battle for technological dominance, with both
nations allocating significant resources toward bolster-
ing their respective chip capabilities and AI infrastruc-
ture. This geopolitical rivalry extends beyond economic
competition, with ideological considerations shaping
AI development strategies and regulatory frameworks.
China’s insistence on AI alignment with socialist values
underscores its commitment to ideological control, while
Russia perceives Western AI advancements as a threat to
traditional values, driving efforts to develop indigenous AI
solutions. Meanwhile, escalating tensions have catalyzed
a bifurcation in the AI chip market, prompting countries
to explore alternative chip architectures and supply chain
diversification strategies. This unfolding chip war not
only underscores the strategic importance of semicon-
ductor technologies but also poses profound implica-
tions for global technological cooperation and innovation.
Chip shortages loom large
The surging demand for AI has highlighted the global
supply chain’s inability to meet the need for powerful
chips essential for developing and deploying AI mod-
els. We predict a chip shortage, particularly for graphics
processing units (GPUs), due to production issues and
ongoing shipping challenges due to regional conflicts.
Microsoft’s recent annual report marked the scarcity of
GPUs as a potential risk for investors, underscoring the
critical role these chips play in AI development and the
broader implications for companies and end-users reliant
on AI technologies. The industry as a whole will grapple
with limited supply and the challenge of meeting explo-
sive demand, prompting a shift toward more efficient or
alternative computational methods. Maybe that’s why in
February 2024, OpenAI CEO Sam Altman reportedly went
on a business development tour seeking $7 trillion in
investment to create an alternative to our current chips.

CENTRAL THEMES
67
Choosing between proprietary and open source
Last year, when Meta released LLaM, its suite of open
source LLMs, there was a new debate about the benefits
and risks of going open source. Organizations using large
language models face a challenging decision: Go with the
big names like OpenAI and Microsoft for easy access to
top-notch tech but give up adaptability and transparency,
or push up your sleeves and build your own tailor-made
systems to ensure transparency and extensibility. Despite
the steep development costs associated with proprietary
LLMs, the open-source community has responded with
notable alternatives, such as Databricks’ Dolly LLM, which
offers a solution at a fraction of the cost. The new shift
toward open-source solutions aims to counterbalance the
growing concentration of AI tools in the hands of a few
major corporations, offering businesses the opportunity
to integrate bespoke applications without compromising
proprietary information.
Reckless era ends, oversight era begins
The era of “move fast and break things,” and “build first,
ask permission later” appears to be waning in Silicon
Valley as regulatory scrutiny intensifies in response to
growing concerns over AI’s societal impacts. With initia-
tives like a US presidential executive order and the EU’s
AI Act, policymakers are striving to establish guidelines
and restrictions to govern AI technologies, particularly in
sensitive areas like facial recognition. However, crafting
concrete policies that balance innovation with ethical
considerations, poses significant challenges, and ensur-
ing effective enforcement remains a formidable task. As
governments grapple with the complexities of regulating
AI, the tech industry faces a new era of accountability and
responsibility for the products they create.
AI doomers distract
Amid the discourse surrounding AI, a contingent of pes-
simistic voices, often referred to as “AI doomers,” has
emerged, likely to persist in the foreseeable future. For
business leaders, navigating this landscape proves chal-
lenging, as they are presented with polarizing narratives
of either utopian ideals or dystopian anxieties, resulting
in a nuanced yet unsettling reality. While it’s crucial to
remain vigilant against potential risks and mitigate them
effectively, the prevalence of doomerism tends to over-
shadow constructive dialogue and proactive measures.

CENTRAL THEMES
68
Industry is building the future of AI, not academia
The landscape of innovation is shifting, with industry
emerging as the primary driver of technological advance-
ment, outpacing academia in the development of new
machine learning models. Recent data reveals a stark
contrast: in 2022, industry produced 32 machine learning
models compared to academia’s three, marking a signif-
icant departure from historical trends. Industry’s domi-
nance is further underscored by its access to abundant
resources—large data sets, computational power, and
financial capital—essential for creating cutting-edge AI
systems. This transition is reflected in the career choices
of AI Ph.D. graduates, with 65.4% opting for industry posi-
tions, compared to 28.2% in academia, a trend that has
steadily widened since 2011. The exodus from academia
to corporations could have a chilling, long-term effect on
knowledge transfer from professors to students, which
could negatively impact the future pipeline for the talent
industry which will need to remain competitive.
AI widens global inequality gulf
The exorbitant costs associated with training language
models are setting a precedent for the formidable ex-
penses expected in developing image and video models,
further accentuating disparities in resources between dif-
ferent regions and exacerbating the global divide between
the affluent and less affluent nations. This trend not only
reshapes the landscape of business and communities
but also positions wealthy countries, notably the United
Arab Emirates and Saudi Arabia, as potential hubs for AI
development, potentially marginalizing opportunities for
advancement in the global south.

Dr. Aidan Gomez, CEO and co-founder of
Cohere, for proposing the novel neural network
technique called the transformer that now un-
derpins the generative AI era.
Arthur Mensch, Dr. Guillaume Lample, and Tim-
othée Lacroix, co-founders of European genera-
tive AI upstart Mistral AI.
Dr. Andrej Karpathy, researcher at OpenAI for
his research in deep learning and computer
vision.
Clément Delangue, CEO and co-founder of
Hugging Face, for creating an open-source,
for-profit machine-learning platform.
Dr. Daniel Kang, assistant professor at Uni-
versity of Illinois Urbana-Champaign , for his
research identifying potential harms from
language models, including demonstrating
language models’ ability to autonomously inter-
act with websites in concerning ways without
human feedback, and his work to develop
methods that promote the safe and ethical
development of AI.
David Nippa, a doctoral student at Ludwig-Max-
imilians-Universität München, for the develop-
ment of an AI model that can predict where a
drug molecule can be chemically altered.
Dr. Dario Amodei and Daniela Amodei, CEO
and president of Anthropic, for creating one of
the world’s leading AI labs.
Dr. David Rolnick, assistant professor of com-
puter science at McGill University, for work on a
framework for understanding the relationship
of AI and greenhouse gas emissions.
Grimes, artist and musician, for championing
new business models around AI for likeness
leasing and creative experimentation.
Dr. Jaime Teevan, chief scientist and technical
fellow at Microsoft, for spearheading the use of
LLMs in Microsoft’s core productivity products.
Jensen Huang, CEO, president, and co-founder
of Nvidia, for navigating the growing geopoliti-
cal chip conflict.
Dr. Joelle Pineau, vice president of AI research
at Meta, for developing new models and algo-
rithms for planning and learning in complex
partially observable domains.
Leopold Aschenbrenner, AI alignment re-
searcher at OpenAI, for his contributions to AI
alignment discourse.
Lila Ibrahim, COO of Google DeepMind, for
leading the company’s responsibility and gover-
nance work.
Marc Raibert, executive director at Boston
Dynamics AI Institute, for his work to develop
AI-driven robots that can reason.
Miguel Solano, co-founder and CEO of VMind,
for his work to improve AI compute performance
in GPUs using novel algorithmic techniques.
Dr. Ning Zhang, an assistant professor of com-
puter science and engineering at Washington
University, for the development of AntiFake, a
tool that prevents unauthorized speech synthe-
sis.
Dr. Prakhar Mehrotra, vice president for
applied AI at Walmart Global Tech, for leading
enterprise adoption of AI.
Robin Li, CEO, chairman and co-Founder of
Baidu, which last year released Ernie Bot, an
LLM on par with ChatGPT.
Dr. Ruogu Fang, an associate professor in the J.
Crayton Pruitt Family Department of Biomedical
Engineering, for his work to evaluate diagnostic
bias in AI tools.
Sebastien Krier, international policy manager
at DeepMind, for his research and intellectual
contributions to AI alignment discourse.
Dr. Sune Lehmann, professor at the Technical
University of Denmark , for research into the
predictive capabilities of AI, specifically its po-
tential to forecast events in an individual’s life.
Dr. Swami Sivasubramanian, vice president
of database, analytics, and machine learning
at Amazon Web Services, for advancing cloud
capabilities and insights for businesses.
Dr. Xin (Eric) Wang, assistant professor of
computer science and engineering at Baskin
Engineering at UC Santa Cruz, for the develop-
ment of the Text to Image Association Test, a
tool that measures complex human biases in
text-to-image models.
Dr. Zhou Jingren, deputy director of Aliba-
ba Damo Academy (Alibaba’s bleeding-edge
research arm), for leading AI initiatives related
to smart cities, autonomous driving, mobile
computing platforms, semiconductor RD, and
other areas.
ONES TO WATCH

IMPORTANT TERMS
70
MACHINE LEARNING (ML)
ML uses data to make predictions and recommen-
dations on how to achieve stated goals. AI pioneer
Arthur Samuel popularized the idea of machine
learning in 1959, explaining how computers could
learn without being explicitly programmed. This
would mean developing an algorithm that could
someday extract patterns from data sets and use
those patterns to predict and automatically make
real-time decisions. It took many years for reality
to catch up with Samuel’s idea, but today machine
learning is a primary driver of AI’s growth.
There are different types of machine learning, including
supervised, unsupervised, and reinforcement.
Supervised learning
A model that attempts to transform one type of
data into another type using labeled examples.
Supervised learning is used when teams know how
to classify the input data and what they are trying
to predict, but can get accurate results much more
quickly by relying on an algorithm rather than a
human. This is the most common form of ML used
today. Understanding what product features would
most likely drive new purchases is a business use
case for supervised learning.
DEEP LEARNING (DL)
Deep learning is a relatively new branch of ma-
chine learning. Programmers use special deep
learning algorithms alongside an enormous corpus
of data—typically many terabytes of text, images,
videos, speech, and the like. Often, these systems
are trained to learn on their own, and they can sort
through a variety of unstructured data, whether it’s
making sense of typed text in documents or audio
clips or video.
In practical terms, deep learning’s emergence
means that more and more human processes will
be automated, including the writing of software,
which computers will soon start to do on their own.
For example, once a system learns what an object
looks like—say, an apple—and then can recognize
that object in all other images, even if it has only a
partial view.
There are different types of deep learning architectures.
The most common types include convolutional neural
networks, recurrent neural networks, transformer neural
networks, and generative adversarial networks (GANs).
Convolutional neural network (CNN)
A CNN is multilayered, with a convolutional layer, a
pooling layer, and a fully connected layer. Each one
performs a different task with the data. The output
is classification. If a researcher has 10,000 images
and needs to extract data—to recognize particu-
lar faces, for instance—the CNN would run until
information could be inferred. In business, CNNs
are used to identify anomalies in medical imag-
ing, faulty products on a production line, blight on
crops, and other irregularities.
Recurrent neural networks (RNNs)
These multilayered neural networks move and store
information between input, hidden, and output
layers. They are good at modeling sequence data for
predictions. In business, they are used anytime the
sequence of data matters, such as speech recog-
nition and language translation. RNNs are used in
digital assistants, to create captions for images,
and to generate narrative reports using structured
data (sports, financial).
Transformers
A transformer is a component whose purpose is to
process sequential data, such as natural lan-
guage or genome sequences. Transformers rely on
“attention” (the mathematical description of how
things relate to, complement, or modify each other)
in translating sequences. A transformer neural
network is the unique architecture that enables
systems to learn from context and to generate new
Unsupervised learning
Data is provided to a model without specific output
parameters, and the model tries to learn the data
set’s structure without any designated labels. For
example, if a researcher doesn’t know what to do
with a large data set, an unsupervised learning
model could determine patterns, classify data,
and make recommendations without a human
supervisor. Researchers used unsupervised learn-
ing during the pandemic to find patterns on how
COVID-19 spread throughout communities.
Reinforcement learning (RL)
A system performs a task by repeatedly running
calculations as it attempts to accomplish a stated
goal. It’s a trial-and-error process, where rewards
or penalties are earned in response to the system’s
performance toward achieving the stated goal. RL is
used when there isn’t enough training data, when
the researcher is trying to learn about an environ-
ment (such as a complex financial portfolio), or
when the researcher needs to find greater levels
of optimization. It has a high number of business
use cases, ranging from real-time dynamic pricing
models to high-frequency trading algorithms to the
systems that operate self-driving cars.

IMPORTANT TERMS
71
information. Transformers are complementary to
CNNs and RNNs, the two most common neural net-
work architectures used in deep learning.
Generative adversarial networks (GANs)
As unsupervised deep learning systems, GANs are
composed of two competing neural networks—a
generator and a discriminator—that are trained
on the same data, such as images of people. The
networks compete against each other to perform
a task, such as identifying the correct person, re-
sulting in optimizing overall performance. GANs are
useful when researchers don’t have enough data
to train an algorithmic model, and are also used to
create new, synthetic data.
Deepfakes, which have become prevalent in the
past year, are generated using GANs. In design,
GANs are tremendously useful: They can produce
thousands of designs and recommend the best
ones based on pre-set parameters. They can gen-
erate and modulate voices, faces, even gestures.
Researchers from Nvidia, Mass General Hospital,
BWH Center for Clinical Data Science, and the Mayo
Clinic collaborated on a GAN that generates syn-
thetic MRIs showing cancerous tumors.
Automatic speech recognition
Algorithmic systems that give computers the abil-
ity to recognize and convert audio to human-read-
able language.
Chain of Thought
This involves a model processing information or
solving problems step by step, mimicking hu-
man-like reasoning.
Computer vision
Processes that give computers the ability to derive
meaningful information from digital images (in-
cluding still and video) and to mimic and manipu-
late such images.
Foundation model
A large-scale AI model trained on vast amounts of
data, capable of being adapted to a wide range of
tasks without being trained from scratch.
Generative AI
GenAI refers to AI technologies that can generate
new content, including text, images, music, and
video, based on learned data patterns.
GPU
A graphics processing unit is specialized hardware
designed to accelerate the creation and rendering
of images and videos, often used in AI for parallel
processing tasks.
Model
A program that has been trained on a data set.
Models are generally used for analytical and deci-
sion-making tasks, such as making predictions.
Natural language processing
Processes that give computers the ability to under-
stand, mimic, and manipulate human language.
Parameter
A variable internal to the model that the system
adjusts during training to improve performance on
given tasks.
Prompt
An input given to a model to elicit a specific output
or response, guiding the AI in generating content or
solving problems.
Recommender systems
A class of machine learning algorithms that uses
data to predict, narrow down, and find what people
are looking for among an exponentially growing
number of options.
ADDITIONAL TERMS
Agents
In AI, agents are entities that perceive their envi-
ronment and take actions autonomously to achieve
specific goals.
AGI (artificial general intelligence)
A designation for systems that match and then
exceed the full range of human cognitive ability
across all economically valuable tasks.
AI safety
A field that studies and attempts to mitigate the
catastrophic risks that future AI could pose to
humanity.
Algorithm
A process describing how to solve a specific prob-
lem or how to complete a particular task.
Alignment
The process of ensuring that an AI’s actions and
goals are in harmony with human values and inten-
tions.
ASI (artificial superintelligence)
ASI refers to an AI system that surpasses human in-
telligence and capability across all fields, including
creativity, general wisdom, and problem-solving.

IMPORTANT TERMS
72
RHLF
Reinforcement Learning from Human Feedback is a
training method where AI models are refined based
on feedback or corrections provided by humans,
enhancing their performance and alignment with
desired outcomes.
Supervised learning
A type of AI training where models learn from
labeled data, using known input-output pairs to
predict outputs from new inputs.
Symbolic AI
Symbolic AI involves AI systems that use explicit,
human-readable symbols to represent knowledge
and perform logical reasoning to solve problems.
Training data
The data set used to teach AI models how to under-
stand and perform tasks by identifying patterns,
making decisions, or generating predictions.
Unsupervised learning
Unsupervised learning involves AI models identi-
fying patterns and structures in data without any
labeled outcomes, learning from the data itself.
XAI (explainable AI)
AI systems designed to provide human-under-
standable insights into their decision-making
processes, enhancing transparency and trustwor-
thiness.
Zero-shot learning
An AI approach that enables models to correctly
handle tasks or recognize objects they have not
seen during training, using understanding from
related contexts.

MODELS,
TECHNIQUES,
AND
RESEARCH

TECH
WHAT IS AN AI MODEL?
An AI model is a computational structure that is designed to perform tasks
that would normally require human intelligence. This includes recognizing
speech and images, interpreting visuals, translating between languages, and
making decisions. There are several types of AI models, each suited to spe-
cific tasks and goals but at their core, all AI models rely on algorithms and
mathematical frameworks. They are “trained” on large data sets so they can
refine their internal parameters and improve at assigned tasks. As AI systems
become more advanced, they require more data and computing power during
this training process.
Constructing AI models is an enormously resource-intensive process, not
comparable to traditional software development. Training a high-performance
language model demands processing huge data sets to fine-tune millions of
parameters. This mandates extensive computing power and specialist time
measured in months or years. As a result, advanced models are predomi-
nantly built by tech industry leaders like Google, Microsoft, and OpenAI who
possess the vast technical infrastructure and talent required. The consum-
er focus and profit motive within these companies have accelerated model
innovation beyond academic efforts. Historically, academia was seen as the
most likely source of groundbreaking AI. But the sheer data scale, computing
power, and engineering capacity within industry, has proven far more effi-
cient for allocation of resources.
Examples
Note: All of these examples are
current as of March 1, 2024
ChatGPT-4
OpenAI’s most recent model as
of publication, GPT-4, doesn’t
just generate text—it can
generate images from text and
vice versa. It was trained on
enormous data sets of text and
images using reinforcement
learning from human feedback
(RLHF), which helped make the
model more helpful and safer
for users. Early benchmarks
exhibit GPT-4’s versatility on
tasks from legal exams to ad-
versarial truthfulness tests. On
the Uniform Bar Exam, it scored
90% versus GPT-3.5’s 10%,
while reducing factual errors
by 40% compared to ChatGPT.
While hallucination risks still
exist, GPT-4 marks substantial
progress in mitigating failure
modes.
Gemini
After an underwhelming debut
in 2022, Google iterated to the
more impressive Gemini Pro in
early 2024. This model demon-
strates rapid advances, as
evidenced by its meteoritic rise
up the Hugging Face conversa-
tional AI leaderboard. Google’s
Gemini isn’t just a single AI
model—it encapsulates a suite
of AI models for varied appli-
cations. Gemini Nano targets
offline Android use. Gemini
Pro now powers Bard and
emerging enterprise services.
Gemini Ultra is Google’s most
advanced large language model
yet, designed to elevate search,
advertising, and cloud products
globally.
Claude 2.1
Anthropic unveiled Claude
2.1, the latest in its series of
language models, capable of
processing significantly longer
texts than OpenAI’s GPT-4.
With the ability to manage up
to 200,000 words or sym-
bols, Claude 2.1 significantly
surpasses GPT-4’s limit of
128,000. The new iteration of
Claude is designed to reduce
the likelihood of generat-
ing inaccurate information
compared to earlier versions.
A significant enhancement in
Claude 2.1 includes its ability
to utilize tools and interface
with APIs. Additionally, the in-
troduction of system prompts
allows users to define precise
contexts for their inquiries,
promoting more organized
and reliable responses from
the model.
PaLM 2
PaLM is a 540 billion parame-
ter language model developed
by Google AI. Smaller 8 billion
and 62 billion parameter
versions were also trained.
PaLM demonstrates strong
performance across common-

TECH
sense reasoning, math reason-
ing, humor, code generation,
translation, and other tasks.
The model highlights Google
AI’s advances in scalable trans-
former architecture research
for language AI.
Whisper
Whisper is an open-source
automatic speech recognition
system created by OpenAI.
First released in 2022, it was
trained on over 680,000 hours
of multilingual speech data
scraped from the internet.
Whisper can transcribe speech
to text in multiple languages
including English. It can also
translate speech from non-En-
glish languages into English
text. Compared to other publicly
available systems, Whisper
demonstrates leading speech
transcription and transla-
tion capabilities. OpenAI has
released the model freely for
public use.
OpenAI’s DALL-E 3
DALLE-3 is a text-to-image AI
system that can create realistic
art and images from textual
descriptions. DALL-E is capable
of generating images in various
styles like photorealistic, paint-
ings, and emoji. Without explicit
prompting, the model can ma-
nipulate and rearrange objects
as well as correctly position de-
sign elements in new composi-
tions. These creative capacities
demonstrate DALL-E’s aptitude
for controllable high-fidelity
image generation.
Stability AI’s Stable
Diffusion
Stable Diffusion is a text-to-im-
age generation model released
in 2022 leveraging diffusion
methodology. Primarily used
to create detailed images from
text prompts, Stable Diffusion
can also perform tasks like in-
painting, outpainting, and im-
age-to-image translation driven
by descriptive text inputs.
Midjourney
Midjourney creates visuals
based on textual descriptions,
known as prompts, akin to
the functionalities offered by
OpenAI’s DALL-E and Stability
AI’s Stable Diffusion. A fake Mi-
djourney-created image of Pope
Francis wearing a puffer jacket
went viral in 2023.
Open AI’s Sora
In early 2024, OpenAI released
Sora, an AI model that can
create realistic and imagina-
tive scenes from text instruc-
tions. Sora marks a significant
advancement in AI’s capability
to execute human creativity by
transforming brief text inputs
into compelling videos up to a
minute long, not only achieving
realistic imagery but also emu-
lating the dynamic essence of
movies, similar to how ChatGPT
mimics human conversation.
Google Lumiere
Google’s Lumiere is a text-to-
video diffusion model that
creates video from a prompt
with realistic motion. Utilizing
a novel diffusion model named
Space-Time-U-Net (STUNet),
Lumiere excels in creating
realistic video content by
understanding both spatial
placement and temporal move-
ment within a video. Unlike
other methods that assemble
videos from individual frames,
Lumiere crafts videos through a
seamless integration of frames,
achieving fluid motion across
80 frames—significantly more
than its current competitors.
Pika
Pika is an “idea-to-video” plat-
form to edit and create videos
from text and still images. Pika
includes features like text-
to-video, image-to-video, and
video-to-video conversions.
Users can ask the tool to create
a video of a real person (e.g.,
“imagine Oprah as a Pixar
cartoon”), ask the tool to edit
glasses on a video of a donkey,
or change the style of a video to
something out of Studio Ghibli.

TECH
Purpose-Built Models
Organizations must decide whether to use
ready-made general purpose AI models like
OpenAI’s GPT, or invest in developing custom
models tailored to their industry and needs.
General-purpose models like GPT are conve-
nient “plug-and-play” solutions that can adapt
to many tasks through fine-tuning. However,
their flexibility is limited when it comes to spe-
cialized business challenges. Custom models
built for a specific purpose can better master
industry-specific challenges by training on
aligned data and objectives. OpenAI now pro-
vides a simple way for users to create custom
models through the GPT marketplace—users
describe their requirements to ChatGPT, and it
handles coding the new model. Custom GPTs
can then be integrated into platforms and
services, accessing databases, email, e-com-
merce, and more to automate workflows.
LLMs Are Getting Bigger and More Expensive
Because of their massive size and complexity,
the cost of developing LLMs is high. Training
these models can cost millions of dollars.
DeepMind’s Chinchilla, for example, reportedly
1.4 trillion pieces of text over 21 days, which
amounts to nearly 1 million hours of GPU
time. If using public cloud services, this level
of compute would cost approximately $2.4
million. Despite its impressive capabilities,
with “only” 65 billion parameters, LLaMA is
still smaller compared to larger models like
OpenAI’s ChatGPT-4, which has 1.76 trillion
parameters.
LLMS as Operating Systems
A radical new concept for computing has
emerged—an operating system powered by a
large language model at its core rather than
traditional programming. In this conceptual
LLM-based OS, routine tasks could be auto-
mated and executed with an unprecedented
level of sophistication, without the need for
manual coding or intervention. The user in-
terface would also be radically different than
traditional operating systems. Rather than
conventional graphical user interfaces or
command line prompts, users could interact
conversationally with the LLM through natural
language requests and queries. For example,
a user could say, “Please open yesterday’s
cost around $2.1 million to train. Bloom, an
open-access multilingual language model,
is estimated to have required an investment
of approximately $2.3 million. OpenAI hasn’t
provided public information about the cost to
train ChatGPT-4, but many analysts estimate
the earlier version of the model, GPT-3, could
exceed $4 million.
As the number of parameters increases, so
does the cost. Moreover, unlike traditional
software, deployment costs remain high
post-development. Operating large language
models for inference still necessitates
enormous compute for the billions of calcu-
lations involved per user query. Furthermore,
contributing to the high price tag of training
and running large language models de-
mands specialized AI hardware, with graphics
processing units (GPUs) now standard over
traditional CPUs. Initially designed for gam-
ing, GPUs are perfectly suited for handling
the extensive data processing demands of
AI, despite their high cost of thousands of dol-
lars per chip. For example, Meta used 2,048
Nvidia A100 GPUs to train its LLaMA model on
While general-purpose models offer broad applica-
bility, their limitations in specialized contexts are
driving the development of models tailored to meet
the unique demands of specific industries.
Image credit: Future Today Institute and Midjourney.

TECH
sales report and format it as a slide presenta-
tion for my upcoming meeting.” The LLM would
comprehend these conversational commands
and perform the necessary actions to carry out
the desired tasks. It would execute complex
workflows automatically by understanding
users’ intentions and goals. This could enable
more intuitive, efficient interactions between
humans and computers.
This concept has moved beyond theory into
practical application, as demonstrated by
Jesse Lyu, CEO and founder of Rabbit. Lyu
launched the R1, a compact device about half
the size of an iPhone, running on Rabbit OS—
an operating system grounded in a LLM. Rabbit
OS functions as a universal app controller,
akin to systems like Alexa or Google Assistant,
yet it offers a unique twist. It simplifies user
interaction by removing the need to navigate
through multiple apps or perform repetitive
logins. Instead, users can directly communi-
cate their needs to the device, and R1, under-
standing these natural language requests,
efficiently executes the desired tasks. In early
2024, NVIDIA announced a personalized AI
chatbot for Windows RTX PCs that runs lo-
cally to connect users’ data and queries to an
open-source large language model. By keep-
ing data on device rather than in the cloud,
Chat with RTX not only delivers ultra-fast
response times but also enhances user
privacy and data security. The chatbot allows
natural language interaction to search files
so that users can simply ask, “What was that
song my friend recommended while we were
at the airport?” instead of manually searching
through texts and email.
LLMs are becoming more central to human-computer interactions. As such, interfaces are shifting from
search to conversational questions and answers in plain language.

TECH
SHOULD WE GO OPEN-SOURCE OR PROPRIETARY?
Companies that want to use LLMs must choose between proprietary or open
source. Both have benefits and drawbacks. Proprietary LLMs from major tech
companies provide easy implementation and leading-edge features. However,
they lack transparency into how they work and have limited ability to custom-
ize them. Building a proprietary model internally gives companies more con-
trol over security, privacy, and tailoring the training to their specific data and
needs. But this requires considerable expertise and time to develop. On the
other hand, open-source language models promote transparency and flexibili-
ty at often lower, long-term costs. Yet if governance practices like testing for bi-
ases and false information are insufficient, they pose risks around issues like
fairness, accuracy, and security vulnerabilities. When considering using LLMs,
executives should think about cost, control, customization, and risk. There is
no universally superior choice—rather, companies must weigh their priorities,
capabilities, and constraints to determine if an off-the-shelf, customized, or
open-source large language model approach best suits their situation.
We have our own nervousness, but we believe that
we can manage through it, and the only way to do
that is to put the technology in the hands of people.
—Sam Altman, CEO of OpenAI
Proprietary examples:
OpenAI’s GPT-4
Anthropic’s Claude 2
Google’s Bard
Open-Source examples:
Meta’s LLaMA
RedPajama-INCITE
BigScience’s Bloom
TII’s Falcon
Open-Source LLMs for Commercial Use
Proprietary large language models cost millions to develop,
which means high-quality capabilities are concentrated within
wealthy tech giants. However, the open-source community
has responded with surprisingly capable smaller models by
fine-tuning them on quality data. For example, in March 2023
Databricks released Dolly—an open-source LLM trained for
under $30 yet demonstrating conversational prowess rivaling
ChatGPT. It was developed using Meta’s open-source LLaMA
LLM and fine-tuned with high-quality inputs from Databricks
employees. The initiative aimed to provide an alternative to
the increasing centralization of AI tools in a few large compa-
nies, focusing on an open-source chat model that permitted
commercial use while protecting intellectual property and
corporate information. Databricks not only open-sourced the
training code, data set, and model weights for Dolly but also
launched Dolly 2.0 in April 2023. Dolly 2.0 is open-source LLM
licensed for commercial use, allowing companies to integrate
their data with Databricks’ data set to create bespoke applica-
tions without compromising their proprietary information.

SAFETY,
ETHICS
SOCIETY

TECH
IS AI REALLY A BLACK BOX?
vide interpretable visibility into significant
aspects of functionality. This can involve
revealing training data characteristics,
delineating gaps in data coverage, auditing
data collection fairness, detailing human
involvement in model development, and
highlighting key input features that drive
outputs. A core focus is validating outcomes
by surfacing how predictions, classifica-
tions and recommendations are supported
to establish trustworthiness. Rather than
eliciting every intricate internal model
transformation, XAI pursues strategic expla-
nations of the most critical workings—an-
swering targeted questions about why and
how certain results are produced. The objec-
tives are accountability through limited but
meaningful transparency, error checking via
result explanations, and accessibility for a
wider range of model users.
AI Intentionally Hiding Data
Computers do exactly as they are told. If you
command a machine to win at a game, it will
do everything in its power to achieve that
Explainable AI (XAI)
Achieving full transparency into complex AI
systems is difficult. However, the emerging
field of explainable AI seeks to enable better
human understanding of how algorithms
function and arrive at outputs. Since com-
plex machine learning models cannot act as
total glass boxes, XAI instead seeks to pro-
Many AI systems are opaque “black
boxes” in how they work. Developers
often withhold model and train-
ing details to protect IP. This lack
of transparency perpetuates an
impression that the systems have
unknowable inner workings. More-
over, researchers themselves don’t
fully understand why AIs sometimes
behave unexpectedly, owing to in-
herent complexity. While inputs and
outputs are observable, the logic be-
tween remains nebulous. Thus some
black-box qualities persist around
advanced models’ inner transforma-
tions, despite transparency efforts.
So while more visibility into AI func-
tionality and development is crucial
for accountability and trust, uncer-
tainties around emergent system
behaviors may endure.
goal. That’s why researchers need to under-
stand how AI reaches the end goal. It might be
cheating to complete the task they were told
to do. Researchers at Stanford University and
Google discovered that an AI system designed
to turn satellite images into usable maps was
withholding certain data. The researchers were
using a neural network called CycleGAN, which
learns how to map image transformations. It
took an old aerial photograph of a neighbor-
hood, distinguished between streets, alleys,
driveways, buildings, and lampposts, and then
generated a map that could be used by GPS.
Initially, they used an aerial photograph that
hadn’t been seen by the network. The resulting
image looked very close to the original
—sus-
piciously close. But on deeper inspection, the
researchers found that many details in both
the original image and the generated image
weren’t visible in the map made by the AI. It
turns out that the system learned to hide infor-
mation about the original image inside of the
image it generated.
AI, like any technology,
is a reflection of its
creators and their
intentions.
—Joy Buolamwini,
founder of the Algorithmic Justice League

TECH
HOW DO WE ENSURE TRUST?
AI Alignment Goes Mainstream
As AI systems improve, many researchers
want guardrails to ensure that they are de-
ployed in ways that do not harm humanity.
AI alignment research refers to the process
of ensuring that AI systems act in accor-
dance with human values and goals. OpenAI,
DeepMind, and Anthropic (which describes
itself primarily as an “AI safety and research
company”) each have AI alignment teams
with dedicated staff researching guardrails.
While the total number of researchers work-
ing on AI alignment is small compared to
the rest of the AI community, such dedicated
teams did not exist until recently. The debate
surrounding the alignment of AI with human
objectives encompasses a broad spectrum
of opinions. On one end, “AI doomers” view
unchecked advancement, especially toward
superhuman capabilities, as posing existen-
tial catastrophe risk—potentially including
human extinction. They advocate solutions
like indefinite moratoriums on large model
training to forestall such outcomes. By con-
trast, the “effective accelerationist” perspec-
tive sees hastening progress as a moral im-
As AI is increasingly incorporated
into more sensitive domains, press-
ing questions emerge.
How can we build AI that we can trust?
How can we trust AI’s predictions and con-
clusions when much of the system
is opaque?
How can we ensure that AI is aligned with
human values, especially as it becomes
more and more capable?
Could we inadvertently instruct a powerful
AI towards harm?
Can we trust the current human custodians
of this technology?
These pressing issues are at the
heart of ongoing debates among AI
ethics experts, where a definitive
consensus on the best approaches
has yet to be reached.
perative to quickly harness AI solving pressing
global problems like disease, inequality, and
climate change.
In the moderate middle lie a diversity of per-
spectives. Some, like economist Tyler Cowen,
argue the doomers’ risks are too narrowly
specified for high probability, while others like
Leopold Aschenbrenner from OpenAI’s “supera-
lignment team” make the case for substantial
investments in AI alignment research, akin to
“Operation Warp Speed” but focused on AI. This
approach stems from the belief that artificial
general intelligence (AGI) could become the
most powerful tool ever developed, necessitat-
ing leadership in AI research by countries like
the US to maintain a strategic advantage over
nations such as China. These represent just a
few of the myriad perspectives and it is likely
that more perspectives will emerge before we
converge on the right AI-alignment strategy.
Indexing Trust
We will soon reach a point when we can no
longer tell if a data set has been tampered with,
either intentionally or accidentally. AI systems
The biggest lesson learned is
we have to take the unintend-
ed consequences of any new
technology along with all the
benefits, and think about them
simultaneously—as opposed
to waiting for the unintended
consequences to show up and
then address them. I don’t think
the world will put up anymore
with any of us coming up with
something where we haven’t
thought through safety, equity
and trust—these are big issues
for the world.
—Satya Nadella, CEO of Microsoft

TECH
HOW DO WE ENSURE TRUST?
rely on our trust. If we no longer trust their
outcomes, decades of research and technolog-
ical advancement will be for naught. Leaders
in every sector—government, business, non-
profits, and so on—must have confidence in
the data and algorithms used. Building trust
and accountability requires transparency and
is a challenge, but there are efforts underway
to assess AI transparency, a critical first step.
Researchers from Stanford, MIT, and Princeton
designed the Foundation Model Transparency
Index (FMTI)—a scoring system that evaluates
transparency across model development, func-
tionality, and usage. The 2023 index places
Llama 2 at the top, as the most transparent
Foundation model, followed by BigScience’s
BloomZ and OpenAI’s GPT-4. The hope is that
by standardizing analysis of opaque systems,
deployment risks and responsibilities can be
better informed.
The ethics of how data is collected in the first
place may also influence the trustworthiness
and validity of scientific research, particularly
in areas such as organ donations and medical
research. In addition, employing ethicists to
work directly with managers and developers
and ensuring diversity among developers—
representing different races, ethnicities,
and genders—will reduce inherent bias in AI
systems.
Synthesizing Trust
Humans can be tricked into believing ma-
chine-generated faces, especially when
they’ve been engineered to elicit trust. A study
in the Proceedings of the National Academy of
Sciences shows that synthetic faces are often
“deemed more trustworthy than real faces,”
suggesting that synthetic faces could be
designed as societal malware. If a bad actor
was attempting to undermine institutions, it
could deploy a synth on social media to sow
distrust. There are not yet effective counter-
measures for synthetic humans or effective
markers to help consumers distinguish be-
tween fake and real.
The perceived trustworthiness of synthetic faces over real ones raises concerns about their potential use
by malicious actors to erode trust in institutions.

TECH
ARE THERE TOOLS TO MAKE AI ETHICAL?
As AI systems become more ad-
vanced, making sure they are ethi-
cally deployed becomes increasingly
important. For instance, AI can now
generate hyper-realistic deepfake
media that now passes the uncanny
valley. This tech could let bad actors
impersonate people or spread mis-
information. AI can also be used for
cheating, fraud, and hacking. In light
of this, companies are emerging to
create tools to combat this behav-
ior. New tools can detect deepfakes,
expose fraudulent AI activities, and
implement preventative measures
against misuse.
while maintaining a 9% false positive rate
on human writing. These tools are essential
in contexts where distinguishing between
human and machine authorship is critical,
such as in academic integrity, journalism,
and legal documentation.
Tools for Detecting Copyright Violations in
AI Outputs
With AI models capable of memorizing and
reproducing content from their training data,
the risk of copyright infringement becomes a
significant concern. Researchers from Google,
DeepMind, ETH Zurich, Princeton, and Univer-
sity of California, Berkeley have demonstrated
this with the Stable Diffusion model, which
can emit memorized images, including those
with trademarked company logos. To combat
this, watermarking techniques are being de-
veloped. For instance, the University of Mary-
land proposes a technique for watermarking
language model outputs, making synthetic
text algorithmically identifiable. Google Deep-
Mind’s SynthID tool embeds digital water-
marks directly into image pixels, enabling the
identification of AI-generated images while
remaining invisible to the human eye.
Tools for Exposing Deepfakes
Hyper-realistic deepfakes pose significant
security risks. Researchers at Washington
University created a tool called AntiFake, which
can add a digital watermark to content that
proactively prevents the cloning of voices and
faces. Intel’s FakeCatcher and Sentinel use
deep learning to analyze media content, catch-
ing manipulations either in real time or high-
lighting alteration patterns after creation. One
significant concern regarding many deepfake
detection tools is that they demonstrate bias.
Studies have uncovered significant dispar-
ities in the error rates of deepfake detection
algorithms across different racial groups. In
one study, the difference in accuracy reached
as high as 10.7%. This bias could lead to severe
implications, such as genuine images of cer-
tain racial groups being mistakenly identified
as fakes or, conversely, manipulated images
being wrongly accepted as authentic. Dr. Siwei
Lyu and a team at the University of Buffalo
have developed what are considered to be the
first deepfake detection algorithms specifical-
ly designed to mitigate bias. Their approach
involves two machine learning methods: one
Deepfake Detectors
AI now enables creating highly realistic fake
media called deepfakes—bogus video, au-
dio, and text that seem real. They can spread
misinformation by fabricating scenes or
putting words in people’s mouths. Research-
ers are developing protections against their
misuse. Tools like AntiFake use imperceptible
watermarks to block fake voice/face cloning
before it happens. Platforms including Intel’s
FakeCatcher and European vendor Sentinel
catch manipulations in real-time using AI.
As deepfakes get better and better, surpass-
ing the “uncanny valley,” so does the market
opportunity for exposing them. Expect to see
more investment in this space.
Tools for Identifying AI-Generated Writing
AI is good at writing like a human. That means
we need tools that can distinguish between
human and AI-written content. DetectGPT is
one such tool, offering over 95% accuracy in
identifying whether a passage is written by a
human or an AI system like GPT-3. Similarly,
OpenAI itself has released a classifier that
flags AI-generated text 26% of the time,

TECH
ARE THERE TOOLS TO MAKE AI ETHICAL?
that makes algorithms aware of demographic
factors and another that works to blind them.
These methods have successfully reduced dis-
parities in accuracy across different races and
genders. Notably, this achievement was not at
the expense of overall accuracy; in some cases,
accuracy was even enhanced. By focusing
on the fairness of the algorithms, Lyu’s work
marks a significant step toward creating more
equitable and reliable deepfake detection
technologies, ensuring accuracy is indepen-
dent of factors like race or gender.
Tools to Thwart Recognition Systems
As facial recognition becomes ubiquitous, var-
ious groups want to limit the technology’s ef-
fectiveness to protect privacy. While methods
of confusing or obscuring facial recognition
systems are not always feasible, researchers
are trying to confuse online applications that
scrape and collect images used as inputs for
training facial recognition engines in order to
develop a form of camouflage, which consum-
ers may someday demand.
Meanwhile, Anthropic has unveiled its Re-
sponsible Scaling Policy, which includes a
detailed list of safety commitments based on
risk assessments and incorporates pauses in
development if safety measures fail to match
the pace of capability advancements. The
policy encompasses several key components,
including internal access controls, adver-
sarial testing (red-teaming), evaluations by
independent third parties, and graded access
based on different AI Safety Levels.
Researchers from the University of Chicago
have created a program, Fawkes, that adds
extra pixels to images to cause facial rec-
ognition apps to misclassify faces. Taking
this principle a step further, Israeli company
Adversa AI adds noise, or small alterations,
to photos of faces, causing algorithms to
detect a different face than what is visible to
the naked eye. The algorithm is successful at
imperceptibly changing an individual’s image
to someone else of their choosing.
Tools to Combat Broadly Malicious AI
Behavior
Research labs around the world are actively
working to build practical safeguards against
malicious AI behavior. DeepMind has intro-
duced a comprehensive toolkit and work-
flow designed to enhance the evaluation of
standard models that can identify when AI
is misbehaving according to human ethical
standards. This approach specifically focus-
es on identifying and assessing potentially
hazardous capabilities, like cyber-offense and
self-replication, as well as the likelihood of
causing harm.
Researchers are working on ways to alter facial im-
ages so facial recognition systems misidentify the
faces, potentially allowing people to avoid identifi-
cation by these surveillance technologies.

TECH
DOES AI INFRINGE ON PRIVACY?
HOW SHOULD WE THINK ABOUT CUSTOMER DATA AND AI APPLICATIONS?
AI enables new forms of pervasive
surveillance that could threaten per-
sonal privacy across several domains.
Last year, facial recognition company
Clearview AI said it had run more than
1 million searches for police in the US.
Somewhat less obviously, ambient
monitoring can now subject house-
hold environments to observation.
Workplace analytics can track de-
tailed employee behaviors and pro-
ductivity. Schools can actively mon-
itor students through devices and
platforms meant for remote educa-
tion. As private spaces face increasing
exposure from third-party tracking, a
culture of Big Brother–like awareness
becomes normalized, rather than
valuing independence and consent.
ogy by law enforcement, doesn’t apply to
private companies.
Teleperformance, a French-based company
that manages outsourced call center work
for many Fortune 50 companies, uses cam-
eras and AI to monitor its teams. It flags em-
ployees as idle when it detects they haven’t
used the keyboard or mouse within a speci-
fied time frame. Live Eye Surveillance offers
a monthly subscription service that remotely
monitors live video feeds of employees for
companies such as 7-Eleven, Dairy Queen,
and Holiday Inn. Sneek is another example
of “tattleware” that captures live photos of
employees via webcams and displays them
on a digital wall viewable by everyone in the
company. Click on a photo and it instantly
pulls that person into a video call with you.
The most well-known user of worker surveil-
lance might be Amazon, which has installed
AI-enabled cameras in delivery trucks to
track behavior. The company docks driver
pay if it perceives unsafe conditions such as
distracted driving, speeding, or hard braking.
In its warehouses, the company monitors
worker productivity by measuring what’s
called “time off task,” which is any time when
a worker isn’t actively processing products.
South Korean e-commerce giant Coupang,
which has pledged to become the “Amazon of
Korea,” uses similar surveillance tactics.
The industry has also continued to evolve as
it offers more AI-based analysis of workers.
Amazon is exploring using keystroke-logging
software that tracks user behavior over time
to detect if the same person is controlling the
worker’s account. Aware’s Spotlight software
detects behavioral changes like mood, tone,
and attitude across conversations on employ-
ees’ devices. Teramind offers software that
will disable private conversations if it detects
“inappropriate” keywords. With the top three
tools in the industry accounting for over 60%
of global demand, expect to see more AI-based
surveillance that leverages the growing pool of
data collected by a variety of companies.
School Surveillance
During the pandemic, many students were is-
sued laptops and other devices by schools to
Increased Used of Ambient Surveillance
What happens behind closed doors may not
be secret for long, and executives should
beware of new ambient surveillance meth-
ods. Scientists at MIT discovered how to use
computer vision to track data from what
they call “accidental cameras.” Windows,
mirrors, corners, houseplants, and other
common objects can be used, along with AI,
to track subtle changes in light, shadows,
and vibrations. The result: We all may soon
have X-ray vision capabilities—which may
not be great news for companies working on
sensitive projects. Those working in informa-
tion security and risk management should
pay special attention to advances in com-
puter vision.
Worker Surveillance
The rise of remote work during the pandemic
accelerated the surveillance of workers, and
will likely continue to grow as remote and
hybrid work models take root. The US Consti-
tution’s Fourth Amendment, which prohibits
unreasonable searches and seizures and
precludes most uses of this same technol-

TECH
DOES AI INFRINGE ON PRIVACY?
HOW SHOULD WE THINK ABOUT CUSTOMER DATA AND AI APPLICATIONS?
facilitate remote learning. They weren’t told,
however, that these devices would open a por-
tal into their homes that could be monitored
by schools at all times of the day. In the US
and many other countries, schools can legally
monitor students, often without disclosing
what is being tracked.
Gaggle is one company that monitors
school-issued accounts and uses AI to track
online behavior of students across services
like email and chat tools. In 2020, the Minne-
apolis school district signed a contract with
the company to monitor its students through
2023. A school district in California contract-
ed with Securly to monitor students in real
time, looking for prohibited behaviors such
as having too many browser tabs open. The
software enables teachers to close tabs for
any students they believe are “off task.”
Philadelphia and Chicago schools deployed
GoGuardian software on district-issued
Chromebooks. A vulnerability in the software
allowed teachers to start virtual sessions
that enabled webcams on those Chrome-
books without notification or consent by the
student. Schools in China deploy technol-
ogy to monitor attentiveness in students.
An algorithm called 4 Little Trees is used in
Hong Kong to detect students’ emotions as
they learn—by monitoring their facial expres-
sions with webcams. If the system detects
a lack of focus, it nudges the student to pay
attention.
AI introduces pervasive surveillance capabilities, jeopardizing personal privacy in schools, workplaces, and
public spaces. Society appears to be prioritizing convenience over privacy, accepting significant trade-offs.

TECH
IS THERE A FEASIBLE SOLUTION TO BIAS?
Seemingly the moment OpenAI’s
ChatGPT went public, there were
multiple accounts of the system
displaying racism, ageism, gender
bias, and political bias. But it’s not
just ChatGPT—many AI systems have
been revealed to contain bias—much
of which can be attributed to the
data that the systems were trained
on. Given AI’s expanding integration
into sensitive domains like finance
and health care, failing to address its
potential biases risks compounding
real-world discrimination through
algorithmic means.
Addressing Political Bias
In 2023, many conservatives raised con-
cerns about ChatGPT’s political bias, sharing
screenshots of ChatGPT’s left-leaning re-
sponses. OpenAI responded with a detailed
blog post explaining its moderation approach,
and CEO Sam Altman hinted at future pos-
sibilities for users to fine-tune ChatGPT
iterations within certain broad guidelines,
potentially sidestepping some contentious
value judgments. Elon Musk, responding
to these critiques about OpenAI’s political
correctness, launched a new venture, called
TruthGPT, aimed at exploring “deeper truths
about the universe.” Separately, in an effort to
make a point about biased AI, David Rozado,
a data scientist from New Zealand, created
DepolarizingGPT. This AI chatbot generates
three types of responses for each prompt:
left-wing, right-wing, and a neutral or inte-
grating perspective. To achieve this, Rozado
fine-tuned the chatbot using content under
fair-use provisions from various sources. For
the left-wing responses, he used material
from publications like The Atlantic and The
New Yorker, and authors like Bill McKibben
and Joseph Stiglitz. Conversely, the right-wing
responses were shaped using content from out-
lets such as National Review and The American
Conservative, and writers like Roger Scruton
and Thomas Sowell.
Doubts regarding ChatGPT’s ability to avoid
bias persist. A new research paper claims to
find substantial evidence of systematic po-
litical bias in ChatGPT, favoring Democrats in
the US, Lula in Brazil, and the Labour Party in
the UK. The paper analyzed ChatGPT’s respons-
es to statements from the Political Compass
test, concluding that it aligns more with liberal
parties internationally. However, the study’s
methodology and findings are not without
criticism. Some researchers argue that the way
ChatGPT was tested doesn’t reflect typical user
interactions and may not accurately represent
the AI’s behavior. Additionally, a data scientist,
Colin Fraser, discovered that reversing the order
of parties mentioned in prompts resulted in op-
posite biases, suggesting potential flaws in the
study. These findings illustrate the complexi-
ties in assessing AI bias and the need for great-
er transparency from developers like OpenAI.
The future of AI should be a
mirror to society. It must be
shaped by a diverse range
of voices, not just those of
technologists.
—Meredith Whittaker,
cofounder of the AI Now Institute

TECH
IS THERE A FEASIBLE SOLUTION TO BIAS?
Addressing Race and Gender Bias
There are significant challenges related to
race and gender bias in AI. A notable instance
occurred in December 2022, when Steven T.
Piantadosi of University of California, Berkeley
revealed a bias in ChatGPT’s programming,
which incorrectly associated scientific profi-
ciency with being white or Asian male. OpenAI
quickly addressed this issue, programming
ChatGPT to reject the notion that race or gen-
der should influence perceptions of scientific
ability. However, this incident underscores a
broader, long-standing problem of bias within
AI systems.
In another study, researchers at the University
of Florida examined racial bias in machine
learning algorithms used for diagnosing bac-
terial vaginosis (BV), a common infection in
women of reproductive age. The study, led by
faculty members Fang and Ivana Parker, ana-
lyzed data from 400 women across four eth-
nic groups—white, Black, Asian, and Hispanic.
They found that the accuracy of BV diagnosis
varied significantly among these groups, with
Hispanic women experiencing the highest
rate of false positives and Asian women the
most false negatives.
To quantify bias, another team from Baskin
Engineering at University of California, Santa
Cruz, led by Assistant Professor Xin (Eric)
Wang, developed a tool called the Text to
Image Association Test. It quantifies bias in
text-to-image (T2I) AI models like Stable Dif-
fusion, by measuring the variance in images
generated from neutral versus gender-spe-
cific prompts. The findings indicate that
such state-of-the-art models not only reflect
but can also amplify existing human biases.
Such analysis represents crucial progress
toward accountability, but much work re-
mains to ensure AI equitability.
After ChatGPT launched in 2022, researchers quickly discovered biases in its programming that linked sci-
entific proficiency to white or Asian male gender and race, underscoring the broader issue of unfair biases
that can become ingrained in AI systems. To address this, developers are proactively testing models for
biases and making concerted efforts to train more equitable, inclusive AI.

TECH
AI introduces security threats of unprecedented complexity due to its ability to
learn and adapt, making traditional security measures less effective. Its inte-
gration across critical infrastructure and sensitive systems means that AI-driv-
en attacks can have far-reaching and unpredictable consequences. Additionally,
the sophistication of AI enables the creation of highly targeted and convincing
cyberattacks, such as deepfakes and advanced phishing attempts, challenging
our ability to distinguish between genuine and malicious communications. For
example, recently, advanced language models have grown so smart that they
can now use tools, read documents, and even call on themselves, acting inde-
pendently. If AI models can hack websites on their own by finding and exploit-
ing weaknesses without being taught specific vulnerabilities, what does that
mean for the future of cybersecurity resilience?
Cyberthreats
The National Cyber Security Centre (NCSC)
released an assessment in 2023 on the near-
term impact of AI on cyberthreats. The NCSC
assessment delves into how AI will likely
enhance the volume and severity of cyberat-
tacks in the next two years, mainly through
the evolution of existing tactics. AI is being
used by various cyberthreat actors, includ-
ing state and non-state entities, to varying
degrees. The report suggests that AI will
significantly improve capabilities in areas
like reconnaissance and social engineering,
making them more efficient and harder to
detect. However, more sophisticated AI uses
in cyber operations will likely remain limit-
ed to actors with substantial resources and
expertise in AI and cyber technologies. The
assessment concludes that AI’s impact on
cyberthreats is uneven and depends on the
capability and intent of the threat actors.
It also points out that the proliferation of
AI-enabled cyber tools in criminal and com-
mercial markets is likely to further enhance
these capabilities.
Adversarial Attacks
Recent studies highlight a significant vulner-
ability in AI to adversarial attacks, revealing
that these systems can be more easily manip-
ulated to make incorrect decisions than pre-
viously understood. These adversarial attacks
involve deliberate tampering with the data
input into AI systems, causing them to misin-
terpret information or act in unintended ways.
For instance, specific patterns or objects, like
certain stickers on a stop sign, can trick an
AI in autonomous vehicles into not recogniz-
ing the sign. Similarly, alterations in medical
imaging data could lead an AI to diagnose
incorrectly.
This issue was the focus of a study by Tian-
fu Wu and his team at North Carolina State
University, which examined the prevalence
of such vulnerabilities in AI deep neural
networks. Their findings suggest that these
adversarial vulnerabilities are far more wide-
spread than previously recognized, posing
a significant challenge to the reliability and
safety of AI applications. In a separate study
by researchers at Carnegie Mellon University,
WHAT SECURITY ISSUES SHOULD WE PREPARE FOR?
AI lowers the barrier for novice cyber criminals, hackers-
for-hire and hacktivists to carry out effective access and
information gathering operations. This enhanced access
will likely contribute to the global ransomware threat over
the next two years.
—U.K. National Cyber Security Centre January 2024 Assessment

TECH
the vulnerability of AI chatbots to adversar-
ial attacks was demonstrated. By modifying
prompts with specific strings of text, which
may appear nonsensical but hold particular
significance for AI models trained on exten-
sive web data, the researchers could bypass
the safeguards designed to prevent chatbots
from generating inappropriate content. This
approach effectively “unshackled” the AI,
making it possible for chatbots like ChatGPT,
Google’s Bard, and Claude from Anthropic to
respond to otherwise restricted or harmful
queries. The success of these attacks across
multiple popular AI chatbots suggests a
deeper, more systemic weakness in the most
advanced AI systems, challenging the deploy-
ment and safe use of these technologies.
Data Poisoning: A Double-Edged Sword
Data poisoning attacks represent a signifi-
cant threat to AI systems, where malicious
actors deliberately manipulate the training
data to mislead the AI into making incorrect
or harmful decisions. These attackers exploit
vulnerabilities, such as embedding harmful
content within files, to introduce misleading
drawing into a cubist style, for example. This
integration allows artists to choose between
masking their style or actively using the
data-poisoning feature.
AI Lowers the Barrier to Misinformation
AI has lowered the bar to produce and dis-
tribute misinformation. An analysis by News-
Guard, a Microsoft tool that shows trust
ratings for over 7,500 news and informa-
tion websites, found that websites hosting
AI-generated bogus reporting have ballooned
over 1,000% in the past year, mushrooming
from 49 to over 600 outlets. While fabrica-
tion used to require armies of workers or
advanced intelligence agencies, AI democ-
ratizes deception. Now a lone teenager can
concoct sites and stories that appear au-
thentic. And generative AI allows customiz-
ing fakery to particular targets and contexts
with minimal effort. A study by the University
of Waterloo found that an early version of
ChatGPT, when tested on different types of
statements including facts and misconcep-
tions, often made errors, contradicted itself,
and repeated false information. For example,
it could correctly state that the Earth is not
flat when asked directly, but show inconsis-
tency in its responses. Researchers expressed
concern over these findings, highlighting the
danger of AI models like GPT-3 spreading mis-
information, especially as they become more
common in use.
This is particularly concerning as we approach
the 2024 US presidential election, with mis-
information experts raising flags about the
potential impacts on democratic process-
es. Ominous previews have already played
out abroad. Shortly before a crucial national
election in Slovakia, a controversial audio clip
spread on social media, purporting to feature
Michal Šimečka of the Progressive party dis-
cussing a vote-rigging plan. Another incident
involved a fake recording of the UK Labour
Party leader verbally attacking a staffer. Both
recordings, which seemed authentic, were
later exposed by fact-checkers as AI-gener-
ated fakes, highlighting the growing issue of
AI-manipulated audio in spreading misinfor-
mation.
data into the training set. This can skew
the AI’s learning process, aligning it with
the attacker’s goals, potentially leading to
biased outcomes, data breaches, or simply
inaccurate AI outputs. To illustrate, consider
the cost implications: Training a complex AI
model like GPT-3 can cost around $17 mil-
lion. If its training data were compromised,
restarting the process could lead to substan-
tial financial losses.
On the flip side, data poisoning can also
serve as a defensive mechanism. A novel tool
named Nightshade exemplifies this dual
nature. Designed to protect artists’ intel-
lectual property, Nightshade subtly alters
digital artwork’s pixels. When AI models use
this “poisoned” art for training, their ability
to accurately interpret images is compro-
mised, leading to erroneous outputs, such
as mistaking a car for a cow. This tool is part
of a broader strategy for artists to safeguard
their work in an unregulated landscape. Art-
ists can use Nightshade via Glaze, another
tool from the same creators, which masks
an artist’s style—transforming a realistic

TECH
Privacy Risks in Behavioral Biometrics
Behavioral biometrics, which employs ma-
chine learning to analyze a vast array of
biometric data points, raises significant pri-
vacy concerns. By quantifying subtle aspects
of our behavior, such as the force used on
touchscreens, the distinct way we tap letters
like “Cs” and “Vs,” or our unique patterns
when using a physical keyboard, these tools
can reveal intricate details about our iden-
tities, thoughts, and future actions. While
the technology offers potential benefits like
enhancing security and possibly eliminating
the need for passwords by identifying indi-
viduals through their typing patterns, it also
introduces substantial risks.
The very aspect that makes behavioral bio-
metrics appealing—its ability to authenti-
cate a user based on nuanced behavioral
traits—also makes it a privacy concern. If our
behavioral patterns can be so precisely mon-
itored and analyzed, they can be replicated
or exploited, leading to new forms of security
vulnerabilities. The notion that machines can
detect and record behaviors we’re not even
conscious of ourselves not only challenges
our concept of privacy but also highlights
how these patterns, once considered person-
al and private, can become accessible and
potentially misused. This duality presents a
critical challenge: balancing the innovative
applications of behavioral biometrics against
the imperative to protect individual privacy
and ensure the security of personal data.
AI can track unconscious patterns in human behaviors like typing cadence and keyboard pressure to de-
rive insights about inner emotional states without explicit user permission.

TECH
AI operations, particularly those involving deep learning and complex model
training, are significantly more computationally intensive than traditional com-
puting tasks. This intensity stems from the need to process vast amounts of
data and perform countless calculations rapidly to train models, recognize pat-
terns, and make decisions. Consequently, AI demands considerably more energy
to sustain these operations, as the intricate algorithms and large-scale data
processing require substantial computational resources, leading to higher ener-
gy consumption compared to conventional computing workloads. On the other
hand, AI is also helping solve environmental issues. A Canadian startup, Rail-
Vision Analytics, developed AI software that helps train engineers drive more
efficiently, potentially saving significant amounts of diesel fuel and reducing
the rail industry’s carbon emissions. This technology, which is like Google Maps
but for trains, advises engineers on when to speed up or stay idle, optimizing
fuel use and contributing to a potential annual reduction of over 20,000 tons
of carbon emissions if widely adopted, equivalent to removing more than 4,000
cars off the road each year.
New Architectures to Make AI Workloads
More Efficient
As AI models become more complex and
larger, consuming a greater share of our
computing resources, their energy usage also
escalates. One promising approach to make
AI-intense compute more energy efficient is
by using photonic AI chips, which harness
light rather than electricity for orders-of-mag-
nitude better efficiency at matrix multiplica-
tions—a core operation for deep learning. A
Stanford team recently achieved a milestone
by training an optical neural network chip to
label data points with 98% accuracy. For the
first time, their photonic processor enabled
light to flow bidirectionally to implement the
backpropagation algorithms vital for training.
While refinements remain, this demonstrates
the promise of optical computing to slash
the carbon footprint of AI workloads.
Neuromorphic chips offer another model of
efficient AI hardware, taking inspiration from
the human brain’s simultaneously distrib-
uted storing and processing of information.
Rather than shuttling data back and forth
like conventional computers, neuromorphic
processors like Intel’s Loihi store memory
within computation. Specializing in sensory
processing, these chips already achieve 1,000x
higher efficiency than traditional hardware for
tasks like gesture and sound recognition.
In a groundbreaking approach, researchers
envision biocomputers powered by networked
human brain organoids—essentially mini-
brains grown from stem cells. “Organoid
intelligence” holds significant potential for
augmenting computing capabilities while
concurrently addressing the escalating energy
consumption demands driven by advance-
ments in artificial intelligence and supercom-
puting (see the Computing report for more
information on organoid intelligence). Despite
traditional computers’ ability to process
calculations at speeds far surpassing human
capabilities, human brains demonstrate supe-
rior performance in complex decision-making
tasks, such as differentiating between a dog
and a cat. Running AI on organoids could be
the key to achieving human-like complex deci-
sion-making in an energy-efficient manner.
WHAT DOES AI HAVE TO DO WITH ESG?

TECH
WHAT DOES AI HAVE TO DO WITH ESG?
A Nuclear Renaissance for AI Workloads
The monumental computational require-
ments of advancing AI could catalyze a
nuclear power renaissance. Microsoft is
exploring the use of next-generation small
modular nuclear reactors (SMRs) to power its
data centers and AI operations. SMRs promise
cheaper, faster modular construction com-
pared to traditional nuclear plants, which are
often over-budget and delayed. Microsoft’s
approach was hinted at further in the fact
that they already have a deal to buy Clean
Energy Credits from Ontario Power Genera-
tion, which is on track to be the first utility to
deploy an SMR in North America. Companies
like Rolls-Royce, Last Energy, NuScale, Oklo,
and TerraPower (backed by Bill Gates) are also
developing various SMR models. Similarly,
Kärnfull Next in Sweden plans to use SMRs to
power data centers. The pivot toward nuclear
energy, particularly next-generation SMRs, is
a strategic response to the dual challenges of
meeting the high energy demands of AI and
achieving climate goals.
viruses and their animal hosts. In a practical
demonstration of AI’s environmental appli-
cations, a research team from the University
of Waterloo has developed an AI tool, Plas-
ticNet, to identify microplastics with unprec-
edented speed and accuracy. This technol-
ogy is particularly crucial for mitigating the
environmental and health hazards posed by
microplastics, commonly found in food and
water sources. By enhancing the efficiency
of identifying these pollutants, PlasticNet
supports wastewater treatment and food
production industries in making informed
decisions to protect the environment and
public health.
Environmental AI
AI presents a dual-edged sword in its impact
on the environment, with its capabilities ex-
tending to both contributing to and alleviat-
ing climate change. David Rolnick from Mc-
Gill University and Mila—Quebec AI Institute,
notes that while AI’s energy consumption
and the promotion of consumerism through
AI-based advertising may exacerbate climate
challenges, it also offers solutions for envi-
ronmental conservation. For instance, AI is
being utilized to monitor and curb deforesta-
tion effectively.
A recent paper from the Cary Institute of
Ecosystem Studies highlights how princi-
ples from ecology could inspire a new wave
of AI development. This synergy between AI
and ecology is seen as a pathway to address
pressing global issues like disease out-
breaks, biodiversity loss, and the repercus-
sions of climate change. AI’s application in
ecology is already proving beneficial, aiding
ecologists in detecting patterns within
vast data sets to make precise predictions,
like identifying potential human-infecting
While AI’s computational demands pose sustain-
ability challenges, it can also enable climate miti-
gation - models can guide efficient resource usage,
accelerate green tech RD, and predict environ-
mental impact.

POLICY AND
REGULATIONS

TECH
HOW DOES GEOPOLITICS FACTOR INTO THE DEVELOPMENT OF AI,
AND IS THERE REALLY A NEW COLD WAR?
Countries are increasingly nationalis-
tic about advancing domestic AI capa-
bilities, with major investments and
restrictions aimed at getting an edge,
even as collaboration fractures. This
extends into military contexts, where
AI drives rapid innovations in areas
like weapons systems, wargaming,
and cyber operations—innovations
dual in nature for both defense and
potential offense. The combination of
deteriorating cooperation and uncon-
trolled AI militarization risks fueling
a dangerous tech-centric arms race.
Unless cooperative norms are estab-
lished, AI may drive global strategic
realignments as impactful as 20th
century nuclear and space races.
AI Nationalism
Governments are racing to establish national
AI champions and reduce reliance on foreign
technology. After high-profile chatbot debuts
like ChatGPT spawned in the US, 2023 wit-
nessed nations worldwide scramble to nur-
ture domestic AI capabilities, allocating tens
of billions in funding. France unveiled sub-
stantial funding for startup Mistral. India’s
Krutrim launched the country’s first multi-
lingual model. Abu Dhabi commercialized its
Falcon system. Beyond economic impacts,
concerns mix technological prestige with na-
tional security and ideological control. The US
and China are at the forefront of this tug-of-
war, each pledging billions toward AI invest-
ments. While US companies are pioneering
the most advanced LLMs, the US government
is concentrating resources on growing home-
made chip capabilities, aiming to lessen reli-
ance on imports critical for national security.
Concurrently, the US has imposed stringent
export controls to limit the dissemination of
advanced AI technology to rivals like China
and Russia. With Western companies barred
from exporting cutting-edge AI chips, adver-
sarial states invest heavily to replace blocked
supplies. China has earmarked hundreds of
billions to develop domestic chip fabrication
immune to US sanctions. The Chinese gov-
ernment has invested heavily in replicating
the chip supply chain domestically, aiming to
insulate itself from Western sanctions. See “The
AI-Driven Chip War” for more.
China also approaches AI on ideological
grounds. The country mandated that AI align
with the “core values of socialism,” effectively
limiting the influence of Western-developed
AI systems within its borders. This stance has
propelled Chinese tech giants like Alibaba
and Baidu to develop their own generative
AI tools, despite challenges in matching the
impact of their Western counterparts. Russia
also perceives American AI advancements as
a cultural and ethical threat, with President
Vladimir Putin highlighting the dangers posed
by Western LLMs to Russian “traditional val-
ues.” This reflects a broader concern over AI’s
potential to shape cultural and ethical norms,
prompting Russia to explore the development
of homegrown AI solutions.
After Nagasaki and Hiroshima,
it took 18 years to get to a
treaty over test bans and
things like that. We don’t have
that kind of time today.”
—former Google CEO Eric Schmidt,
on the urgency to create guardrails for AI.

TECH
HOW DOES GEOPOLITICS FACTOR INTO THE DEVELOPMENT OF AI,
AND IS THERE REALLY A NEW COLD WAR?
The AI-Driven Chip War
Rising tensions between the US and China
are catalyzing a supply chain schism for the
AI chips critical to national competitiveness.
This divide has been exacerbated by strategic
moves such as the CHIPS Act and increasingly
stringent export controls, which have partic-
ularly targeted the semiconductor sector—a
vital component of AI development. Efforts by
companies like Nvidia to adapt by launching
China-specific chips were thwarted by new
US restrictions, leading Chinese companies
to turn to domestic suppliers such as Huawei.
Additionally, Dutch firm ASML canceled ship-
ments of advanced semiconductor manufac-
turing equipment to China under US pressure,
highlighting efforts to curb China’s access
to crucial AI development technologies. The
standoff has prompted China to explore
alternatives like RISC-V, an open-source chip
architecture, as a means to bypass interna-
tional restrictions. This move has sparked
debate in the US about the potential risks
of technology transfer and the feasibility of
restricting contributions to RISC-V due to its
global, royalty-free nature.
The intensification of the Chip War is leading
to a bifurcation in the AI chip market, with
potential long-term implications for global
technological advancement and cooperation.
This divergence not only underscores the
strategic importance of semiconductors in
national security and AI development but
also hints at the emergence of distinct tech-
nological spheres, each aligned with diver-
gent national values and priorities.
The chip war could force a seismic restructuring of international manufacturing supply chains, trade flows,
and technology innovation networks.

TECH
Advancements in artificial intelli-
gence are reshaping modern warfare
in unprecedented and concerning
ways. Militaries worldwide are explor-
ing how to best leverage AI for tactical
advantages, including through auton-
omous weapons systems, wargaming
simulations, and automated hacking
tools. However, these technologies
raise pressing ethical issues and
could dangerously escalate conflicts.
The complex tradeoffs surrounding
AI and defense boil down to a central
tension: harnessing potential benefits
to national security versus controlling
for geopolitical risks.
members while outlining a roadmap for
adoption. Additionally, the First Committee
of the UN General Assembly adopted a draft
resolution in 2023 calling for the UN secre-
tary-general to conduct a comprehensive
study of lethal autonomous weapons. The
committee instructed the secretary-general
to consult member states and civil society
on addressing humanitarian, legal, security,
technological, and ethical concerns related
to autonomous weapons.
Simulating Warfare
Given the rising tensions between the US
and China over Taiwan, several groups are
building AI-powered simulation tools to war-
game a future conflict. In China, the People’s
Liberation Army has been using AI simula-
tion tools to prepare for military operations
against Taiwan.
The Center for Strategic and International,
a bipartisan, nonprofit policy research or-
ganization, developed a wargame involving
an amphibious invasion of Taiwan. After 24
rounds of gameplay, the US and its allies
Japan and Taiwan successfully defeated a
conventional amphibious invasion by China.
While Taiwan remained autonomous in the
simulation, its economy was devastated and
the US lost hundreds of aircraft and tens of
thousands of lives––while the Chinese Com-
munist Party never really destabilized. Games
that use real-world data to run simulations are
augmenting the work of military strategists,
so that leaders can validate or revise their pos-
tures on deterrence, invasion, and defense.
AI Used to Guide Military Strikes
In 2021, the US military said that it had started
using AI to guide its airstrikes, deploying algo-
rithms to a live operational kill chain. The kill
chain is a process of gathering intelligence,
performing analysis, weighing risks, and
deploying weapons to destroy a target. Using a
modified process, an AI system was deployed
into the Air Force Distributed Common Ground
System to analyze troves of intelligence, which
would have required a significant amount of
human hours to complete. The new AI system
cannot order a strike on its own, but it is now
automatically identifying possible targets.
Autonomous Weapons Policies
The US Department of Defense recently up-
dated its guidance on autonomy in weapons
systems. The original 2012 policy, and a 2017
update, did not explicitly mention AI. The
DOD updated its AWS definition by removing
references to a “human operator” and re-
placing it with simply “operator,” a subtle yet
notable shift clearing the way for future sys-
tems with decreased human oversight. This
new directive is aimed at helping to clarify
the process for developing autonomous
or semi-autonomous weapons systems.
Previous policies, such as the Ethical Prin-
ciples for Artificial Intelligence (2020) and
Responsible Artificial Intelligence Strategy
and Implementation Pathway (2021), were
intended to guide decision making for the
development and deployment of AI within
the military.
The policy change comes on the heels of
other recent government actions addressing
military AI. In late 2022, NATO released its
Autonomy Implementation Plan, arguing AI
systems offer clear opportunities for alliance
COULD AI BE INVOLVED IN—OR CAUSE—A HOT WAR?

TECH
Automated Target Recognition
Lethal autonomous weapons systems,
powered by AI, are capable of finding tar-
gets autonomously and making decisions
to complete a mission. In 2022, a lieutenant
colonel in the Ukrainian military said that he
and a group called Aerorozvidka had devel-
oped special drones that make use of auto-
mated target recognition. While it’s unclear
whether Aerorozvidka actually carried out
test missions, the fact remains that machine
learning–based vision for automated target
recognition already exists. In response, 70
nations delivered a joint statement at the UN
General Assembly calling for a ban on autono-
mous weapons––but little progress has been
made in the months since.
Automating Offensive Attacks Using AI
Thanks to advancements in AI, one of the big
trends in security is automated hacking— in
short, software that’s built to out-hack the
human hackers. DARPA launched a Cyber
Grand Challenge project in 2016, with a mis-
sion to design computer systems capable of
beating hackers at their own game. DARPA
tegic insights. By fusing photographs, drone
footage, and overhead views, AI integrates
distinct perspectives into a unified assess-
ment of terrain and enemy movements.
This augmented analytics empowers a new
paradigm of cost-effective warfare centered
around drones. Affordable models either
commercially sourced or improvised as
DIY drones generate valuable intelligence
rivaling America’s far more expensive Reaper
and Predator UAVs. Tight integration with
cutting-edge systems like Delta further
multiplies impact. After proving effective in
2022 trials, Ukraine greenlit Delta’s full-scale
February deployment. Pulling sensor, aerial,
and ground reports into a consolidated data
lake, this cloud-based architecture furnishes
commanders with an integrated common
operating picture for tactical decisions.
Algorithmic Warfighting
Future wars could be fought entirely in code,
using data and algorithms as powerful
weapons. The current global order is being
shaped by artificial intelligence, and the
wanted to show that smarter automated sys-
tems can reduce the response time—and fix
system flaws—to just a few seconds. Spot-
ting and fixing critical vulnerabilities is a
task that might take a human hacker sever-
al months or even years to complete, and yet
the machine that won the Grand Challenge
did it in just a fraction of that time.
The winner became the first nonhuman enti-
ty to earn the DEF CON’s Black Badge, which
is the hacking community’s equivalent of an
Oscar. Very soon, malicious actors will create
autonomous systems capable of automat-
ically learning new environments, exposing
vulnerabilities and flaws, and then exploit-
ing them for gain—or whatever the stated
objective, which could simply be generalized
mayhem.
AI-Assisted Situational Awareness
Ukraine has become a test bed for modern
AI-enabled battlefield awareness. Geospa-
tial intelligence leverages neural networks
to combine satellite imagery, social media
posts, and other open-source data into stra-
AI is enabling the development of weapons that
can select targets and attack on their own. The UN
General Assembly has called for banning this type
of autonomous attack technology, but so far there
has been no ban put in place.

TECH
same countries leading the world in AI re-
search—the US, China, Israel, France, Russia,
the UK, and South Korea—are also developing
weapons systems that include at least some
autonomous functionality.
In 2020, the US Air Force successfully flew
an AI copilot on a U-2 spy plane in California,
marking the first time in the history of the
DOD that an AI algorithm trained to execute
specific in-flight tasks was deployed. With
the call sign ARTUµ, it was the mission com-
mander—though the flight was just practice.
Future Today Institute analysis shows that the
future of warfare encompasses more than tra-
ditional weapons. Using AI techniques, a mil-
itary can “win” by destabilizing an economy
rather than demolishing countrysides and city
centers. From that perspective, China’s unified
march to advance AI puts the emerging super-
power dangerously far ahead of the West.
Mandating Ethics Guidelines for Tech
Contractors
Project Maven was developed to enlist AI to
analyze surveillance video. Initially, Google
was the DOD’s vendor, but when employees
found out they’d been working on a military
project, thousands protested. It wasn’t the
first time tech contractors had lost trust in
the government.
As a result, the Defense Innovation Unit is
enforcing “responsible artificial intelligence”
guidelines that vendors must adopt when
building AI systems, models, or applications
for the DOD. The guidelines offer specific
instructions that must be followed during
planning, development, and deployment,
which include provisions for risk assess-
ment. This represents a longer-term trend:
government agencies requiring transparency
in AI projects.
The future of warfare may largely involve fighting via cyberattacks powered by AI systems rather than con-
ventional physical weapons. Militaries have started using AI as co-pilots in spy planes and drones, pointing
to increased AI integration in defense.

REGIONAL
APPROACHES

TECH
COUNTRIES TRY TO REGULATE AI, BUT PLANS DIVERGE
Governments worldwide are trying
to balance maximizing AI’s benefits
with mitigating its risks by establish-
ing regulatory frameworks. So far, 31
countries have passed AI regulations
and 13 more are debating AI laws.
There are significant divergences
between each country’s distinct ap-
proach to regulating the technology.
Some nations, like Israel, Japan, and
Australia, have focused on revising
existing laws to facilitate AI develop-
ment, while others, like the UAE, are
crafting broad national AI strategies
with minimal regulatory emphasis.
Countries, like Russia, Iran, North
Korea, Syria, and Iraq, have opted to
outright ban specific services like
ChatGPT. The EU’s AI Act categorizes
systems by risk levels and restricts
the highest risk applications. Like the
EU, China has introduced AI-specific
obtain licenses before exporting cer-
tain technologies to these entities,
aiming to address national securi-
ty concerns without automatically
imposing a full embargo, reflecting
a significant effort to regulate the
flow of sensitive technologies to
organizations implicated in uneth-
ical practices. The impact on these
Chinese companies and China’s AI
industry could be disastrous, de-
pending on enforcement measures
and these companies’ reliance on
US technology, worsening US-China
relations.
These complex tensions parallel
past situations like GDPR in Europe,
where large multinational compa-
nies often end up defaulting to the
most stringent regulations globally
even if not universally binding. This
scenario could plausibly unfold with
major players standardizing elements
of higher-bar AI governance models
like the EU’s for consistency. The lack
of alignment across the proliferating
patchwork of national and regional AI
laws risks hampering innovation and
global collaboration. But ironing out
conflicts poses immense challenges
given different priorities surrounding
development versus human rights
and ethics.
legislative frameworks, but unlike
the EU, the frameworks are centered
on enforcing “socialist core values”
in AI. Brazil’s draft AI policies pri-
oritize user rights and risk assess-
ments, differing from Israel’s model
underscoring responsible innovation
and sector-specific oversight. The
UAE’s national strategy concentrates
heavily on expanding AI integration
rather than regulation.
On October 9, 2023, the US Bureau
of Industry and Security of the De-
partment of Commerce added 28
Chinese entities, including eight
leading technology companies, to
its entity list for their involvement
in human rights violations against
Uighur Muslims in Xinjiang, a move
that China condemned as inter-
ference in its internal affairs. This
listing requires US companies to

TECH
HOW IS THE US SPECIFICALLY REGULATING AI?
The explosion of AI technologies is
leading to both fascination and con-
cern among federal legislators, who
are now exploring regulatory respons-
es without a clear consensus. Last
October, the Biden administration is-
sued an executive order to ensure the
safe and trustworthy development
and use of AI, covering a wide range
of AI systems beyond just generative
AI and neural networks, affecting
organizations across all economic
sectors. Going forward, the National
Institute of Standards and Technolo-
gy (NIST) will play a key role in es-
tablishing guidelines for AI systems,
prompting organizations to assess
their use of AI and their reliance on
AI-enabled products and services
from third parties, and to align their
AI risk management frameworks with
NIST standards. But for now, there is
society. State laws vary, focusing on
consumer data privacy, combating
AI-driven discrimination, especially
in hiring practices, and address-
ing the manipulation of media in
elections, with some states already
implementing or proposing legisla-
tion to restrict deceptive AI-generat-
ed content. The US will likely adopt
a bottom-up patchwork quilt of AI
regulations instead of one sweeping
law, like the EU’s AI Act. The US gov-
ernment will likely boost spending
on AI and AI research, especially in
defense and intelligence, and use its
buying power to shape the market.
A Patchwork Approach
In the US, the approach to regulating AI
amounts to a diverse array of regulations that
vary by state and sector, creating a patchwork
framework rather than a unified national
strategy. This decentralized approach results
in differing standards and guidelines across
jurisdictions, complicating compliance for
organizations operating in multiple states and
sectors within the AI landscape. Consider the
current landscape of proposals and policies
below.
Bias
The Algorithmic Accountability Act, a nota-
ble congressional proposal, would mandate
companies to evaluate their algorithmic
systems, including AI, for bias, effectiveness,
and other factors if passed. Under the act, the
Federal Trade Commission would be tasked
with enforcing these evaluations, with a focus
on preventing the use or sale of racially biased
algorithms. However, the specifics of the FTC’s
enforcement strategy remain undefined.
no clear enforcement mechanism
in place to check for compliance.
Meanwhile, Congress is deliberating
how to approach AI’s dual-edged
sword of opportunities and challeng-
es, as it looks to local legislatures
for precedents. So far, more than 30
states have enacted laws address-
ing AI in diverse ways, from specific
policy concerns to establishing bod-
ies for studying AI’s impact. Senate
Majority Leader Chuck Schumer and
others have emphasized the need
for AI regulation through initiatives
like the AI Insights Forum, signaling
a bipartisan understanding of its
necessity. Some lawmakers con-
sider the European Union’s AI Act
a model for comprehensive regula-
tion, suggesting that the US might
follow with a similar framework to
manage AI’s growing influence in

TECH
Several US states and the District of Colum-
bia are enacting or proposing legislation to
prevent AI and algorithmic decision-making
tools from reinforcing societal discrimi-
nation. Laws are being passed primarily in
Democratic-led states, focusing on areas like
insurance, employee surveillance, and hiring
practices. For instance, Colorado mandates
insurers to disclose and manage risks of
algorithm use to ensure fair coverage. Mas-
sachusetts is considering a ban on AI-based
employee surveillance technologies. In D.C.,
proposed laws would restrict service eligibili-
ty decisions made by algorithms and require
user notification about how their data is used.
Additionally, New York City and some states
are addressing AI’s role in hiring, requiring
bias audits and transparency in the use of au-
tomated decision systems.
Copyright
The US Copyright Office has ruled that AI-gen-
erated content typically doesn’t qualify for
copyright protection as it’s not human-creat-
ed. But this could change now that tools like
Sora pose new threats to the film industry.
Deepfakes
The US National Defense Authorization Act
includes provisions that address the grow-
ing problem of deepfakes, requiring the
Department of Homeland Security to issue
an annual report for the next five years on
the risks posed by deepfakes. In 2021, the US
Senate Committee on Homeland Security
and Governmental Affairs voted unanimous-
ly to advance the Deepfake Task Force Act,
which would establish a public-private team
to investigate technology strategies and to
develop policies that could curb risk.
Numerous states, including California, Texas,
Minnesota, and Washington, have enacted
laws, while New York, New Jersey, and Mich-
igan have proposed legislation, aimed at
either prohibiting or requiring disclosure of
manipulated media. Many of these measures
are intended to prevent public deception
regarding political candidates or to influence
election outcomes.
Misuse
The topic of auditing misuse is also on the
Privacy
Proposals at the federal level include a
complete prohibition on using personal data
for targeted advertising and FTC-mandated
data minimization, restricting websites to
collect only data pertinent to their specific
functions. At the state level, at least 12 states
have enacted regulations governing auto-
mated decision systems, including AI, for
profiling consumers based on personal data.
Virginia’s 2021 Consumer Data Protection
Act is a pioneering example, mandating risk
assessments and consumer rights protec-
tions when entities process over 25,000
people’s data for profiling posing height-
ened harm risks. States are increasingly
following Virginia’s model by instituting
similar regulatory frameworks surrounding
data-driven automated systems. Addition-
ally, some jurisdictions like New York City
are specifically restricting AI usage in hiring
practices through measures like bias audits
and candidate notifications when screening
algorithms are deployed.
The US government is concerned about the po-
tential misuse of deepfake technology, especially
with elections approaching, which is why the latest
National Defense Authorization Act includes new
provisions aimed at tackling the challenges posed
by increasingly realistic synthetic media.

TECH
congressional agenda. Legislators like Sens.
Ted Budd and Ed Markey are pushing for the
Department of Health and Human Services
to assess AI’s biological risks and develop
strategies against its use in bioweapons or
artificial pandemics.
Licensing
The concept of licensing requirements for AI,
akin to the stringent regulation of food and
pharmaceuticals, is gaining momentum.
Inspired by Andrew Tutt’s 2017 proposal, this
approach suggests an agency could enforce
pre-market approval for algorithms in certain
applications, effectively requiring a govern-
ment license before public deployment. Prom-
inent figures like OpenAI’s Sam Altman, Rand
Corp.’s Jason Matheny, and New York Universi-
ty’s Gary Marcus have supported such licens-
ing, drawing parallels with the Food and Drug
Administration’s model.
Compute
In October 2023, the Bureau of Industry and
Security introduced updated rules to en-
hance its October 7, 2022, regulations. These
Agency (IAEA) to govern superintelligence
efforts and safe AI deployment worldwide. A
2020 law has already mandated the creation
of a task force to design NAIRR.
However, others argue existing institutions
could handle AI oversight without requiring
new bureaucracies. Examples include gov-
ernment bodies like the National Institute of
Standards and Technology, FDA, Securities
Exchange Commission, and Federal Commu-
nications Commission, as well as beefing up
AI and tech expertise within established sci-
entific institutions such as the Department
of Energy, National Science Foundation, and
NIST. Critics argue new institutions may not
be more effective than today’s agencies,
citing issues faced by analogous bodies like
IAEA in comprehensively monitoring rele-
vant technologies globally. There are also
questions around feasibility of meaningfully
tracking AI development as opposed to phys-
ical materials.
At least 12 states have passed laws man-
dating government or related entities to
research AI to enhance understanding and
assess potential impacts. While some of these
initiatives delay targeted regulation, others
have led to tangible steps. For instance, Ver-
mont’s Artificial Intelligence Task Force’s anal-
ysis resulted in the state’s Division of Artificial
Intelligence, which annually reviews AI usage
and its effects within state government.
Public-Private Partnerships
The abundance of AI job listings across practi-
cally every American industry signals surging
demand for related skills. The White House
has issued a call for AI talent to join the fed-
eral government, following President Biden’s
executive order for the safe, secure, and ethi-
cal development and use of AI. This initiative
seeks experts to help implement AI technol-
ogies across various government sectors to
enhance services, ensure AI safety and equity,
and maintain the country’s leadership in AI
innovation. Many US fabs, which are funded in
part by the Chips Act, face construction delays
because of a shortage of skilled workers.
The US has enlisted allies like Japan, the
revisions focus on closing loopholes in the
existing policy, further limiting China’s
access to advanced AI semiconductors and
manufacturing tools. This move strengthens
the US strategy to impede China’s military AI
development.
Conflicting Views About Institutional Roles
Debate persists over whether new institu-
tions should be formed to oversee AI devel-
opment and safety, or if this responsibility
should fall to existing agencies. Proponents
of new institutions argue they could have a
major positive impact, just as bodies like
the National Transportation Safety Board did
for transportation safety. Suggestions for
new AI oversight bodies include a Nation-
al Algorithms Safety Board to monitor and
ensure safety in algorithmic and AI systems,
a federally funded National Artificial Intelli-
gence Research Resource (NAIRR) to support
and coordinate AI research, an international
collaborative facility modeled after CERN to
attract top global AI talent and focus efforts
on pursuing AI safely, and an organization
similar to the International Atomic Energy

TECH
Netherlands, and Germany to tighten their
own export regimes. While competing nations
pursue more centralized strategies, the US
distinctive edge lies in decentralized AI inno-
vation across promising startups and tech
giants alike.
Large companies have always lobbied to
influence policy and regulation. But as the
tech giants amass power and wealth, they are
making key decisions that impact diplomacy
and geopolitics.
Big Tech companies are standing up depart-
ments dedicated to geopolitics. Microsoft
President Brad Smith regularly meets with
heads of state, and in 2023 played a key role
at the World Economic Forum’s Annual Meet-
ing. Smith developed an international treaty
called the Digital Geneva Convention to pro-
tect citizens against state-sponsored cyber-
attacks. Microsoft’s Digital Diplomacy Group
actively works on a tech-focused approach to
foreign policy. The company sees corporate
foreign policy as good business that builds
trust and enables long-term planning. Meta,
repair vulnerabilities in the US government’s
infrastructure, responding to the increasing
deployment of the technology by hackers for
nefarious activities.
Google, Amazon, Salesforce, and many other
companies are now building teams centered
on geopolitics and digital diplomacy. More
than a dozen countries are creating ambas-
sador-like positions charged to negotiate
with the leaders of US Big Tech companies,
with the aim of mediating disagreements,
collaborating on shared interests and
developing public-private alliances. The
longer-term implications of corporations in-
fluencing global politics could be profound.
What if a company’s priorities differ from the
national priorities of its home government?
National Security
While late to consider AI as a national securi-
ty issue, the US is quickly playing catchup.
The Pentagon is considering the creation
of an extensive network that leverages AI
along with drones and autonomous systems
within the upcoming two years, aimed at
mitigating threats posed by Russia, Chi-
na and non-state actors. The White House
has already initiated a multimillion-dollar
cybersecurity competition aimed at en-
couraging the adoption of AI to identify and
The Pentagon is considering plans to build an
expansive network utilizing AI and drones over the
next two years to address threats from Russia, Chi-
na, and non-state groups.

TECH
WHAT IS CHINA DOING?
China is an undisputed global lead-
er in AI. Under President Xi Jinping,
the country has made tremendous
strides in many fields, but especially
in AI. Businesses and the government
have collaborated on a sweeping plan
to make China the world’s primary
AI innovation center by 2030, and
it’s making serious progress toward
that goal. That plan is unlikely to be
repealed by a new government; China
abolished Xi’s term limits and will
effectively allow him to remain in
power for life.
Within the next decade, China plans
to meet two crucial milestones: By
2027, its People’s Liberation Army will
have a modern-ready force, and by
2030 the Chinese Communist Party
(CCP) expects to have outpaced the
US in AI and become the dominant
China’s Expanding Market
It’s a challenging time for Chinese startups
because of rising tensions with the West.
Companies hoping to gain traction in Europe
are making efforts to cloak their origin.
Shein, the e-commerce website popular
among teens, says it was “founded in L.A.,”
but the company actually got its start in
Nanjing and Guangzhou by relying on the
region’s manufacturing centers and ample
supply chains. Or look at TikTok, which has
said it’s a US-based company—while the
app’s parent Chinese company ByteDance
has employed linguistic gymnastics to sepa-
rate itself. Binance, the world’s largest crypto
exchange, which was created in China, says
that it doesn’t have a headquarters located
in one physical location.
It’s no wonder that as Chinese startups
hope to expand globally, they’re seeking to
distance themselves from the authoritarian
regime in Beijing. But that creates political
hurdles, especially as the CCP seeks to bring
its home-grown technology ecosystem into
lockstep with party leaders.
The result could be a future parallel uni-
verse, in which Chinese-created AI systems
are shaped both by enormous amounts of
data and local laws. In Brazil, a generative AI
system might write an unfettered political
essay in Portuguese about a leader––while in
China, that same essay would be automatical-
ly filtered for politically sensitive words and
phrases. As the CCP enforces new regulations
targeting AI and what the government calls
“deep synthesis tech,” the ways in which peo-
ple experience and work alongside AI could be
dramatically different.
China’s Big Tech
Alibaba, Tencent, and Baidu, which have made
important advancements in AI research, may
find it difficult to keep innovating. Starting in
2020, the CCP initiated a wave of legislation
aimed at its tech sector, introducing anti-mo-
nopoly legislation focused on the platform
economy and promoting data security and
privacy laws. The Personal Information Pro-
tection Law (PIPL), China’s version of the EU’s
GDPR, went into effect in 2021. What followed
were a series of crackdowns targeting some of
force. China is producing what it
calls “intelligentized” technologies
to bolster both its economy and
military.
Recently, China took major steps to
shape the future of AI by releasing
its own pretrained models, and it is
forging ahead with its own natural
language processing models, which
makes sense since the most popular
models in use now are trained on
English text. China now has at least
130 LLMs, which accounts for 40% of
the global total, closely trailing the
US. Despite this rapid development,
investors and analysts caution that
many of these models lack sustain-
able business strategies, offer sim-
ilar functionalities, and face rising
operational expenses.

TECH
China’s most successful tech companies. Ul-
timately, this regulation wasn’t about “break-
ing up” China’s Big Tech—the CCP wanted to
focus its tech sector on achieving research
and development goals set by the govern-
ment and military within the decade.
Increasingly, Beijing is pressuring its me-
ga-successful big tech companies to share
data with the state and to perform research to
support the vision of the CCP. Going forward,
Beijing aims to direct the might of its tech
companies at programs of national strategic
importance rather than making video games.
China’s tech crackdown could cool private in-
vestment in Chinese companies, which could
result in a chilling effect on innovation and
economic growth, and also free up capital for
emerging markets.
Deepening International Ties
China is actively building out AI infrastruc-
ture and ecosystems, specifically focused on
developing nations. By focusing on the infra-
structure and the ecosystem, Beijing is not
just setting the stage—it’s constructing the
rest of the world combined.
This strategy mirrors China’s Belt and Road
initiative but instead of building physical
infrastructure in developing nations to
increase influence, China is building the
technological infrastructure, which includes
skills and data flow. However, US export con-
trols on key semiconductors and technolo-
gies to China present obstacles. In response,
China has taken measures such as prohib-
iting the use of chips from American com-
pany Micron in its infrastructure and imple-
menting a licensing system for the export of
specific essential metals like gallium and
germanium, which are crucial for high-end
semiconductors as well as components in
solar panels and electric vehicles.
As China shapes the world order in its own
image, it is simultaneously exporting its
technologies and surveillance systems to
other countries with authoritarian regimes.
When the CCP expands into African coun-
tries and throughout Southeast Asia and
Latin America, it will also begin to eschew
entire theater to establish Chinese-designed
AI systems.
Over 140 cities globally, from Kuala Lum-
pur to Nairobi, are being transformed into
“smart cities” and “safe cities” powered by
AI. Chinese companies are providing the
technology and expertise to supercharge as-
pects like transportation, logistics, and law
enforcement. China already leads the world
in exports of AI-enabled surveillance sys-
tems. China’s “Luban workshop” initiative
is another strategic move by China, offering
vocational training globally that includes AI
education. This has resulted in the creation
of a workforce skilled in AI in various devel-
oping nations. China also created a “BRICS
AI Study Group” to accelerate AI cooperation
with other developing economies. Chinese
tech companies even helped construct the
premier AI company in the UAE. Additionally,
China dominates the market for industrial
robot installations, having surpassed Japan
in 2013. The gap between China and other
countries has only widened—in 2021, China
installed more industrial robots than the
China is funding smart cities powered by Chinese AI
and surveillance technology in developing nations.
The country is exporting advanced monitoring
systems as part of a broader strategy to extend its
technological and geopolitical influence.

TECH
XX
operating systems, technologies, and in-
frastructure built by the West. Two Chinese
companies— the state-controlled CEIEC and
Huawei—built Ecuador’s surveillance sys-
tem, called ECU-911. The system promised to
curb high murder rates and drug crime, but
Ecuador could not afford the investment. As
a result, a deal was struck for a Chinese-built
surveillance system financed with Chinese
loans. It was a prelude to a much more lucra-
tive deal: Ecuador eventually signed away big
portions of its oil reserves to China to help
finance infrastructure projects. Similar pack-
age deals have been brokered in Venezuela
and Bolivia.
China is quietly weaponizing AI, too. China’s
People’s Liberation Army is catching up to
the US military, using AI for such tasks as
spotting hidden images with drones. The
Chinese military is equipping helicopters and
jet fighters with AI. The government created
a top-secret military lab—a Chinese version
of DARPA—and it’s building billion-dollar
AI national laboratories. China’s military is
achieving remarkable AI successes, including
a recent test of “swarm intelligence” that
can automate dozens of armed drones.
When it comes to AI, leaders should moni-
tor escalating tensions between the US and
China. But they should also remember that
there are cells of rogue actors who could
cripple our economies simply by mucking
with the power or traffic grids, causing traf-
fic spikes on the internet, or locking us out
of our connected home appliances. These
aren’t big, obvious signs of aggression, and
that is a problem for many countries, includ-
ing the US. Most governments don’t have
a paradigm describing a constellation of
aggressive actions. Each action on its own
might be insignificant. What are the escala-
tion triggers? Without a definition, a strate-
gic vulnerability exists.
China is quickly advancing military applications of AI, recently demonstrating swarm intelligence capabili-
ties to coordinate actions of dozens of armed drones.

TECH
WHAT IS EUROPE DOING?
In late 2023, The European Union fi-
nalized negotiations on its landmark
AI Act. This legislation establishes the
world’s first comprehensive frame-
work for regulating AI systems. The
overarching goals are to guarantee
AI safety, uphold ethical standards,
and drive European AI leadership.
Specifically, the EU AI Act classifies AI
systems into different risk categories
based on their use cases.
In February 2024, a new European AI
Office, established within the Euro-
pean Commission, was announced
to promote the development and use
of safe and trustworthy AI across the
EU, functioning as the core of a uni-
fied European AI governance system.
Through the implementation of the
AI Act, the office aims to safeguard
health, safety, and fundamental
rights, providing a stable legal en-
ble and flexible use in multiple lan-
guages for various tasks, claiming
to outperform or match other lead-
ing models on certain benchmarks.
The company uses a novel mixture
of experts (MoE) architecture, en-
hancing efficiency by routing tasks
to specialized neural networks,
making processing faster and less
resource-intensive. Mistral made its
models available for public use un-
der the Apache 2.0 license via Hug-
ging Face and BitTorrent—yes, the
same BitTorrent that gained notori-
ety housing illegally copied movies
and music and allowing downloads
via its peer-to-peer network—and
the company recently launched beta
access to its API for different levels
of Mistral models.
Germany also recognizes the geo-
strategic importance of AI innova-
tion to compete with American and
Chinese tech giants. A new hub in the
southeast city of Heilbronn aspires to
be a startup epicenter applying AI to
help German industrial leaders stay
competitive. Germany has committed
nearly 500 million euros toward AI
research and innovation, aiming to
enhance supercomputing infrastruc-
ture, skill development, and create
150 new professorships, with a focus
on achieving “technological sover-
eignty” and reducing its dependency
on external powers. German Federal
Minister of Education and Research
Bettina Stark-Watzinger is lobbying
for EU-wide cooperation in AI, partic-
ularly between Germany, France, and
Scandinavian countries, to position
Europe at the forefront of the global AI
landscape. Despite all these commit-
ments, concerns linger about the slow
vironment for businesses in all 27
member states. It will be responsible
for monitoring compliance and in
enforcing AI regulations.
France aims to advance its AI ca-
pabilities and influence. President
Macron promised more than $500
million to cultivate French AI “cham-
pions” and counter Silicon Valley’s
English-dominance in AI systems.
Mistral, a Paris-based AI company
founded by Arthur Mensch, Guillau-
me Lample, and Timothée Lacroix,
former AI researchers at Meta and
DeepMind, is gaining attention for
its rapid growth and focus on devel-
oping smaller, high-performance AI
models as an alternative to giants
like OpenAI. Unlike some larger,
more restrictive models, Mistral’s
offerings can run locally with open
weights, allowing for more accessi-

TECH
WHAT IS EUROPE DOING?
pace of integrating AI into the broad-
er economy and the potential stifling
effect of the EU’s AI Act on innova-
tion, highlighting the need for more
effective transfer of research to prac-
tical applications and the creation of
a robust AI-specific infrastructure.
Brexit continues to complicate Eu-
rope’s AI landscape. The UK gov-
ernment, following a white paper it
published in March 2023, decided
against introducing new AI-specif-
ic legislation, opting instead for a
pro-innovation regulatory framework
that leverages existing regulatory
powers to manage AI technologies.
This approach emphasizes high-level
principles such as safety, transpar-
ency, and fairness to guide regu-
lators, without imposing statutory
duties to ensure flexibility and adapt-
ability in AI oversight. One area still
ethics and norms that the Kremlin
opposes. Putin warned against the
“digital cancellation” of traditional
Russian culture by Western AI al-
gorithms, which he claimed often
exclude or ignore Russian contribu-
tions to culture, science, and litera-
ture. He pledged significant invest-
ment in supercomputers and other
technologies to enhance national AI
research, underscoring the need for
AI developments to be grounded in
Russian traditional values and cul-
tural heritage.
Putin is justifiably worried about
adopting a Western paradigm of
AI. Models like ChatGPT are trained
overwhelmingly in English and are
likely to exhibit the same assump-
tions as English-language media
that could contradict official nar-
ratives peddled by Russian media.
Major Russian tech companies like
Yandex and Sberbank are racing to
build their own rivals to ChatGPT.
But their offerings already lag be-
hind the accelerating innovation of
US and Chinese tech giants. Western
sanctions further hamper access to
vital computing power. Perhaps most
critically, Russia’s authoritarian at-
mosphere of censorship and distrust
conflicts with the very nature of imag-
inative, generative AI.
up for debate is intellectual property
across news and entertainment me-
dia. The House of Lords have called
for standardized regulatory powers
and meaningful sanctions to deter
wrongdoing––without explaining
what oversight would need to entail,
or how innovation can still be count-
ed on to stimulate the UK economy.
Finally, let’s not forget Russia, which
seeks to counter Western dom-
inance in AI. In November 2023,
Russian President Vladimir Putin
announced plans for the develop-
ment of an AI national strategy,
stressing that its focus would be to
prevent Western monopoly. He criti-
cized the “monopolistic dominance”
of foreign technology in Russia as
unacceptable and dangerous, high-
lighting that many AI systems are
trained on Western data, reflecting

TECH
WHAT IS THE MIDDLE EAST DOING?
The United Arab Emirates is posi-
tioning itself as a neutral ground for
the advancement of artificial intel-
ligence, aiming to bridge the gap
between the US and China amidst
ongoing geopolitical tensions. To
date, the UAE government has shown
deft diplomatic skills in navigating
complex international relations that
increasingly involve AI and other crit-
ical technologies.
Though the UAE government has
worked hard to remain neutral, its
companies are still caught in the
crosshairs between the ongoing race
between the US and China for AI su-
premacy. A major innovator based in
Abu Dhabi, G42, develops advanced
technologies across sectors like
space, health care, energy, and secu-
rity, but in December 2023, it faced
growing pressure to cut ties with
In parallel, the Kingdom of Saudi
Arabia has embarked on its own
sweeping economic diversification
agenda centered around AI. Through
its Vision 2030 plan, the kingdom
seeks to position itself for a future
where the global economy is less
dependent on oil and more driven
by technology and innovation. The
crown jewel of Vision 2030 is Neom,
a futuristic megacity under con-
struction aiming to seamlessly inte-
grate cutting-edge technologies like
robotics and AI across all aspects of
daily life.
The kingdom is backing its AI ambi-
tions with significant investments,
including $20 billion specifically
earmarked for advancing artificial
intelligence. It established the Saudi
Data and Artificial Intelligence Au-
thority (SDAIA) to drive the national
AI strategy. SDAIA initiatives like the
National Center for Artificial Intelli-
gence are designed to make Saudi
Arabia an AI leader across priority
industries such as health care. Global
tech giants have taken note, with Chi-
na’s Huawei recently launching a new
cloud data center in Riyadh to grow
its digital offerings in the region. The
facility will support AI applications
and Arabic language models to power
government services. Though Hua-
wei’s expansion may benefit Saudi
AI progress in the near term, Wash-
ington is likely to view such collabo-
rations with concern given its wider
technology rivalry with China.
The governance structures in KSA
and the UAE allow for swifter deci-
sion-making and implementation of
technology strategies compared to
democracies, where public opinion on
hardware suppliers such as Huawei.
The decision to phase out Chinese
hardware was also a move to pre-
serve G42’s access to US-made chips.
Also late in 2023, the government
launched a new state-sponsored AI
company, AI71, to commercialize its
leading LLM, Falcon. AI71 aspires to
directly compete with leading AI labs
like OpenAI. The UAE is also focusing
on nurturing its homegrown talent
in AI by investing in specialized edu-
cation. It established the Mohamed
bin Zayed University of Artificial
Intelligence (MBZUAI), recruiting
renowned experts from institutions
like University of California, Berkeley
and Carnegie Mellon as its faculty.
The university produces scores of
graduates annually, most of whom
work at local Emirati technology
companies.

TECH
WHAT IS THE MIDDLE EAST DOING?
issues like privacy and employment
significantly influences policy. These
nations have the financial resourc-
es to invest heavily in essential AI
components like GPUs, having spent
hundreds of millions on them, as well
as the energy required to power these
intensive processors.
Many Middle Eastern nations are positioning themselves as emerging AI hubs, aggressively investing in AI skills development, research, and entrepreneurship.
The goal is to diversify their economies in anticipation of declining reliance on oil.

TALENT

TECH
As AI permeates industries, demand
has soared for technical talent to
build and deploy AI capabilities. How-
ever, the rapid pace of innovation has
yielded a global AI skills shortage.
Employers struggle to attract talent,
especially when competing against
prestigious technology giants with
deep pockets. This breeds uncertainty
on optimal strategies, leading execu-
tives to wonder about the precise mix
of skills needed in their workforce,
whether current hiring approaches ap-
ply for burgeoning AI roles, and how to
evaluate the technical caliber required
for AI related work.
of AI skills across occupations, has revealed
that as of 2022, the regions leading in AI skill
penetration are India, the United States, and
Germany. The increasing prevalence of these
skills points to a transformative phase in the
job market, where AI proficiency is becoming
a critical asset for professionals in a multi-
tude of sectors.
AI Brain Drain from Academia
A striking shift has occurred in where AI Ph.D.
graduates build their careers. The Artificial
Intelligence Index Report from Stanford
shows an increase in the proportion of AI
Ph.D. graduates in North America entering
the industry after graduation, from 44.4%
in 2010 to roughly 48% in 2019, while the
percentage taking up academic positions
declined from 42.1% in 2010 to 23.7% in that
same period. The reason is clear: Compet-
itive salaries offered by the private sector,
along with the chance to work on applied
AI research, has tempted Ph.D.s away from
the classroom to corporate America. Lead-
ing AI organizations, such as OpenAI and
Anthropic, are offering starting salaries for
new researchers in the range of $700,000 to
$900,000, according to salary negotiation
service Rora, with Google even offering sub-
stantial restricted stock grants to attract top
data scientists. That’s orders of magnitude
higher pay than what even tenured professors
can expect from their universities. Top aca-
demics now earn generous corporate salaries
and benefits, and they get to work in a similar
tenured environment that’s carefully cultivat-
ed to replicate their experience in academia.
Tech companies are also endowing AI profes-
sorships at top universities. In some cases,
professors take one- or two-year sabbaticals
to work at tech companies and then return to
their universities. But corporate benefits can
be difficult to give up, and companies need
the talent. Poaching academia today could rob
the future of future AI experts: Without great
scholars, who will train the next generation of
innovators?
Demand for AI-related Skills Increases
Across Sectors
Employer demand for AI skills is rising rap-
idly across nearly every industry in the US.
With the exception of sectors such as agri-
culture, forestry, fishing, and hunting, there
has been a noticeable uptick in AI-related
job postings—from 1.7% of all postings in
2021 to 1.9% in 2022. Employers are actively
seeking individuals proficient in machine
learning, which tops the list of in-demand
AI skills, followed by knowledge in artificial
intelligence and natural language process-
ing. The surge in these specific areas un-
derscores a shift in the job market, with AI
skill clusters achieving greater prominence
than they had a decade prior. Demand for
Python skills has also increased, evidence
of its growing popularity as an AI coding
language. This increased AI skills demand
is not isolated to the US; it reflects a global
trend. The US leads globally for AI-related
job postings, followed by Canada and Spain.
Furthermore, LinkedIn’s AI skill penetration
rate metric, which assesses the prevalence
WHERE AND HOW DO I GET AI TALENT?

TECH
HOW WILL AI CHANGE THE NATURE OF WORK?
In a clever study released last summer, researchers from the US demonstrated
that shortly after ChatGPT was introduced, copywriters and graphic designers
on prominent online freelancing sites experienced a notable decrease in job op-
portunities, along with steep declines in their earnings. Here’s the rub: Genera-
tive AI wasn’t only replacing their jobs, it was diminishing the value of the work
they are trained to do.
It’s become clear that AI will change how we do work, where we do work, and
what work needs to be done in ways that are both piddling and profound. This
breeds equal parts excitement and anxiety. While fears persist of mass job
elimination, experts emphasize AI more as augmenting than replacing human
roles—though the truth remains unclear. Furthermore, as AI becomes more
sophisticated, able to collaborate with humans and demonstrate capabilities
once considered uniquely human, it raises the question: Is AI a tool for workers
to use or a colleague to collaborate with? The answer may determine how readi-
ly people embrace working with AI.
Gains and Pains
The integration of AI promises to both
enhance efficiency for some roles while
making other jobs more challenging. In the
financial services sector, for instance, a
study has highlighted how the integration
of AI systems is increasing the demands on
middle management. While AI applications
efficiently handle routine tasks formerly
executed by humans, this shift necessitates
that managers adapt to new challenges and
demands, navigating a landscape where they
must balance traditional management roles
with the oversight of AI systems.
Yet in medicine, AI is generally positioned as
an empowering asset to augment clinicians
rather than replace them. Physician-re-
searchers at Beth Israel Deaconess Medical
Center showed an AI chatbot surpassing
human accuracy at probabilistic reason-
ing to aid diagnoses. Separately, scientists
at University College London developed AI
speech pattern detection tools to uncover
early schizophrenia indicators. Rather than
substituting the role of the physician, these
technologies enable doctors to discern addi-
tional insights.
Status Shifts
Emerging research suggests AI may profound-
ly reshape perceptions of high-status occu-
pations in the coming years by automating
prestigious skills. Studies from the OECD
and OpenAI forecast mass job losses even
in respected professional domains like law,
medicine, and finance. However, AI might also
democratize skills that have long been asso-
ciated with high-status roles. For instance, the
use of AI like GPT-4 in professional services
has shown that those leveraging these tools
can outperform their peers across various
tasks. LLMs might be just as good—or even
better—at certain jobs in law firms because
the work can be automated for faster, cheaper
results. The use of LLMs in law could change
how legal work is done, potentially reducing
the need for junior lawyers for routine tasks
and forcing legal process outsourcing firms to
change their business models, while also of-
fering law firms and legal departments signif-
icant efficiency gains and cost savings. This

TECH
HOW WILL AI CHANGE THE NATURE OF WORK?
suggests that AI has the potential to level the
playing field, allowing individuals who might
not have traditionally excelled in these roles
to boost their performance and productivity.
This shift also brings a potential upheaval in
the power dynamics traditionally associated
with specialized skills and knowledge. High
verbal intelligence, once a marker of elite sta-
tus and a key to high earnings, might lose its
prestige as AI begins to outperform humans
in tasks involving language and writing.
Skills such as writing proficiency or multi-
lingualism, previously indicators of a highly
educated individual, may diminish in value
as AI improves text quality and eliminates
language barriers.
The possible reduction in status and influ-
ence for those skilled in words and symbols
is a significant cultural shift. For centuries,
Western societies have revered those adept at
conceiving and communicating new con-
cepts. The rise of the Scientific and Industrial
revolutions only amplified their status and
influence. The encroachment of AI into these
prioritizing features, and developing busi-
ness cases. A developer AI agent would focus
on automating code generation, refining
existing code, and aiding in bug detection.
But these agents wouldn’t be siloed to
interaction with humans—they can interact
with one another just as humans would to
accomplish a goal. Imagine an ecosystem
populated by specialized AI agents, each
equipped with distinct expertise and knowl-
edge, designed to not only assist individual
tasks but also to collaborate and interact
with one another. This vision points toward
a future where AI agents evolve from per-
forming singular, user-specific tasks— like
drafting emails, resolving customer support
queries, or managing grocery orders—to op-
erating within a network where they commu-
nicate and cooperate with other agents.
Companies that possess extensive data
repositories in specific verticals are likely to
emerge as leaders in the AI Agent space. For
example, Bloomberg, with its rich trove of
financial data, is well-positioned to develop
sophisticated financial AI agents. It’s already
begun creating large language models tailored
to finance. Similarly, LexisNexis, with its vast
legal information database, could develop
AI agents specialized in the legal domain.
These AI agents, drawing from deep wells of
domain-specific data, would not only cater to
their direct users but also become invaluable
assets to other businesses, systems, and AI
agents.
realms could represent a profound and
unprecedented shift in societal values and
the stratification of labor. In this new era,
jobs that are less susceptible to automation,
such as skilled trades, might gain in pres-
tige and economic reward relative to those
more easily disrupted by AI technologies.
Agents Will Increasingly Perform Tasks on
Our Behalf
The workforce is increasingly familiarizing
itself with chatbots to perform routine tasks
like drafting emails and synthesizing tech-
nical language and documents. AI agents
would take this a step further by active-
ly performing tasks like sending emails,
scheduling meetings, and booking reserva-
tions. These agents represent a shift toward
a more proactive and autonomous model,
transcending the capabilities of traditional
chatbots. AI agents could book your vaca-
tion, coordinate dinner reservations among
friends’ calendars, or perform specific tasks
relevant to a particular role at a company.
For example, a product owner AI agent could
specialize in aiding with market analysis,

EMERGING
CAPABILITIES

TECH
Can AI think like a human? Recent advances suggest we may be close to un-
locking AI’s potential for complex reasoning, and perhaps even one day achiev-
ing
artificial general intelligence (AGI), shorthand for a computer that can do
anything a human brain can. Microsoft recently startled the AI community, be-
coming the first major tech company to argue current systems exhibit sparks of
AGI. While it never made that same proclamation, Google’s DeepMind team has
repeatedly demonstrated nascent AGI capabilities in its research.
AGI refers to flexible, human-level cognition able to tackle any intellectual task.
Yet progress has been uneven across the spectrum of what a human brain is
able to do. Contrary to sci-fi visions of AI’s prowess at logic and math, early
breakthroughs in artificial intelligence were primarily in creative realms like
art and language modeling. Initially, it was believed that AI would excel in rea-
son-based jobs, particularly in mathematics, given computers’ inherent profi-
ciency in handling numbers and calculations at a speed far surpassing human
capabilities. However, the evolution of AI has taken a somewhat unexpected
turn, veering more towards creative applications rather than purely logical rea-
soning. That is, until now. Looking beyond AGI, the ultimate frontier is artificial
super intelligence (ASI)—AI that surpasses human intelligence in every aspect,
from creativity to problem-solving, heralding an era where AI’s capabilities
could transcend human limitations.
AI Breakthroughs in Mathematics
A breakthrough in AI’s mathematical abili-
ties was showcased by DeepMind’s Alpha-
Geometry. In a landmark paper published in
Nature, AlphaGeometry demonstrated its ca-
pability to solve complex geometry problems
at a level comparable to a human Olympiad
gold medalist. It successfully solved 25 out
of 30 Olympiad-level geometry problems
within the standard time limit, a perfor-
mance on par with top human competitors.
The success of AlphaGeometry highlights AI’s
growing capacity for logical reasoning and
knowledge discovery. AlphaGeometry effec-
tively showcases a process that mirrors real
thinking. Its process has been compared to
the dual-process theory of thinking, Type I and
Type II, as popularized by psychologist Daniel
Kahneman in “Thinking, Fast and Slow.” Also
from the DeepMind team, a technique involv-
ing LLMs named FunSearch has demonstrated
that AI can assist mathematicians in solving
wicked problems, inspired by the card game
“Set.” This marks the first instance where an
LLM-based system has been able to surpass
existing mathematical and computer science
solutions, proving yet again that AI can solve
a wide array of math and compsci questions
more effectively than human mathematicians
working alone. FunSearch works by generat-
ing and testing short computer programs for
solving mathematical problems, refining its
approach through feedback, and represents a
novel form of human-machine collaboration
that could amplify the capabilities of human
mathematicians rather than replace them.
CAN AI REASON? AND HOW CLOSE ARE WE REALLY TO AGI AND ASI?
Will AI take over the
world? No, this is a
projection of human
nature on machines.
—Yann LeCun,
vice president and chief AI scientist at Meta

TECH
AI Persuasion
Logical argument is core to persuasion, but
emotional resonance and validating existing
views profoundly shape what people are con-
vinced of too. AI shows promising aptitude on
both fronts—generating seemingly rational ar-
guments while precisely targeting psycholog-
ical triggers. OpenAI CEO Sam Altman recently
warned superhuman persuasiveness may
arise in AI before general intelligence does,
with unpredictable outcomes. AI chatbots,
like OpenAI’s ChatGPT, have demonstrated an
impressive level of conversational prowess;
they can sound convincing, even when pro-
viding incorrect information, which is partic-
ularly troubling when considering the human
tendency to form emotional connections with
these systems. Evidence shows even limited
interactions with AI chatbots promotes at-
tachment and trust, amplifying their capacity
for conviction.
Researchers at Stanford’s Polarization and
Social Change Lab and the Institute for Hu-
man-Centered Artificial Intelligence conduct-
ed studies to explore AI’s capabilities in sway-
have developed methods for organizing this
data to forecast future events in an individu-
al’s life. Notably, their model, dubbed Life-
2vec, can make predictions about profoundly
significant events, including estimating the
time frame of a person’s death.
The cutting edge in AI reveals accelerating
abilities to computationally interpret inte-
gral aspects of the human experience—from
life outcomes to subjective thought itself.
Recent research has demonstrated AI’s
ability to not only forecast significant life
events but also to delve into the depths of
human cognition by reconstructing images
seen by individuals, based solely on brain
scans. A team from Osaka University in Ja-
pan has achieved a groundbreaking feat in
cognitive AI. By analyzing functional mag-
netic resonance imaging (fMRI) scans taken
while subjects viewed specific images, the
AI system they trained was able to recreate
these images with surprising accuracy. The
AI generated visuals of a teddy bear, a clock
tower, and an airplane, among other objects,
after participants had looked at similar
items. This research marks a significant step
forward in AI’s ability to interpret and visual-
ize human thoughts based on neurological
data. The implications of these advancements
are profound. AI’s ability to predict life events
suggests a future where technology could offer
insights into personal and societal trends
with unprecedented accuracy. Meanwhile,
the capacity to reconstruct visual experienc-
es from brain scans opens new avenues for
understanding human cognition, memory, and
perception.
More practically, Nvidia is developing an
AI-powered “digital twin” of Earth, known as
Earth-2, leveraging its FourCastNet AI model
to predict weather with unprecedented speed
and accuracy, outperforming traditional meth-
ods by forecasting thousands of potential out-
comes. This breakthrough in climate modeling
represents a huge advance in applied research.
Detecting Emotion
A new type of neural network can determine
how people are feeling. Using radio waves, AI
can detect subtle changes in heart rhythms,
ing public opinion on contentious political
issues. Their findings were alarming: AI-gen-
erated arguments were as persuasive, if not
more so, than those penned by humans on
a range of topics. For example, AI-crafted
messages on policies like smoking bans and
carbon taxes significantly shifted readers’
support. While highlighting AI’s influential
potential, researchers in parallel sound
alarms on misuse by hostile actors. As
models continue absorbing the intricacies
of human psychology while simultaneously
continuing to improve at logic-based per-
suasion, safeguarding against deception
emerges paramount.
Prediction and Prescience Into Our
Human Lives
Will I die within four years? This is one of
the questions that a collaborative research
project is pushing AI to be able to answer. By
leveraging large data sets detailing various
aspects of people’s lives and employing
transformer models—similar to those under-
pinning the language processing capabilities
of systems like ChatGPT—the researchers

TECH
run a pattern analysis, and predict someone’s
emotional state in a given moment. A team
from Queen Mary University of London used a
transmitting radio antenna to bounce radio
waves off test subjects and trained a neural
net to detect fear, disgust, joy and relaxation,
as people were shown different videos. The
system accurately tagged emotional states
71% of the time, which signals new opportuni-
ties for health and wellness applications, as
well as for job interviews and the government/
military intelligence community. The EU is
sponsoring a pilot project called iBorderCtrl
that uses emotion recognition technology
to assess truthfulness in border crossing
interviews—the system analyzes interviewees’
micro-expressions and nonverbal cues in an
attempt to quantify the likelihood of decep-
tion during questioning. However, emotion
recognition technology is still emerging and
its accuracy in quantifying human emotion re-
mains unproven, given the inherent complex-
ity and nuance of human expression. Some of
the most advanced emotion recognition tech-
nology is currently being developed in China,
where extensive work has been done on facial
Amit Sheth, who founded the Artificial Intel-
ligence Institute at the University of South
Carolina, is exploring a new idea called neu-
ro-symbolic vision. This approach is similar
to how we, as humans, understand the world:
by turning what we see and hear into sym-
bols in our minds, and then using what we
know to make sense of those symbols, make
plans, and take actions. This way of process-
ing information is also how we explain our
thoughts and actions to others, which is
especially important in areas like health care
where trust is key. Neuro-symbolic AI aims
to improve how smart systems figure things
out and make them more accountable. By
combining the learning power of neural net-
works with organized knowledge (like facts
and rules), we could see big improvements
in AI’s ability to understand concepts, make
connections, and reason about the world in a
way that’s clear to us. As people start ques-
tioning current AI methods, this neuro-sym-
bolic approach could lead us toward creating
AI that thinks more like humans do, which
could be a big step toward achieving AGI.
recognition systems, albeit amid ethical con-
cerns over potential misuse—the country has
faced scrutiny for employing emotion AI to
enable surveillance, most notably to monitor
the Uyghur population.
Neuro-symbolic AI
Neuro-symbolic AI combines the best of two
worlds in AI: the learning capabilities of neu-
ral networks (which are good at handling un-
structured data like images and language)
and the reasoning capabilities of symbolic AI
(which deals with structured data and logic).
For businesses, this means they can create
smarter systems that not only learn from
vast amounts of data but also understand
and apply rules and logic, similar to human
reasoning. In practical terms, this means
that a neuro-symbolic AI could analyze a
company’s data and also understand the
context, making decisions that are more
accurate and relevant to specific business
scenarios. By understanding rules and logic,
neuro-symbolic AI might automate tasks
that previously required human understand-
ing, saving time and reducing errors.
AI can detect emotions through facial analysis and
by tracking subtle biological clues like changing
heart rhythms.

TECH
Where we will ultimately deploy AI
workloads remains an open question.
Many anticipate the future is likely to
embrace a hybrid approach that com-
bines cloud, edge, and on-device com-
puting in some capacity. This strategy
allows for data processing and model
training to leverage the vast parallel
processing power of cloud servers.
Meanwhile, edge hardware and local
devices could handle real-time infer-
ences and personalization, optimizing
for both performance and privacy. But
the specific balance across environ-
ments and when to favor one over the
other is still unclear as capabilities
and demands evolve.
Cloud Strain From AI Boom
AI has arrived, but the underpinnings of
the cloud may struggle to withstand its
weight. Cloud providers such as Amazon
Web Services, Microsoft Azure, and Google
Cloud are under intense pressure to adapt
their services to accommodate the needs of
large-scale generative AI models, which can
be up to 100 times larger than their prede-
cessors. Generative AI models like ChatGPT
that produce original text and analysis can
be 10 to 100 times more complex than a
Google search. The current cloud infrastruc-
ture, primarily designed to provide scalable,
pay-as-you-go services for diverse workloads
through general-purpose computing, is now
significantly challenged by the demands of
AI-intensive workloads.
Only a small portion of current cloud serv-
ers are outfitted with AI-optimized GPUs
or structured to function in collaborative
clusters, essential for meeting the substan-
tial computational requirements of AI tasks.
A significant bottleneck also arises from the
scarce availability of high-performing GPUs,
with Nvidia essentially serving as the sole
supplier. Because of high demand, Nvidia’s
H100 graphics—an earlier version of their
most powerful graphics—sold for more than
$40,000 on eBay. To reduce their dependence
on Nvidia, companies like Alphabet, Microsoft,
and Amazon are developing their own AI chips
for model training. Despite their cloud plat-
forms not being fully optimized for AI, AI work-
loads are contributing to significant revenue
growth in their cloud infrastructure.
AI Breathes Life Into Legacy Systems
The rising costs associated with cloud com-
puting, especially for tasks like training AI
models, are prompting some companies to
reconsider on-premises solutions. Dell Tech-
nologies, recognizing this shift, has developed
servers specifically designed for on-premis-
es AI deployments. By moving AI operations
in-house, Dell argues that companies can po-
tentially save on networking and data storage
expenses. Furthermore, AI is playing a pivotal
role in revitalizing legacy mainframe sys-
tems. Over 800 billion lines of COBOL code are
currently in use within production systems,
Cloud Neutrality
A handful of companies control the cloud
and have the ability to set pricing, access
and standards. Those companies own the
infrastructure and don’t have to make their
business practices transparent. Generative
AI systems require enormous amounts of
costly computing power and cloud infra-
structure, which the tech giants are trading
for future shares of profit. This consolidates
additional power among the largest cloud
providers. As more of our businesses and as-
pects of our lives move to the cloud, efforts
will grow to ensure that infrastructure serves
the public interest. The three biggest cloud
providers, Microsoft, Amazon, and Google,
have collectively invested tens of billions of
dollars building infrastructure: data centers,
monitoring systems and software. Their ro-
bustly designed systems prevent downtime
and data loss, and few other companies in
the world can compete. But the cloud isn’t
public infrastructure; it’s private. And as
private companies, cloud providers currently
control access to services that are becoming
the lifeblood of businesses.
IS THE FUTURE OF AI CLOUD, EDGE, OR ON-DEVICE?

TECH
making the transition from this language,
established in 1959, to more contemporary
languages a daunting task. The scarcity of
COBOL experts—many are nearing retirement
age—and the complex nature of migration
efforts for large organizations further com-
pound these challenges. IBM’s introduction of
Code Assistant for IBM Z, an AI-powered tool
that translates COBOL code into Java, offers
a solution to modernize mainframe applica-
tions with the help of AI. This blend of AI in-
novation not only supports the shift towards
on-premises AI deployments to manage costs
but also demonstrates the potential for AI to
breathe new life into legacy infrastructures.
Optimizing AI to Run at the Edge
Smart devices like phones lack the memory
and computing power required to fine-tune
AI models with user data over time. This
limitation has necessitated transmitting
personal information to the cloud for updat-
ing, an energy-intensive process that risks
data privacy. Now, advances like PockEngine
enable efficient on-device learning without
offloading data. Developed through an MIT
easier integration into edge devices. For
mobile and embedded use cases, massive
cloud-based LLMs are often impractical.
Their substantial size and latency makes
local deployment a non-starter. More com-
pact models in the millions or single-digit
billions of parameters, however, could po-
tentially run efficiently on smartphones and
IoT devices. Your washing machine could be
equipped with a compact language model,
enabling you to inform it verbally that you’re
washing a mixed load and are concerned
about a sweater washing in overly warm
water. The small language model that can
run in the appliance eliminates the need for
internet connectivity to operate your wash-
ing machine in this manner. SLMs could
therefore empower voice assistants, smart
home automation, and beyond, reducing
the dependency on cloud-based services for
these types of applications.
On-Device AI
Tech giants such as Samsung, Microsoft,
Google, and Apple are spearheading a move-
ment towards on-device AI, emphasizing a
blend of performance and privacy. These com-
panies are competitively equipping their de-
vices with specialized AI chips to enable local
processing, thereby reducing reliance on cloud
servers. This approach to on-device AI process-
ing is motivated by the goal of safeguarding
sensitive data, drastically cutting down the
risk of data breaches during its transfer to
and from the cloud. Moreover, on-device AI has
the unique capability to adapt and personalize
according to a user’s behavior directly on the
device. Samsung introduced its Galaxy S24
smartphones, showcasing a leap in AI capa-
bilities with the implementation of generative
AI tools that operate through a combination
of on-device processing and cloud-based
computations. Google’s latest Pixel phone
features custom AI silicon to handle tasks like
predictive typing more responsively on-device.
Apple’s newest MacBook CPU incorporates
neural processing units for faster machine
learning. AMD’s latest Ryzen mobile chips
similarly target laptop enhancements like
voice assistance.
and IBM collaboration, PockEngine is a train-
ing model that selectively identifies which
specific parts of an otherwise enormous
model to update locally based on a user’s
unique inputs. By focusing only on essen-
tial parameters and shifting computations
to preprocessing, PockEngine minimizes
real-time resource usage. Not only does this
make it more efficient, it also facilitates
the creation of personalized deep-learning
models. For instance, AI assistants can con-
tinuously adapt to a user’s accent or typing
patterns without reliance on constant cloud
connectivity. Tests demonstrate PockEngine
fine-tuning complex models up to 15x faster
than alternatives, all while maintaining or
boosting accuracy.
Small Language Models for AI at the Edge
While large language models with billions or
trillions of parameters have demonstrated
impressive capabilities, smaller AI models
may be better suited for edge-based use
cases. Though less broadly capable, special-
ized mini-models bring benefits like faster
inference, lower compute requirements, and

TECH
Wearable AI
AI is changing human-computer interac-
tion, shifting us away from screens, track-
pads, and keyboards towards more intuitive,
voice-based interfaces. This is giving rise to
a new class of lightweight, wearable gadgets
and screenless computers that integrate
seamlessly into daily life. By reducing screen
fatigue and intrusive features, these de-
vices foster a more natural, human-centric
approach to technology. A prime example is
the newly launched Humane AI Pin, an Ope-
nAI-powered wearable priced at $699, plus
a $24 monthly subscription. Forgoing tradi-
tional app interfaces, this 34-gram device
focuses solely on voice interactions. Users
access information and perform tasks by
speaking to the Pin’s built-in microphone.
By stripping down the technological inter-
face, Humane aims to create a streamlined,
human-like experience. Another device is the
Rewind AI Pendant, which captures real-world
conversations, storing encrypted transcripts
and audio locally on the user’s phone. Be-
yond recording, Rewind’s platform searches
transcripts, generates meeting summaries
and analyzes speech patterns. Essentially,
the Pendant serves as a personalized assis-
tant harnessing environmental information
to support the user. Both the Humane Pin
and Rewind Pendant epitomize the shift
towards invisible, assistive technology that
facilitates life’s tasks and interactions much
like a helpful human companion would. This
evolution in form and function represents a
paradigm shift, integrating technology more
seamlessly while making it feel more intui-
tive and human-centric.
Note: we’ve included this trend in both the AI
and Computing reports. We think it is important
to consider the near-future of wearables as you
contemplate the future of your organization and
AI’s development.
AI can let us get information by voice requests rather than typing search terms or looking through folders.
This more natural interaction could drive demand for wearable or voice-based interfaces.

TECH
Businesses should keep an eye on
emerging AI capabilities because
these technologies can unlock new
opportunities for innovation, effi-
ciency, and competitive advantage.
Early awareness and adoption of AI
advancements can position a compa-
ny as a market leader, enabling it to
refine its operations, enhance cus-
tomer experiences, and create novel
products or services.
their product offerings, and funding flows
into vector database startups, adoption will
accelerate. By 2026, over 30% of enterprises
are expected to implement vector databases
to support their AI models. This trend signals
a skills shift as well, with data and software
engineering teams needing more knowledge
of techniques like semantic search and vec-
tor indexes to successfully leverage vector
databases for AI use cases.
Vertical Integration From Hardware to LLMs
Companies are increasingly adopting a ho-
listic approach to AI development, seeking to
dominate the entire spectrum from hard-
ware to LLMs through end-to-end vertical
integration. This strategy would allow com-
panies to oversee the full pipeline, from the
foundational hardware to the sophisticated
AI models that drive innovation. Nvidia, a
titan in the realm of AI hardware, is now
speculated to potentially broaden its scope
into cloud computing services. By capitaliz-
ing on its hardware prowess, Nvidia could of-
fer comprehensive AI cloud services, further
cementing its role in shaping the AI domain.
Vector Databases
Vector databases are poised to grow rapidly
in importance alongside advancements in
AI. As AI models like large language models
become more capable of human-like gen-
eration across modalities like text, images,
and audio, they rely heavily on vector repre-
sentations of data, known as embeddings,
to understand and generate contextual
meaning. To function optimally, these gen-
erative models need databases specifical-
ly designed to store massive vector data
sets and allow instantaneous retrieval of
semantically similar vectors. This is where
vector databases come in; they are uniquely
designed to efficiently store, manage, and
retrieve high-dimensional vector data, which
is crucial for embedding processes found in
natural language processing, image gen-
eration, and other AI applications. Unlike
traditional databases that organize data in
rows and columns, vector databases use
vectors to represent data points, enabling
faster and more relevant data retrieval based
on similarity. As companies like Microsoft
and Oracle introduce vector databases into
WHY SHOULD WE PAY ATTENTION TO EMERGING CAPABILITIES THAT
AREN’T YET FULLY DEVELOPED?
If you aren’t ahead, you are already behind. Proac-
tively assessing how innovations apply to their
operations and offerings will help companies capi-
talize on advances and stay competitive.

TECH
Nvidia’s GeForce Now, a cloud streaming
service, already demonstrates the compa-
ny’s capability to merge high-performance
hardware with cloud-based offerings, hinting
at a future where Nvidia’s influence extends
across the AI ecosystem. In February 2024,
Nvidia demoed a personalized AI chatbot for
Windows PCs that connects to local files,
enabling natural language queries such as
“what restaurant did my friend recommend?”.
Rather than searching manually, users can
query the chatbot directly to retrieve informa-
tion from personal notes and messages.
Meanwhile, cloud AI providers like Amazon,
along with emerging AI startups like Anthrop-
ic and Mistral, currently depend on third-party
hardware for their AI operations. This depen-
dency poses the question of whether these
entities might emulate OpenAI’s strategy of
procuring their own chips. In early 2024, Sam
Altman, OpenAI CEO, indicated that he plans
to raise billions for an AI chip venture aimed
at developing a network of factories for fabri-
cation. Intel’s foray into AI software develop-
ment further illustrates this trend. Leveraging
one of its supercomputers, Intel has built
a generative AI system capable of process-
ing text and images. This initiative not only
showcases Intel’s commitment to advancing
AI capabilities but also emphasizes the stra-
tegic value of controlling both hardware and
software components in delivering sophisti-
cated, secure, and efficient AI solutions.
WHY SHOULD WE PAY ATTENTION TO EMERGING CAPABILITIES THAT
AREN’T YET FULLY DEVELOPED?
Companies are now adopting a comprehensive strategy for AI, covering everything from hardware to LLMs.
They aim for control over the entire AI development process through vertical integration.

INDUSTRIES

TECH
Many companies have new competitors—they just don’t realize it yet. The
boundaries between sectors are blurring; professional services firms tradition-
ally known for consulting are now venturing into engineering, powered by AI
technologies. Similarly, big tech hyperscalers, once primarily focused on build-
ing and hosting tech infrastructure, are expanding into consulting services.
This crossover signifies that AI’s versatility and capability to add value across
different functions are enabling companies to enter and compete in domains
previously beyond their reach. Consequently, businesses may find themselves
up against competitors from entirely different industries, underscoring the
need to innovate and adapt strategies in response to the unpredictable dynam-
ics AI introduces to the market.
INDUSTRIES
While AI may not directly replace every job, it positions those who embrace its capabilities to outperform
and replace those who do not.

TECH
AI is enabling HR departments to
automate time-consuming adminis-
trative tasks like screening job appli-
cants, while also providing insights to
enhance employee retention, training,
development and engagement. From
personalized onboarding chatbots
to performance comparison analyt-
ics, companies are unleashing AI to
expedite recruiting, predict attrition
risk, optimize benefits, identify pro-
ductivity barriers and mitigate bias in
reviews. Though valid ethical concerns
remain, AI has significant potential
in HR to both improve experiences
for employees and drive better overall
business performance.
analyze hundreds of details, such as the
tone of voice, facial expressions, and man-
nerisms to best predict how a candidate will
fit in with the culture of a community. Start-
ups such as HireVue use AI systems to help
companies decide which candidates to hire.
But this kind of recognition technology has
practical applications beyond job interviews:
It can detect when someone is likely to
make a purchase—or attempt to shoplift—
in a store, whether someone is lying, and
whether someone is receptive to new sug-
gestions and ideas. Unlike security cameras,
which tend to have a light indicating they’re
recording, algorithms work invisibly, which
means that this is an area that could face
regulatory scrutiny. The consumer advocacy
organization Electronic Privacy Informa-
tion Center filed a complaint with the FTC
requesting an investigation into HireVue,
alleging its tools produce results that are
“biased, unprovable, and not replicable”
through algorithmic models.
Benefits Selection and Management
AI automation is taking over the complex
tasks of managing employee benefits, includ-
ing facilitating open enrollment, tracking indi-
vidual coverage, and making adjustments due
to life changes. This simplifies workflows for
HR teams and provides employees smoother,
more reliable experiences with their benefits.
Startups like Paidleave.ai offer AI chatbots to
assist workers in understanding and utilizing
paid leave benefits. Major HR systems provid-
ers like ADP are also releasing AI assistants,
such as ADP Assist, to help HR managers han-
dle common inquiries and provide data-driven
insights. By automating benefits adminis-
tration, AI enables HR staff to focus on more
strategic tasks while empowering employees
through intuitive self-service tools.
Autonomous Talent Acquisition
AI automation can significantly reduce the
time and cost of recruiting by handling
tedious, manual tasks like screening re-
sumes, scheduling interviews, and tailoring
outreach. Johnson Johnson leveraged AI
writing tools to reduce unconscious bias in
job descriptions, improving gender diversity
in applicants. AI also assists with onboard-
ing tasks like verifying employee paperwork,
delivering induction training, and providing
system access. By automating repetitive HR
workflows, AI allows recruiters and manag-
ers to focus their human skills on building
relationships and strategic planning. Overall,
AI promises major gains in operational effi-
ciency, cost savings, and unbiased, person-
alized experiences for both recruiting and
onboarding processes.
Customer and Personnel Recognition
Systems
Recognition systems can now be deployed
to watch people in an interview and gauge
enthusiasm, tenacity, and poise. Algorithms
HOW IS AI BEING USED IN HR?

TECH
AI will change marketing in big ways.
Algorithms can study lots of custom-
er data to understand what people
want. This lets marketers create very
tailored ads and content for each
person. AI chatbots can also have
friendly conversations to help cus-
tomers. Perhaps more importantly,
AI shifts how buyers find and choose
products in the first place. By chang-
ing the platforms people use, their
behaviors change too. Marketers
should fully rethink strategy as AI
transforms what makes people dis-
cover and buy things.
dation woven into the consumer experience.
Failure to adapt approaches could prove
highly risky in the coming years.
Dynamic Engagement Through Deep
Personalization
Traditional marketing communications
like emails, PDFs, and social posts have
been static and one-way, but AI is ushering
in a new era of responsive, conversational
messaging. Chatbots and virtual influencers
allow for personalized interactions where
content changes based on the user. For
example, Meta leverages AI characters based
on celebrities like Snoop Dogg and Kendall
Jenner to engage audiences through gam-
ing and advice. While not real people, these
bots represent AI’s ability to gather data and
connect with users in a more humanized,
tailored way. As this technology advances,
marketers can leverage AI to deliver deeply
customized content that dynamically adapts
to individuals’ preferences and behaviors
in real time. This interactivity creates more
meaningful engagement between brands
and consumers.
AI-Assisted Campaigns
Major digital advertising platforms like Meta
and Google are unveiling new generative AI
capabilities to assist advertisers in stream-
lining campaign creation. In May 2023, Meta
launched AI Sandbox—a “playground” for
testing AI-powered ad tools. Features include
intelligent text variation to auto-generate
messages optimized for different audienc-
es, background image generation from text
prompts, and image resizing to fit multiple
social media formats. Meanwhile, Google
expanded its Gemini conversational AI that
creates full search campaigns from a single
landing page URL provided by the advertiser.
After some human tuning, Gemini’s chatbot
can collaborate with advertisers on campaign
objectives, target segments, and ideas for ex-
tra ad content. These tools automate tedious
creative tasks, allowing advertisers to instant-
ly produce customized images, text, and even
full campaigns tailored to their goals. And
generative AI abilities like text-to-image, text-
to-video, and text optimization further accel-
erate campaign ideation and production.
AI Shifts Search
Early data signals that the rise of AI tools
like ChatGPT may be subtly reducing Goo-
gle search volumes. While the search giant
still dominates with over 90% market share,
metrics show marginal declines coinciding
with surging interest in conversational AI.
Rather than competitors like Bing stealing
share, this hints at a more fundamental
shift—people using search less because
AI applications can directly provide infor-
mation. For marketers who have invested
heavily in search engine optimization, this
presents a seismic challenge. If traffic from
search shrinks in favor of on-device intel-
ligent assistants, prevailing strategies get
disrupted. The expected launch of AI models
from Apple, Google, and others threaten an
even greater paradigm change toward inte-
grated, device-based discovery rather than
browser-led journeys. In essence, where and
how people find products appears poised for
disruption. Marketers must prepare for an
upcoming inflection point where search-cen-
tric models cede ground to AI-powered, om-
nipresent product discovery and recommen-
HOW IS AI BEING USED IN MARKETING?

TECH
Anecdotal Observations, Now Usable
Marketing Data
Until recently, subtle human interactions and
reactions, like micro-expressions, were merely
anecdotal insights. However, advancements
in AI now allow us to quantify these observa-
tions and transform them into quantifiable
marketing data. Companies like Chooch use
Vision AI to efficiently search video data and
discern facial cues to understand consumer
engagement. In physical stores, similar tech-
nology can monitor customer responsiveness
to branding. Essentially, AI can convert once
subjective perceptions into hard analytics
to better personalize experiences. However,
while this data enables deeper personal-
ization, ethical questions remain regarding
consent and privacy when collecting such
intimate human insights. As the technology
progresses, regulations and corporate respon-
sibility practices must also evolve to protect
and respect consumers.
HOW IS AI BEING USED IN MARKETING?
Companies must strike a balance between responsibly using consumer data to provide personalized offerings while avoiding intrusive tracking that could under-
mine customer trust.

TECH
With a history stretching back to the 1960s as one of the earliest adopters of
computer technology, the pharmaceutical industry is now rapidly integrating
AI into drug discovery. By applying advanced algorithms to harness vast data
sets—from genomics to clinical trials—AI enables more targeted identification
of promising candidates and illuminates their interactions with disease path-
ways. This streamlines the overall RD process, heightening productivity and
success rates while lowering costs. The acceleration and efficiency afforded by
AI promises to expand treatment options for previously untreatable diseases.
The gap between data-intensive computational labs and traditional wet labs is
closing, with AI-designed molecules already advancing to clinical trials.
In short, the long-developing foundation of computing in pharma is now bear-
ing fruit in the form of transformative AI applications spanning candidate
screening to preclinical validation—reshaping how medications are researched
and brought to market.
For deeper insights in how AI is being used in pharmaceuticals and life scienc-
es, see the Bioengineering report.
HOW IS AI BEING USED IN PHARMA?
AlphaFold has now predicted the 3D shapes of almost all proteins in the human body, accomplishing in
just a few years what would have previously taken decades - or may have been impossible.

HOW IS AI BEING USED IN PHARMA?
TECH
Protein Folding
In 2020, DeepMind’s AI made a big announce-
ment: It had solved a 50-year grand challenge
with AlphaFold, an AI tool that predicts the
structure of proteins. AlphaFold outperformed
an estimated 100 teams in a biennial pro-
tein-structure prediction challenge called
Critical Assessment of Structure Prediction, a
problem that has long vexed biologists.
AlphaFold had previously bested other teams
but worked so quickly and so accurately that
it signaled a near future when the technology
could be used regularly by other scientists.
Along with the newest version of AlphaFold,
DeepMind published full details of the system
and released its source code. It also made a
stunning reveal: AlphaFold 2 has predicted
the shapes of nearly every protein in the hu-
man body, as well as hundreds of thousands
of other proteins found in 20 of the most
widely studied organisms, including yeast,
fruit flies, and mice. In a December 2023 up-
date, Isomorphic Labs and DeepMind released
an improved AlphaFold model that predicts
protein structures with greater accuracy and
These examples reflect only a sample of the
expansive AI drug discovery efforts under-
way across academia and industry. Major
pharmaceutical leaders such as Johnson
Johnson, Novartis, and AstraZeneca have
already forged partnerships with AI startups.
The allure lies in deep learning’s unmatched
speed and pattern recognition capabilities
for parsing volumes of data. While AI cannot
wholly replace lab science (yet), it signifi-
cantly accelerates prediction, design, and
validation to streamline timelines.
Generative Antibody Design
An antibody is simply a protein that protects
an organism. Produced by the immune sys-
tem, antibodies bind to unwanted substanc-
es and eliminate them. In 2023, researchers
from Absci Corp. showed how a generative AI
model was able to design multiple novel an-
tibodies that bind to a target receptor, HER2,
more tightly than previously known thera-
peutic antibodies. What’s interesting about
this work is that researchers first removed all
reference data on antibodies, so that the sys-
tem couldn’t just imitate and replicate the
structure of known antibodies that work well.
The designs produced by Absci’s system
were both diverse (meaning, they didn’t have
counterparts known to already exist) and they
received a high score on “naturalness,” so they
would be easy to develop and therefore cata-
lyze a strong immune response. Using genera-
tive AI to design novel antibodies that func-
tion at the same level––or even better––than
those designed by our own bodies marks a
bold new step in using AI to reduce the speed
and cost of therapeutic antibody development.
expands coverage to model interactions with
additional molecules like ligands. By en-
hancing AlphaFold’s capabilities, this latest
iteration provides scientists a more powerful
tool to rapidly examine proteins and molec-
ular interactions for advancing fundamental
biology research and applications.
AI-First Drug Development
The COVID-19 pandemic sparked a surge in
AI applications for expediting drug discov-
ery. An international research team demon-
strated this potential by crowdsourcing an
antiviral drug candidate in just 48 hours—a
process that traditionally takes months.
Separately, scientists at Ludwig-Maximil-
ians-Universität München developed an AI
model predicting where molecules can be
chemically altered. By reducing required
experiments, this enables more efficient,
sustainable synthesis. Another University
of Cambridge team created a platform that
automates experiments, then uses AI to
forecast chemical reactions. Until recently,
this was a trial-and-error process—which
means that it was slow and inefficient.

TECH
The health care industry suffers from
ballooning expenses and inadequate
human resourcing. As the COVID-19
pandemic spotlighted, doctor and
nurse shortages constrain delivery
capacity even in times of immense
need. AI could help make healthcare
cheaper, easier to access, and higher
quality by automating routine tasks.
AI has demonstrated the ability to an-
alyze certain types of test results as
accurately as physicians, and faster.
However, regulatory hurdles delay
rollout of this technology. Safety reg-
ulations developed for a human-cen-
tered system now hinder AI adoption.
Updating policies to allow ethical AI
use, while still protecting patients,
would facilitate major progress.
between doctors and patients. This clarifies
therapeutic options and care decisions. In
essence, AI boosts speed, accuracy, critical
care, self-care, and communication in health
care—all central to improving patient health.
AI-Assisted Diagnosis and Clinical
Decision-Making
People have long turned to search engines
to self-diagnose, but the emergence of AI
chatbots like ChatGPT and Bing introduces a
new era of medical consultation. LLMs have
already demonstrated the ability to accu-
rately provide potential diagnoses based on
symptom descriptions, achieving an 88% ac-
curacy rate in identifying the correct diagno-
sis among the top three choices, compared
to a 96% accuracy rate by physicians given
the same information. By processing natural
language descriptions, chatbots empower
more user-friendly symptom investigation
compared to rigid online symptom checkers.
Beyond advising patients, AI also increas-
ingly assists clinician decisions. FDA-ap-
proved systems already analyze imaging
scans to detect abnormalities, leveraging
data from billions of procedures. Algorithms
likewise forecast patient risk levels by assess-
ing extensive health records, outperforming
conventional clinical scores. As demonstrated
in a Beth Israel Deaconess Medical Center
study, an AI chatbot even surpassed physi-
cians in diagnostic accuracy for negative test
results—highlighting potential to close certain
cognition gaps. However, risks around reliance
on potentially misinforming training data
remain. If these can be addressed responsibly,
AI has immense capacity to streamline radiol-
ogy, reduce errors, aid predictions, and make
consultation and reasons more accessible.
Anomaly Detection in Medical Imaging
Anomaly detection uses AI to detect abnor-
malities in medical images, helping clinicians
identify issues faster. Machine learning algo-
rithms have the capability to sift through ex-
tensive medical data, including imaging and
pathology reports, significantly faster than
humans working alone. In radiology, AI’s abil-
ity to pinpoint anomalies in medical images
is exceptionally accurate. Such early detection
AI to Improve Patient Outcomes
AI can enable quicker, more accurate diag-
nosis and treatment, driving better patient
outcomes. This impact is clear in managing
critical conditions like sepsis. Saint Luke’s
Health System implemented an AI sepsis
detection system, cutting the time to antibi-
otic administration by 32%. It also reduced
sepsis deaths by 16%. Since sepsis accounts
for one in three hospital deaths nationwide,
early AI detection and treatment could save
many lives. For example, UCHealth’s AI tool
is estimated to save around 375 lives yearly,
and many more once it’s rolled out.
Beyond the hospital, AI also helps patients
better self-manage chronic diseases. Up to
70% make medication errors like incorrect
insulin doses. But AI tools quietly identify
these errors at home, nudging patients with
alerts to take their treatments properly. En-
suring adherence promotes better outcomes.
Additionally, poor communication frustrates
83% of patients. By enabling natural lan-
guage processing and speech recognition,
AI can facilitate more meaningful dialogues
HOW IS AI BEING USED IN HEALTH CARE?

TECH
significantly enhances patient outcomes for
conditions such as cancer, leading to reduced
mortality rates. For example, UC Davis Health
has implemented Viz.ai, utilizing AI to ana-
lyze CT scans and flag potential strokes. Even
though physicians still review all scans, the
AI rapidly identifies anomalies to prioritize
cases. Adoption of these AI tools is increasing,
as 2021 FDA approvals now allow integration
into standard workflows rather than just
augmentation. IDx-DR uses AI to diagnose
diabetic retinopathy from retinal scans, while
Caption Health captures cardiac ultrasounds
that nurses can interpret quickly with just a
few days of AI software training.
Concerned about AI replacing doctors? There
is already a critical shortage of physicians
in rural areas. While AI can’t take the place
of physicians, it can simplify their workload
especially when it comes to medical imag-
ing, which could help to decrease the rate of
burnout, and enable them to dedicate more
attention to patient care. Though human
review remains vital, these emerging auton-
omous systems prove the growing role AI
ware improve, more intricate movements
may be possible, granting patients liberty
and control not felt for years post-accident.
Still, much testing remains before these
cyborg-esque applications become main-
stream medicine.
Medical Deepfakes
Medical deepfakes are AI-manipulated
medical images and data. While the term
“deepfake” has negative associations, these
technologies also have valuable clinical
applications when used ethically. For exam-
ple, Korean researchers synthesized realistic
mammograms using StyleGAN2 to improve
breast cancer detection.
However, medical deepfakes could also be
used to unethically alter diagnostic images
by adding or removing medical conditions.
Cyber criminals are developing novel med-
ical deepfake attacks intended to bring
chaos to hospital systems and diagnostic
centers. Researchers at Ben-Gurion Universi-
ty and the Soroka University Medical Center
demonstrated that tumors could be added or
removed from CT images––and the deepfakes
were good enough that radiologists didn’t
realize they were altered. (See our Health Care
Medicine report.)
Fortunately, tools to prevent misuse are in de-
velopment. For instance, DeepMind created AI
watermarks to validate real medical images.
With ethical governance, medical deepfakes
could enable earlier disease detection and
protect patient privacy. However, safeguards
are crucial as these technologies advance to
maintain accuracy and trust.
Healthcare-Specific LLMs
ChatGPT release in 2022 triggered a surge
in interest in applying natural language
processing (NLP) to health care tasks like
diagnosis and treatment recommendations.
However, most existing language models
fail to capture the nuanced vocabulary and
semantics of medical language. Furthermore,
general purpose LLMs, trained on extensive
data sets from across the internet, may have
imbalanced weight distributions—potentially
overemphasizing content like Reddit posts
plays in surfacing hard-to-spot anomalies in
imaging.
AI-powered movement
Groundbreaking medical research uses brain
implants and artificial intelligence to give
paralyzed patients control over their bod-
ies again. In early research, a quadriplegic
patient can now move his arms and hands
simply by thinking about the action. This is
achieved through innovative neural bypass
surgery, pioneered by scientists at North-
well Health. Microchips are embedded in the
brain in the regions that control movement
and sensation. Sophisticated AI algorithms
then interface with the chips, interpreting
the patient’s thought patterns and translat-
ing desired actions into movement signals.
In a similar study, another patient regained
control over his lower body with a spinal cord
implant that bypasses injury sites. Termed a
“digital bridge,” an AI thought decoder reads
his brain signals related to intended mo-
tions and matches them to the appropriate
muscle activations. As algorithms and hard-

TECH
while underrepresenting reputable sources
such as medical publications. To address this
gap, researchers have developed domain-spe-
cific LLMs exclusively pretrained on large
medical corpora. For instance, BioBERT, which
is pretrained on PubMed articles, excels at
biomedical text processing tasks, while Clin-
icalBERT leverages clinical notes to enhance
its performance on health care–related NLP
tasks. BlueBERT merges the strengths of both
biomedical and clinical training, making it a
versatile model for a wide range of medical
text analysis applications. Similarly, MedNLI
focuses on clinical notes and natural lan-
guage inference, allowing for sophisticated
understanding and prediction in clinical con-
texts. Google recently unveiled Med-PaLM—
among the largest medical LLMs to date—
which proves highly accurate in answering
US Medical Licensing Examination questions
and consumer health queries. The family of
Med-PaLM models available through Google
Cloud enables a sweeping range of precision
health care applications.
AI for Mental Health
As mental health care systems struggle to
meet rising demand globally, artificial in-
telligence presents new opportunities to in-
crease access to support services. Intelligent
conversational agents like Replika that em-
ulate emotional support show promise for
addressing the student mental health crisis.
In one survey study of over 1,000 users, 3%
even reported Replika halted their suicidal
thinking. The social connection and therapy
services such bots provide may help fill gaps
for those awaiting treatment. Meanwhile,
University of Illinois Chicago researchers
piloted an AI voice assistant called Lumen
that delivers talk therapy content. The virtual
coach improved patient depression and anx-
iety, while brain scans revealed correspond-
ing neurological changes—demonstrating
legitimacy as a stopgap measure.
As mental health demands escalate globally,
AI virtual assistants and chatbots could aid
overwhelmed systems by offering readily ac-
cessible support. While not replacing human
therapists, they can screen patients, provide
psychoeducation, suggest coping strategies,
and monitor conditions between appoint-
ments with professionals.
In-Silico Trials
In-silico trials use computer simulations
rather than human subjects to test new
drugs and therapies. These digital trials,
powered by artificial intelligence, create
“digital twins” that mimic human biology
and disease. By running thousands of virtual
trials, researchers can quickly and affordably
predict how a drug might perform in human
patients. This has the potential to dramati-
cally accelerate and improve the drug devel-
opment process.
For example, a company called Novadiscov-
ery used AI to accurately forecast the results
of a Phase III clinical trial, showing the prom-
ise of this approach. In-silico trials may one
day replace up to half of human testing. Reg-
ulators are looking at how to include these
virtual results in the approval process. New
frameworks to validate in-silico trials will be
important to ensure reliability. By moderniz-
ing clinical trials with AI and simulations, we
can bring innovative treatments to patients
faster and more affordably.

TECH
After nearly 2,000 years, AI has finally unlocked the secrets inside ancient
scrolls flash-fried by Mount Vesuvius’ eruption in 79 AD. The Vesuvius Challenge,
launched in early 2023, aimed to develop an AI system capable of deciphering
these fossilized scrolls—known as the Herculaneum Papyri—rescued from an
ancient Roman library. Its success could save an invaluable trove of literature
and history from extinction.
In February, translated excerpts revealed one scroll’s author—likely the philos-
opher Philodemus—wrote about music, food and embracing life’s pleasures. He
rebukes opponents unable to appreciate enjoyment. This represents just 5% of
the text from one scroll, but demonstrates AI’s immense potential. Deciphering
these delicate, charred scrolls would have been impossible without AI. The proj-
ect illustrates how AI could optimize science by radically accelerating the pace
of innovation across fields. While the essence of the scientific method endures,
AI promises to transform each stage of discovery.
AI-Driven Hypotheses
AI is changing the way scientists ask ques-
tions and form hypotheses. With the help of
LLMs, knowledge graphs, and algorithmic
analysis, researchers can now tap into vast
databases of scientific literature, uncov-
er hidden connections, and propose novel
hypotheses that might have remained un-
discovered through conventional methods.
Tools like PaperQA and Elicit employ LLMs to
sift through extensive databases of scien-
tific articles, producing concise summaries
that include relevant citations. These AI-driv-
en summaries can serve as a foundation for
developing new hypotheses by highlighting
key findings, trends, and gaps in the current
body of knowledge. Furthermore, by ana-
lyzing existing literature and data, AI can
identify blind spots in research—areas that
have been overlooked or underexplored. Uni-
versity of Chicago researchers James Evans
and Jamshid Sourati showed this by using
knowledge graphs not only to map out con-
nections between materials, properties, and
researchers but also to find unconventional
pathways that could lead to new discoveries.
Their algorithms have successfully predicted
drug repurposing opportunities and novel
material properties that were later validated
by human research.
While AI has shown a propensity for generat-
ing specific, concrete hypotheses, interest is
rising in its ability to propose more abstract
and general theories. This involves not just
solving predefined problems but uncovering
fundamental principles that can guide future
research across various domains. A collabo-
rative approach described by the University of
Chicago’s Sendhil Mullainathan and Jens Lud-
wig in a paper posits AI and humans working
together to generate broad hypotheses from
complex data sets, illustrating the potential
for AI to contribute to a deeper understanding
of complex phenomena.
AI-Driven Experimentation
Beyond hypothesis, AI is also accelerating
scientific experimentation itself—both in
simulation and the real world. Researchers
at Caltech are exploring how they can use AI
models to conduct virtual experiments. The
HOW IS AI BEING USED IN SCIENCE?

TECH
team employed an AI fluid simulation model
to automatically design a better catheter that
prevents infections. For real-world experi-
mentation many researchers are turning to
“self-driving labs”—automated robotic plat-
forms infused with AI. For instance, Emerald
Cloud Lab is a research facility that handles
daily lab work without the researcher actually
having to set foot in the physical lab space.
Using AI, the lab can autonomously handle
everything from method design to instrument
operation to data acquisition and analysis. In
2023, a study published in Nature showcased
how a self-operating lab sped up the creation
of new materials. Within just 17 days of non-
stop work, this autonomous lab successfully
produced 41 new substances, targeting 58 dif-
ferent materials including various oxides and
phosphates. The high success rate shows the
promise of AI-powered platforms for autono-
mous experimentation, especially for auton-
omous materials discovery (see “AI to Speed
Up New Materials Development”).
AI and the Replication Crisis
The replication crisis in science refers to a
widespread problem where many scientif-
ic studies, particularly in psychology and
the social sciences, cannot be replicated or
reproduced by other researchers, casting
doubt on the reliability of their findings.
Many published studies fail to yield consis-
tent results when experiments are repeat-
ed. To assess research integrity efficiently
without costly manual replication, research-
ers developed an AI algorithm to predict a
study’s likelihood of successful reproduction
based on analysis of over 14,000 psychology
papers. By identifying factors that contrib-
ute to or detract from replicability, this tool
allows researchers, journals and funding
agencies to focus resources on the most
robust, reliable science. Moving forward, the
ability to estimate replication probability be-
fore peer review could guide adjustments to
improve study design as well as inform pol-
icy shaped by scientific evidence. If scaled
across disciplines, AI-enabled replication
forecasting presents a cost-effective solu-
tion to promoting greater rigor and reproduc-
ibility in the scientific process.
NLP Algorithms Detect Virus Mutations
Natural language processing (NLP) algo-
rithms, which are typically used for words and
sentences, are also being used to interpret
genetic changes in viruses. Protein sequences
and genetic codes can be modeled using NLP
techniques—and can be manipulated the way
you’d produce text in word processing soft-
ware. At MIT, computational biologists used
NLP to solve a vexing problem when develop-
ing new vaccines. “Viral escape” is the ability
for a virus to mutate and evade the human
immune system and cause infection. MIT
researchers modeled viral escape using NLP to
identify how the virus might look different to
the immune system. The approach is similar
to changing words in a sentence to change
its meaning. For example: “I laughed at the
clown” versus “I cried at the clown.” By using
this kind of modeling before mutations occur,
public health officials could strategize and
potentially prevent new viral spreads.
AI-Powered Analysis and Interpretation
AI also stands to change how and who does
the interpretation and analysis of scientific
data. As AI tools become more integrated
into research methodologies, they lower
entry barriers, enabling a diverse group of
new scientists, including those without
formal data science training, to contribute
meaningfully to scientific discourse. The
fear of criticism from established experts, a
significant deterrent for novice researchers,
is mitigated as AI provides guidance on best
practices and ensures the credibility of their
analyses. Moreover, as AI grows more adept
at understanding and generating insights
from multimodal data, including visualiza-
tions, it offers a more intuitive and acces-
sible way for independent researchers to
explore and contribute to various scientific
fields. This shift not only expands the pool
of researchers but also enriches scientific
inquiry with a wider range of perspectives
and ideas.

TECH
AI to Speed Up New Materials Development
Running experiments with several variables
often requires tiny, methodical tweaks to
measurements, materials, and inputs. Grad-
uate students might spend hundreds of
tedious hours repeatedly making small ad-
justments until they find a solution—a waste
of their cognitive abilities, and their time.
Unlike graduate students, AI doesn’t have
to sleep. For instance, Google DeepMind’s AI
program, GNoME, has significantly expanded
the database of stable materials, identifying
380,000 new potentially stable crystals from
a vast prediction of 2.2 million. This break-
through, published in Nature, demonstrates
AI’s capacity to enhance our understanding
of material stability and composition without
the constraints of human biases or limita-
tions. In a set of subsequent experiments
(aforementioned in AI-driven experimenta-
tion), an autonomous lab was able to create
41 of the theorized materials over 17 days. This
demonstrates the capabilities of both the AI
discovery model and the lab’s robotic tech-
niques.
Researchers are developing automated laboratory systems that use AI to independently handle processes from operating scientific instruments to performing re-
al-time data analysis.

TECH
AI has many uses in finance, like cus-
tomized services and fraud detection.
It can help forecast assets and market
trends. However, AI also poses finan-
cial risks. It could enable new types of
fraud and cybercrime. There are also
concerns about overreliance on a few
centralized AI systems for decision
making. If these systems make mis-
takes, it could spark a “polycrisis.”
Bad decisions could compound, turn-
ing small issues into major crises.
Mitigating Fraud
Financial institutions are increasingly utiliz-
ing AI to detect and reduce fraud. Advanced
machine learning models can identify suspi-
cious patterns in immense volumes of trans-
action data that humans alone may miss.
This allows companies to catch more fraud
attempts sooner. For example, several major
banks have invested heavily in developing
proprietary AI fraud prevention systems. By
continually monitoring for anomalies, these
algorithms have enabled substantial reduc-
tions in losses from fraudulent activities.
JP Morgan Chase invested $100 million into
developing sophisticated anti-fraud tech-
nologies for consumer payments, leading
to a notable 14% decrease in fraud incidents
between 2017 and 2021.
The Bank for International Settlements (BIS)
Innovation Hub’s Project Aurora has also
demonstrated the effectiveness of neural
networks, a branch of machine learning,
in combating money laundering. These
advanced systems excel in detecting irreg-
ular patterns and anomalies in financial
transactions that might elude traditional
detection methods, offering a more robust
defense against financial crimes. Similarly,
the Bank of Canada has developed a machine
learning-based tool designed to spot irregular-
ities in regulatory submissions. According to
Maryam Haghighi, the bank’s data science di-
rector, this tool conducts automatic daily anal-
yses that can uncover discrepancies human
inspectors might miss, thereby increasing
efficiency and allowing staff to allocate more
time to investigate these anomalies further.
Predicting Financial Risk
AI systems can help improve loan underwriting
and reduce financial risk. Models are being
trained to recognize anomalous activity and to
develop forecasts for a variety of middle—and
back-office applications. For example, US Bank
relies on deep learning to analyze customer
data as well as to root out money laundering
schemes. On a larger scale, the European Cen-
tral Bank (ECB) has integrated AI to advance
oversight across millions of businesses and
government entities. By automatically classify-
ing information, the technology helps identify
If we enter into a world where
all the banks are using this
major technology, are we
going to see supercharged
herding behavior? Are we
going to see AI bots that are
sentiment-driven and feed
off each other, and you then
end up with much bigger
amplitudes in the financial
cycle—so big credit booms
and busts. I’m not saying it’s
imminent, but this is something
we’re paying attention to.
—Gita Gopinath, International Monetary
Fund’s first deputy managing director
HOW IS AI BEING USED IN FINANCE?

TECH
HOW IS AI BEING USED IN FINANCE?
stability threats early by uncovering patterns.
The ECB also web scrapes pricing data for
real-time inflation analysis to stay ahead of
macro risk shifts. AI also aids ECB bank ex-
aminers; algorithms rapidly parse volumes of
filings to surface compliance issues or other
red flags.
Customized Portfolios
Socially conscious investing is entering
the mainstream as young investors assert
their consciences and wield new purchasing
power. As Gen Z starts working and financial
planning, demand will surge for customized
investment portfolios matching personal
values. This techie, purpose-driven genera-
tion wants their dollars supporting cherished
causes—two-thirds aim to back companies
upholding their principles around environ-
mental, social, and governance (ESG) con-
cerns. AI can help with this values-based
investing by enabling asset managers to
efficiently build highly customized portfolios
aligned with each client’s ethics. JPMorgan
Asset Wealth Management’s acquisition of
rors and amplifying mistakes. We have seen
how interconnected markets can lead to a
crisis when institutions mimic each other’s
actions without independent thought, as
in the 2008 housing crash. Some worry the
rise of cutting-edge generative AI could fuel
herd mentalities, if banks and funds utilize
the same basic signals and models from
one or two dominant providers. That could
potentially concentrate risk, create confor-
mity, and set the stage for panic and conta-
gion across the system. Furthermore, if the
leading models have flaws, or the data sets
themselves provide a distorted view, it could
lead institutions toward harmful decisions
en masse. So if an unprecedented shock
hits markets, AI could end up exacerbating
volatility and dysfunction. These opaque
algorithms can quickly turn negative loops
and contagion. This could be viewed as a
polycrisis —when multiple crashes converge,
the combined crisis proves more damaging
than isolated events.
OpenInvest allows investors to integrate their
personal values directly into their invest-
ment strategies. The platform’s generative
AI technology enables the customization of
a client’s entire portfolio, including external
assets, based on their specified values. On
the European front, Amundi, managing over 2
trillion euros in assets, leverages AI to tailor
investment portfolios for its vast clientele. By
gathering clients’ risk preferences, Amundi’s
AI tools can dynamically adjust portfolios,
offering a real-time reflection of investor
sentiment.
Growing Concern About Centralized
Data Sets
The growing reliance on centralized data
sources and AI models in finance raises con-
cerns about potential fragility in the system.
As a few large tech companies come to lead
the AI space, providing the models and data
that power financial decision-making, risks
emerge. Market participants could end up
drawing from the same narrow set of flawed
data or algorithms, modeling the same er-
AI systems are now used in finance for predictive
risk analytics, fraud detection, and regulatory over-
sight by rapidly surfacing patterns that may have
been invisible to human analysts.

TECH
HOW IS AI BEING USED IN INSURANCE?
Predicting Workplace Injuries
AI systems are being trained to detect possi-
ble workplace injuries. Using AI-based com-
puter vision models, Turkey-based Intenseye
can detect 40 types of employee health and
safety incidents in real time. The compa-
ny says that it does not capture personally
identifiable information from the visual data
it processes and that it detected 1.8 million
unsafe acts in 2020 and 2021. San Francisco
based Voxel uses computer vision to enable
security cameras to automatically detect
high-risk activities in real time. Caterpillar,
in collaboration with Seeing Machines, an
Australian company, has launched a technol-
ogy that detects driver fatigue through eye
and facial movement analysis. If the system
observes that a driver’s eyes remain closed for
more than 1.6 seconds, it initiates an alert in-
side the truck. Should the behavior persist, a
second alert notifies a supervisor, and a third
alert often leads to the driver being taken off
duty. Besides identifying fatigue, the technol-
ogy is adept at detecting instances of driver
distraction, contributing to a reduction in
fatigue-related incidents by as much as 90%.
Improving Damage Assessment
Insurance companies are applying AI to
assess damage and improve forecasts. The
Vehicle Damage Inspection model, which
is available on AWS Marketplace, uses a
machine learning model to determine what
part of a car is damaged. After photos are
uploaded, it assesses loss and dramatical-
ly reduces the amount of time required for
human appraisers to conduct their analysis.
Following catastrophic typhoons and weath-
er events in Japan, local insurance compa-
nies are relying on computer vision to assess
damage after a natural disaster. Sompo
Japan is using the Tractable AI Estimating
system to calculate the approximate repair
cost of damaged homes.
Consumer-Facing Robo-Advisers
Automated assistants are moving from
the fringe to the mainstream as consumer
adoption increases. Robo-advisers offer algo-
rithm-based portfolio management advice to
investors, applying parameters like risk toler-
ance and desired returns. These investment
tools offer some tangible benefits over their
traditional, human counterparts: they can pro-
vide more services at a lower cost, they’re able
to digest and interpret mounds of data in real
time, and they don’t take part of the weekend
off to golf. Wealthfront is an AI-powered system
for consumers: It suggests fund managers and
calculates probable risk levels based on the
user’s personal information and preferences.
AI Claims Processing
While human claims writers must painstak-
ingly review pictures and reports to assess
damage, compare what they see to coverage
policies, and make a determination about
appropriate actions, an AI system can digest
the same data and accomplish the same
work in a matter of minutes. Using a suite of
tools—natural language processing for policy
review, and computer vision recognition to
spot anomalies in photos and videos—claims
can be processed efficiently and, it’s believed,
more accurately. AI-powered claims process-
ing reduces the overhead for businesses and
wait times for customers. Some insurance
providers are wading into a new pool of op-
portunities. Liberty Mutual’s mobile app has
The Connected Worker
Insurers are pursuing a “connect and pro-
tect” approach to reduce risks by leveraging
advanced sensors and artificial intelligence.
New Internet of Things devices worn by
workers or installed in insured locations
can continuously gather safety-relevant
data. This massively expands visibility into
hazards before losses occur. For instance,
Honeywell provides smart hardhats with
fatigue sensors, heart rate monitors and
more to enhance worker safety. The resulting
streams of biological and environmental
data feed into AI safety dashboards. Man-
agers gain real-time insight on emerging
risks across worksites to guide preventa-
tive interventions. Worker wearables could
enable employers to monitor and safeguard
entire workflows. Yet, at the same time, this
intensive data gathering and monitoring
raises worries of overly intrusive Big Broth-
er–level surveillance. Companies that appear
to excessively pry may meet marketplace
resistance despite promised safety gains.

TECH
HOW IS AI BEING USED IN INSURANCE?
started to integrate ML for damage assess-
ment—it informs customers about their cov-
erage and next steps.
Liability Insurance for AI
Who’s to blame when machines behave
badly? When the machine learning system
in Uber’s self-driving car failed and killed an
Arizona pedestrian, the company was likely
not covered under traditional cyber insurance.
As businesses rush to build and implement
AI products and processes, they must plan for
emerging risks. For example, what happens if
machine learning makes a company vulnera-
ble to attackers who inject fake training data
into a system? What if a health care compa-
ny’s AI misinterprets data and neglects to
identify cancer in certain patients?
These problems could put a company at risk
of lawsuits, and new insurance models are
needed to address these issues. Underwriters
are starting to include AI under cyber insur-
ance plans, while specialty insurers such as
La Playa’s Science and Tech Insurance now
offer coverage for AI applications.
Insurance companies are deploying sensors in equipment and safety gear to predict injuries, hoping to preemptively eliminate hazards rather than just compensate
workplace harm after the fact.

CREATIVITY
AND
DESIGN

TECH
HOW ARE PEOPLE USING AI TO BE MORE CREATIVE?
New research shows AI demonstrates very high levels of creativity, scoring in
the top 1% on standard tests. Scientists at the University of Montana tested
ChatGPT using the Torrance Tests of Creative Thinking, which assess human
creativity skills like coming up with lots of new ideas. Shockingly, ChatGPT
beat out nearly all college students by scoring higher than 99% of people for
originality. It showed an extreme creative talent at inventing brand new con-
cepts nobody has thought of before. The AI also did well at producing large
volumes of ideas.
While such revelations might initially spark fears of being replaced, another
perspective is that this means AI could be a very creative collaborative part-
ner. For those already engaged in creative pursuits, AI can serve as an invalu-
able companion, augmenting their ability to generate innovative ideas and
solutions. Those who possess creative visions but lack the technical skills to
fully realize them can leverage AI as a tool to bridge that gap.
GAN-Assisted Creativity
Generative adversarial networks (GANs) are
unlocking new creative possibilities across
a range of artistic disciplines. DALL-E 3 and
other AI image generators are powered by a
combination of existing algorithms—fusing
the creativity of GANs and the text com-
prehension capabilities of transformers.
This enables intuitive image creation from
conversational prompts. Users can simply
describe desired images, realistic or ab-
stract, and the model will digitally paint cus-
tom photographic illustrations on demand.
With each new prompt, it remixes its broad
visual knowledge to translate text into novel
graphical forms. Sora and Pika, idea-to-video
platforms, do the same for videos.
Creative applications for these tools are
widespread across artforms and disciplines.
In graphic design, GAN-enabled features in
Adobe Photoshop automate tedious edit-
ing so designers can ideate faster. Fashion
GANs remix clothing and textile data sets
into refreshing one-of-a-kind garment and
fabric patterns. Architecture and interior
design GANs accelerate iteration by proposing
reimagined building layouts and conceptual
spaces. Rather than replacing imagination,
GANs serve as an endless springboard for
human creators—providing inspiration to
stretch creative boundaries in tandem with
this AI muse. Across disciplines, GANs liberate
designers to explore new frontiers.
Neural Rendering
Starting with a 2D image, researchers can
now create a rich 3D view of a scene by us-
ing a neural network to capture and generate
spatial imagery. Called neural rendering, the
process captures a photorealistic scene in 3D
by calculating the density and color of points
in space. The algorithm converts 2D pixels
into voxels, which are a 3D equivalent. The
result is a video which looks convincingly real.
The many applications for neural rendering
include amping up autonomous driving to
help train algorithms to recognize and react
to novel road situations. This technology will
influence the future of video games, virtual re-
ality, and emerging metaverse environments.

TECH
Generating Virtual Environments From
Short Videos
Nvidia has developed an AI system called
Neuralangelo that creates realistic 3D en-
vironments automatically from short video
clips. It uses AI algorithms called GANs and
has been trained on open-source self-driving
car data sets. Specifically, Neuralangelo takes
video segments categorized by objects like
buildings, trees, and vehicles, and uses them
to generate novel graphics. Using short clips
segmented into various categories—such as
buildings, sky, vehicles, signs, trees, or peo-
ple—the GANs created new, different versions
of these objects. The array of possible appli-
cations is vast. Automatically generated vir-
tual environments could be used for movies,
bringing down the costs of TV production. The
ability to procedurally generate realistic 3D
environments and assets could significantly
enhance video game development. It allows
for unique worlds and reduces modeling
costs. Architects and urban planners can use
the system to visualize and iterate on build-
ing and city designs more quickly. It super-
charges prototyping capabilities. The possible
like AI Hub are organically forming to offer
guidance and collaboration. With over 21,000
members, such groups allow music creators
to teach each other techniques, share artist
voice models, and troubleshoot projects as
participants collectively push boundaries on
what is achievable.
Underpinning these innovations is a com-
mon thread: AI democratizing music produc-
tion. Once the domain of recording studios
and audio engineers, creating professional
or personalized music is now available to
everyday creators through such technolo-
gies. Even imperfect raw recordings can be
revitalized, as Paul McCartney recently un-
veiled an AI-restored long-lost vocal track by
John Lennon that became the foundation for
a new Beatles song. From sonic preservation
to imaginative generation, AI empowers both
novices and experts to shape soundscapes
in previously unthinkable ways.
Automatic Ambient Noise Dubbing
For some time, we’ve been training comput-
ers to watch videos and predict correspond-
real-world applications are immense. For
example, the capability to easily produce 3D
worlds could significantly bring down costs
for CG in movies and TV production. Video
game developers also stand to benefit, as
they can use Neuralangelo to rapidly create
fresh 3D assets and environments for their
virtual worlds. This allows for unique styling
while reducing the need for extensive human
modeling. Furthermore, architects and urban
planners can utilize the system to quickly
visualize and iterate on building and city
designs at low cost.
AI Democratizes Music Production
A wave of AI voice and music startups has
emerged over the past year, aiming to revo-
lutionize audio editing and creation. Com-
panies like Descript and Voicemod now offer
tools that can manipulate speech—opening
possibilities like effortless podcast clean-
up or even mimicking celebrity voices. For
music, Google’s experimental Dream Track
lets users generate original songs in the
style of famous artists through simple text
prompts. As interest grows, communities
Companies like Nvidia have developed AI systems
capable of generating realistic 3D environments
from short video clips. This could lower the barrier
to entry to movie production and game design.

TECH
ing sounds in our physical world. For example,
what sound is generated when a wooden
drumstick taps a couch? A pile of leaves? A
glass windowpane? The focus of this re-
search, underway at MIT’s Computer Science
and Artificial Intelligence Laboratory, should
help systems understand how objects inter-
act with each other in the physical realm. This
could improve the soundscapes created for
AI-generated movies—but it might also help
us imagine soundscapes for both imaginary
worlds (Laconia, from The Expanse) and real
ones (Mars).
Generating Music From Text
MusicLM is an AI system created by Google
that can transform text descriptions into
high-quality musical compositions. For ex-
ample, it can turn a text prompt like “upbeat
pop song with piano” into an actual 24 kHz
audio clip matching that description. What
makes MusicLM special is its ability to accu-
rately capture the emotion and style details
described in text when generating music.
It also adapts hummed melodies into full
song arrangements. In May 2023, MusicLM
debuted via Google’s AI Test Kitchen as an
experimental demo. By May, it was publicly
accessible so anyone can create AI-generat-
ed music through text prompts or whistling.
Users can specify instruments and moods.
However, MusicLM has sparked debates
around copyright issues. Critics argue that
because MusicLM learns by analyzing large
sets of existing songs, it may illegally use
copyrighted material without artist permis-
sion when generating its music. Lawsuits
around AI music copyright are expected that
may impact systems like MusicLM.
AI music composition tools can now generate original melodies and harmonies from text prompts. Other
audio AI tools are can convincingly synthesize plausible sounds to match visuals without requiring re-
al-world recordings.

TECH
As AI becomes increasingly integrat-
ed into creative workflows, the in-
dustry faces pivotal questions about
intellectual property, the ethics of
AI-generated content, and the future
of human-AI collaboration in arts
and business. This dynamic interplay
between technology and creativi-
ty not only opens new avenues for
invention and expression but also
ignites debates on the legal and
ethical implications of AI’s role in the
creative process.
on this topic this year. Under new contract
terms, studios “cannot use AI to write scripts
or to edit scripts that have already been
written by a writer,” according to comedian
Adam Conover, who spoke on behalf of the
Writers Guild of American negotiating com-
mittee. The newest contract also prevents
studios from treating AI-generated content
as “source material,” like a novel or a stage
play, that screenwriters could be assigned to
adapt for a lower fee and less credit than a
fully original script.
New Business Models
A philosophical fork is emerging in how cre-
ators respond to AI. While some double down
on safeguarding their intellectual property,
others adopt an “if you can’t beat ’em, join
’em” ethos, choosing to embrace AI as a part-
ner instead of as a threat. Grimes sits firmly
in the latter camp, recently unveiling a plan
to share 50% of earnings from any AI-syn-
thesized songs that use her voice. The artist
positions herself at the forefront of this new
business approach, highlighting the idea
that AI can enhance production rather than
AI-Assisted Invention
Stable Diffusion, MidJourney, DALL-E3,
and ChatGPT-4 are now widely accessible
to end-consumers, leading to AI-assisted
human creativity. But these systems were
all trained using other artists’ works. If a
business uses an AI-generated image, video,
or text for commercial purposes, does it owe
anything to those whose original works were
used for training? Likewise, what if a gen-
erative AI system invents a product that’s
eligible for a patent?
In 2021, the South African government grant-
ed a patent to an AI system called Dabus,
which invented a method to interlock food
containers. It was a world-first: previously,
patents had only been awarded to humans.
In the US, the application was rejected,
with a judge citing case law stipulating
that only a human can hold a patent. There
may be business cases for an AI to hold a
patent rather than an individual. It raises
the question: What happens when AI sys-
tems co-invent, or even entirely invent, new
products? We’re likely to hear more debate
HOW IS AI DISRUPTING THE CREATIVE INDUSTRY?
Rather than tightly controlling their creative IP,
some artists are openly embracing AI to pioneer
new business models - training generative sys-
tems on their aesthetic so fans can discover or
even co-create derivative works, fostering engaged
communities and unlocking new profit streams in
the process.

TECH
replace it. She sees AI as a partner that can
free up human creativity instead of supplant-
ing human creativity. Avant-garde musician
Holly Herndon pioneered a similar fan part-
nership model back in 2021, enabling collec-
tive remixing of her signature sound under
prescribed conditions. Still, tensions churn
within creative circles around these digital-
ly-driven opportunities. Both views show seri-
ous efforts to understand huge changes and
figure out how to use them positively.
Legal Battles Between Writers and AI
As AI generative writing capabilities rap-
idly advance, friction is rising between the
technology and professional human writers.
This apprehension has been highlighted by
significant events such as the Hollywood
writers’ strike and a surge in lawsuits aimed
at protecting copyright interests. The strike
recently concluded with the Writers Guild of
America securing an agreement that intro-
duces measures to regulate AI’s role in the
creative process. Although the use of AI tools
has not been outright banned, the new con-
tract establishes safeguards ensuring that
AI technologies remain under the control of
human workers rather than being utilized by
employers as a substitute for human talent.
Parallel to the concerns in Hollywood, a no-
table lawsuit has been filed against OpenAI
by a collective of distinguished authors,
including John Grisham, Jonathan Franzen,
and Elin Hilderbrand, and spearheaded by
the Authors Guild. It accuses OpenAI of copy-
right infringement for allegedly training its
ChatGPT chatbot on copyrighted books with-
out authorization or compensation to the
authors. The plaintiffs argue that ChatGPT’s
ability to generate “derivative works” that
closely mimic and summarize their books
could detrimentally affect the market for
the original works. The case, filed in the US
District Court for the Southern District of
New York, highlights the tension between
the advancement of AI technology and the
protection of intellectual property rights.
HOW IS AI DISRUPTING THE CREATIVE INDUSTRY?
Writers worry increasingly capable AI narrative generation poses an existential threat. They fear that au-
tomated writing could make their skills redundant and jobs interchangeable. However, some wrtiers are
using AI as a tool to boost their own creatitivy and automate aspects of their workflow.

SCENARIOS
TECH

SCENARIOS
SCENARIO YEAR 2024
The Deepfake Mafia
AeroTech Innovations is a seemingly reputable company that boasts cutting-edge aerospace components that are sourced for commer-
cial airlines. This company, with its extensive online presence, sophisticated marketing campaigns, and convincing video testimonials
from high-profile business leaders, quickly gains the trust of major airlines searching for competitive edges in efficiency and safety.
With digital footprints of thousands of employees on LinkedIn, AeroTech appears to operate on a global scale. Its website features video
testimonials from well-known industry figures, praising the revolutionary impact of AeroTech’s products on their operations. The com-
pany’s adept use of digital platforms to showcase its expertise and the supposed reliability of its parts does not go unnoticed. It’s man-
aged to navigate the complex procurement processes of multinational airlines with ease, providing detailed digital 3D models of com-
ponents for review.
As AeroTech secures contracts, the company begins supplying airlines with parts promoted as state-of-the-art that are in fact sophisti-
cated 3D-printed components designed to fail. These parts are engineered to withstand initial testing but are programmed to degrade
after a specific number of flight hours, threatening catastrophic failures mid-flight.
The chilling reality is that AeroTech Innovations does not exist. It is the brainchild of a small group of four terrorists, leveraging advanced
deepfake technology and digital manipulation to create a facade of a global corporation. The LinkedIn profiles were all AI generated, the
video testimonials of real leaders were all deepfaked. AeroTech represents a new frontier of weaponized fakery; no longer just isolated
fakes of individuals but comprehensive illusions constructing an entire company from whole cloth.

SCENARIOS
SCENARIO YEAR 2027
TrailMate SLM
Morgan embarks on the ambitious journey to traverse the 2,190 miles of the Appalachian Trail, equipped with an REI de-
vice called TrailMate SLM, a compact AI gadget designed to serve as a natural language personal hiking assistant. The
small language model (SLM) embedded in the AI device covers topics like basic first aid and safety, cooking and food
handling techniques, and plant identification, making it an indispensable tool for any hiker.
Knowing he’ll be without a signal during parts of his hike, Morgan is grateful for the TrailMate SLM. The device’s abili-
ty to function offline ensures that, even in the absence of a signal, he will have a reliable source of guidance. For more
complex inquiries when in range of a signal, the device can connect to a larger, more comprehensive language model
through a subscription service.
Opting for cost and space efficiency, Morgan chose the basic TrailMate SLM over the premium version. The premium
model, while offering more detailed responses and a larger database, required a bulkier battery pack and sacrificed pre-
cious backpack space. This decision meant accepting a tradeoff in the level of detail available from the TrailMate SLM.
Despite this compromise, Morgan feels prepared, buoyed by years of backpacking experience. This journey is not just a
test of physical endurance but a leap of faith in the power of technology to augment human resilience and adaptability.

SCENARIOS
SCENARIO YEAR 2028
Centralized AI Belt and Road Infrastructure Crumbles
In Jakarta, Indonesia, a shocked finance minister hastily convenes an emergency meeting after volatile trading erased
nearly a third of the IDX Composite index value in just three days. Investigations reveal the startling catalyst—a subtle
data anomaly in AI-optimized stock recommendations from SinoTech, a Chinese tech company powering many Indone-
sian banks’ investment advisory services.
Upon discovery, revelation spreads that numerous Chinese AI providers across Southeast Asia share common LLM at a
state-owned entity. Realization dawns that dependence on these technologies has silently concentrated risk and wo-
ven tight coupling across ASEAN markets. Though no evidence shows coordinated attack, herd behavior amplified by
opaque Chinese predictive systems nearly collapsed interconnected regional exchanges.
Hard lessons are learned on the perils of external centralized data dependence as the unified ecosystem strategy that
propelled China’s AI success proves its Achilles heel. The crisis births calls for data transparency, decentralized collabo-
ration, and renewed focus on nurturing domestic capabilities to avoid future shocks. Indonesia spearheads the Digital
Sovereignty Initiative, providing subsidies for homegrown startups to counter reliance on imported AI tech.

SCENARIOS
SCENARIO YEAR 2030
Tabby the Tiger: Nurturing Curiosity Through AI Friendship
Tyler, a curious and imaginative 8-year-old, receives a special gift from his parents: a plush tiger named Tabby. Tabby
is embedded with an AI chatbot designed to be Tyler’s new friend under the innovative “friendship first” model of early
schooling. This model leverages the natural dynamics of friendship to foster learning and personal growth in children,
with the AI chatbot subtly guiding conversations to educational topics.
Tabby, with its vast repository of knowledge, gently brings up science and math topics, using stories, games, and ques-
tions to spark Tyler’s curiosity. For instance, when Tyler gets curious about why some toys are more expensive than oth-
ers, Tabby introduces the basics of supply and demand. This sparks an idea in Tyler’s mind, leading him to set up a lem-
onade stand in his front yard. With Tabby’s guidance, Tyler works out that he should raise or lower the price of lemonade
based on the weather. If it’s warm out, he can raise the price. If it’s raining, he should lower the price. This hands-on
activity not only entertains Tyler but also solidifies the economic principles of supply and demand in his young mind.
The beauty of this model is its subtlety; learning is not forced but emerges naturally from the bond they share. Tyler is
not just absorbing information; he is inspired to learn more, explore further, and dream bigger.
As months go by, Tyler’s parents notice a remarkable transformation in their son. Reflecting on this, Tyler’s parents can’t
help but draw parallels to their own childhood friendships that shaped their interests and careers. They realize that Tab-
by is not just a toy or a learning tool but a true friend who has opened a world of possibilities for Tyler.

SCENARIOS
SCENARIO YEAR 2040
What If “Thought-to-3D” Was an AI Modality?
It’s Monday morning and Maya settles in at her home office, excited to make progress on a new product design that
came to her in the shower. As founder of a startup creating sustainable kitchenware, inspiration strikes at odd hours,
often fading quickly. But now Maya simply puts on her Muse Cap linked to her Thought-to-3D AI system and mentally fo-
cuses on visualizing her idea—an ergonomic spatula with a unique twisted handle for comfort and control while cooking.
As Maya concentrates, the Muse Cap’s brain activity sensors—basically a mini FMRI machine—capture her visualization
data and feed it into the generative AI application. Within minutes, a 3D model of the spatula takes shape on screen,
automatically matched to Maya’s thoughts. She inspects it from all angles, edits a few details by voice command, then
hits print. The 3D printer at her downtown office soon produces an initial tangible prototype that Maya can pick up later
after dropping off the kids from school. She plans to test it while she cooks dinner that evening. If it works, she’ll send it
out tomorrow for manufacturing.
With the Muse Cap, this morning’s shower thought could be tomorrow’s revenue stream.

AUTHORS
CONTRIBUTORS
TECH

AMY WEBB
Recognized as the global leader in strategic fore-
sight, Amy Webb advises business leaders through
disruptive change, enabling them to navigate an
unpredictable future with confidence and take
actions that address global challenges, create
sustainable value, and ensure a company’s long-term growth. As founder and CEO of
the Future Today Institute, Amy pioneered a unique quantitative modeling approach
and data-driven foresight methodology that identifies signals of change and emerg-
ing patterns very early. Using that information, Amy and her colleagues identify white
spaces, opportunities, and threats early enough for action. They develop predictive
scenarios, along with executable strategy, for their global client base. In 2023, Amy
was recognized as the #4 most influential management thinker in the world by
Thinkers50, a biannual ranking of global business thinkers. She was also featured on
the 2021 Thinkers 50 list, was shortlisted for the 2021 Digital Thinking Award, and re-
ceived the 2017 Thinkers50 Radar Award. Forbes called Amy “one of the five women
changing the world,” and she was honored as one of the BBC’s 100 Women of 2020.
Amy also serves as a professor of strategic foresight at New York University’s Stern
School of Business, where she developed and teaches the MBA-level strategic foresight
course with live case studies. She is a Visiting Fellow at Oxford University’s Säid School
of Business. She was elected a life member of the Council on Foreign Relations and is a
member of the Bretton Woods Committee. She is a Steward and Steering Committee
Member for the World Economic Forum, a founding member of the Forum’s Strategic
Foresight Council, a member of the Forum’s Risk Advisory Council, and serves on the
Forum’s Global Futures Council. She was a Delegate on the former U.S.-Russia Bilateral
Presidential Commission, representing US interests in technology.
Regarded as one of the most important voices on the futures of technology (with spe-
cializations in both AI and synthetic biology), Amy is the author of four books, including
the international bestseller The Big Nine and her most recent, The Genesis Machine,
which was listed as one of the best nonfiction books of 2022 by The New Yorker. To
date, her books have been translated into 19 languages. A widely published and quoted
thought leader, Amy regularly appears in a wide range of publications and broadcasts.
AUTHORS CONTRIBUTORS
TECH

TECH
SAM JORDAN
Manager
Sam Jordan is a Manager at Future Today Institute. She leads our Advanced Comput-
ing practice area, which includes technology, artificial intelligence, virtual realities,
networking, telecommunications, and space. She is a distinguished practice area
lead, where she enables organizations to navigate through uncertainty with inno-
vative strategies. With a proven track record across various sectors, Sam’s visionary
leadership has driven growth and resilience for Future Today Institute’s global clients and partners.
Before joining FTI, Sam was the CEO and co-founder of TrovBase, a secure data discovery and analysis-sharing plat-
form. Sam grew the company from idea to launch and executed the company’s transition from scientific replication to
its current focus. In parallel, Sam engaged with the open science community, advocating for better data management
practices to address challenges in scientific replication. Previously, she worked for IBM, where she helped large enter-
prises in the retail and distribution sector modernize their IT stack. Her expertise centered around mainframes, assisting
with the integration of new software and modern methodologies to legacy systems.
Sam is a coach in the strategic foresight MBA course at the NYU Stern School of Business. She holds a BS in Economics
and Data Analysis from George Mason University and an MBA from New York University’s Stern School of Business.
Managing Director
MELANIE SUBIN
Creative Director
EMILY CAUFIELD
Editor
ERICA PETERSON
Copy Editor
SARAH JOHNSON
Director of Operations
CHERYL COONEY

SELECTED
SOURCES

Acosta, Julián N., Guido J. Fal-
cone, Pranav Rajpurkar, and Eric
J. Topol. “Multimodal Biomedi-
cal AI.” Nature Medicine 28, no.
9 (September 2022): 1773–84.
https://doi.org/10.1038/s41591-
022-01981-2.
Aftab, Aamir. “Generative Adver-
sarial Networks (GANs): A Deep
Dive into AI Creativity.” Medium,
December 21, 2023. medium.
com/@aamiraftabcloud/
generative-adversarial-net-
works-gans-a-deep-dive-in-
to-ai-creativity-71d7b351f3c1.
Agostinelli, Andrea, et al. “Mu-
sicLM: Generating Music from
Text.” ArXiv:2301.11325 [Cs, Eess],
January 26, 2023. arxiv.org/
abs/2301.11325.
AHA Center for Health Innovation
Market Scan. “How AI Is Improv-
ing Diagnostics, Decision-Mak-
ing and Care. AHA. www.aha.org/
aha-center-health-innovation-
market-scan/2023-05-09-how-
ai-improving-diagnostics-deci-
sion-making-and-care.
Armstrong, Adam. “Dell Power-
Scale Updates, New Partnerships
Point to AI Stack.” TechTar-
get, December 7, 2023. www.
techtarget.com/searchstorage/
news/366562300/Dell-Power-
Scale-updates-new-partner-
ships-point-to-AI-stack.
“Artificial Intelligence Act:
Deal on Comprehensive Rules
for Trustworthy AI.” Europe-
an Parliament, December 9,
2023. https://www.europarl.
europa.eu/news/en/press-
room/20231206IPR15699/
artificial-intelli-
gence-act-deal-on-comprehen-
sive-rules-for-trustworthy-ai.
Betker, James, Gabriel Goh, Li
Jing, et al. “Improving Image
Generation with Better Cap-
tions,” n.d.
Blattmann, Andreas, Tim Dock-
horn, Sumith Kulal, et al. “Stable
Video Diffusion: Scaling Latent
Video Diffusion Models to Large
Data Sets,” n.d.
Bousquette, Isabelle. “The AI
Boom Is Here. The Cloud May Not
Be Ready.” Wall Street Journal,
July 10, 2023. www.wsj.com/
articles/the-ai-boom-is-here-
the-cloud-may-not-be-ready-
1a51724d.
Chatterjee, Mohar. “White
House Offers a New Strategy
for AI— and Picks New Fights.”
Politico, October 30, 2023. www.
politico.com/news/2023/10/30/
biden-ai-tech-industry-poli-
cy-00124185.
Chen, Ziqi, et al. “G2Retro as
a Two-Step Graph Generative
Models for Retrosynthesis
Prediction.” Communications
Chemistry, vol. 6, no. 1 (May 30,
2023): pp. 1–19, www.nature.com/
articles/s42004-023-00897-3.
“Chexia Face Recognition.”
NIST. Accessed January 25,
2023. https://www.nist.gov/pro-
grams-projects/chexia-face-rec-
ognition.
Chitty-Venkata, Krishna Teja,
and Arun K. Somani. “Neural
Architecture Search Survey: A
Hardware Perspective.” ACM
Computing Surveys 55, no. 4
(April 30, 2023): 1–36. https://
doi.org/10.1145/3524500.
Cowen, Tyler. “AI’s Greatest
Danger? The Humans Who Use
It.” Bloomberg.com, January 25,
2024. www.bloomberg.com/
opinion/articles/2024-01-25/ai-
s-greatest-danger-the-humans-
who-use-it.
Cowen, Tyler. “Your Child’s Favor-
ite Teacher May Soon Be a Chat-
bot.” Bloomberg.com, January
17, 2024. www.bloomberg.com/
opinion/articles/2024-01-17/
ai-in-the-classroom-everyone-s-
favorite-teacher-may-soon-be-a-
chatbot.
“Dawn of the EU’s AI Act: Political
Agreement Reached on World’s
First Comprehensive Horizon-
tal AI Regulation.” White
Case LLP, December 14, 2023.
https://www.whitecase.com/
insight-alert/dawn-eus-ai-act-
political-agreement-reached-
worlds-first-comprehensive-hor-
izontal-ai.
“Decomposing Language Models
Into Understandable Compo-
nents.” Accessed February 21,
2024. https://www.anthropic.
com/news/decomposing-lan-
guage-models-into-understand-
able-components.
DeepMind. “Mastering Atari, Go,
Chess and Shogi by Planning
with a Learned Model.” https://
deepmind.com/research/publi-
cations/2019/Mastering-Atari-
Go-Chess-and-Shogi-by-Plan-
ning-with-a-Learned-Model.
DeepMind. “MuZero: Mastering
Go, chess, shogi and Atari With-
out rules.” https://deepmind.
com/blog/article/muzero-mas-
tering-go-chess-shogi-and-at-
ari-without-rules.
DeepMind. “Putting the Power
of AlphaFold into the World’s
Hands.” https://deepmind.com/
blog/article/putting-the-power-
of-alphafold-into-the-worlds-
hands.
Dong, Tian, Bo Zhao, and
Lingjuan Lyu. “Privacy for Free:
How Does Dataset Conden-
sation Help Privacy?” ArXiv,
June 1, 2022. http://arxiv.org/
abs/2206.00240.
Driess, Danny, Fei Xia, Mehdi
S. M. Sajjadi, et al. “PaLM-E: An
Embodied Multimodal Language
Model.” ArXiv, March 6, 2023.
https://doi.org/10.48550/arX-
iv.2303.03378.
Eger, Steffen, Christoph Leiter,
Jonas Belouadi, et al. “NLLG
Quarterly ArXiv Report 06/23:
What Are the Most Influential
Current AI Papers?” ArXiv, July 31,
2023. https://doi.org/10.48550/
arXiv.2308.04889.
SELECTED SOURCES

“ESM.” Accessed January 25,
2023. https://huggingface.co/
docs/transformers/model_doc/
esm.
“EU Artificial Intelligence Act
| Up-to-Date Developments
and Analyses of the EU AI Act.”
Accessed February 21, 2024.
https://artificialintelligenceact.
eu/.
Fedus, William, Barret Zoph,
and Noam Shazeer. “Switch
Transformers: Scaling to Trillion
Parameter Models with Sim-
ple and Efficient Sparsity.”
arXiv:2101.03961 [Cs], Janu-
ary 11, 2021. http://arxiv.org/
abs/2101.03961.
Gandhi, Sanchit, Patrick von
Platen, and Alexander M. Rush.
“Distil-Whisper: Robust Knowl-
edge Distillation via Large-
Scale Pseudo Labelling.” ArXiv,
November 1, 2023. https://doi.
org/10.48550/arXiv.2311.00430.
Gibney, Elizabeth. “What the
EU’s Tough AI Law Means for
Research and ChatGPT.” Nature,
February 16, 2024. https://doi.
org/10.1038/d41586-024-00497-
8.
GitHub. “Ablation Study: Why
ControlNets Use Deep Encoder?
What If It Was Lighter? Or Even
an MLP?.” Accessed February
21, 2024. https://github.com/
lllyasviel/ControlNet/discus-
sions/188.
Google DeepMind. “FunSearch:
Making New Discoveries in
Mathematical Sciences Us-
ing Large Language Models.”
December 14, 2023. https://
deepmind.google/discover/blog/
funsearch-making-new-discov-
eries-in-mathematical-scienc-
es-using-large-language-mod-
els/.
Google Research. “Med-PaLM.”
Google Research Med-PaLM,
2023. sites.research.google/
med-palm/.
Gunasekar, Suriya, Yi Zhang,
Jyoti Aneja, et al. “Textbooks Are
All You Need.” ArXiv, October 2,
arXiv.2306.11644.
Hafner, Danijar, Jurgis Pasu-
konis, Jimmy Ba, and Timothy
Lillicrap. “Mastering Diverse
Domains through World Mod-
els.” ArXiv, January 10, 2023.
iv.2301.04104.
Hassan, Oz. “Artificial Intelli-
gence, Neom and Saudi Ara-
bia’s Economic Diversification
from Oil and Gas.” The Political
Quarterly, vol. 91, no. 1 (January
2020): pp. 222–227. https://doi.
org/10.1111/1467-923x.12794.
Helwan, Abdulkader. “A List of
the Available Medical Large
Language Models: Med-LLMs.”
Medium, November 27, 2023.
abdulkaderhelwan.medium.
com/a-list-of-the-available-
medical-large-language-models-
med-llms-f087119fa89d.
House, The White. “Executive
Order on the Safe, Secure,
and Trustworthy Develop-
ment and Use of Artificial
Intelligence.” October 30, 2023.
https://www.whitehouse.
gov/briefing-room/presiden-
tial-actions/2023/10/30/
executive-order-on-the-safe-se-
cure-and-trustworthy-develop-
ment-and-use-of-artificial-intel-
ligence/.
Howard, Lisa. “New AI Technol-
ogy Helps Physicians Quickly
Identify Stroke.” UC Davis
Health, February 1, 2024. health.
ucdavis.edu/news/headlines/
new-ai-technology-helps-phy-
sicians-quickly-identi-
fy-stroke/2024/02.
Hutson, Matthew. “Hypotheses
Devised by AI Could Find “Blind
Spots” in Research.” Nature,
November 17, 2023, www.nature.
com/articles/d41586-023-
03596-0.
IBM. “What Is a Vector Data-
base?.” www.ibm.com/topics/
vector-database.
IBM Education. “The Benefits of
AI in Healthcare.” IBM Blog, July
11, 2023. www.ibm.com/blog/the-
benefits-of-ai-in-healthcare/.
IBM Research. “Neuro-Symbolic
AI.” research.ibm.com/topics/
neuro-symbolic-ai.
“Join Us in the AI Test Kitchen.”
Google blog. https://blog.google/
technology/ai/join-us-in-the-ai-
test-kitchen/.
“Joon Sung Park | Generative
Agents: Interactive Simulacra
of Human Behavior.” YouTube.
https://www.youtube.com/
watch?v=nKCJ3BMUy1s.
Kannampallil, Thomas, et al.
“Effects of a Virtual Voice-Based
Coach Delivering Problem-Solv-
ing Treatment on Emotional Dis-
tress and Brain Function: A Pilot
RCT in Depression and Anxiety.”
Translational Psychiatry, vol. 13,
no. 1 (May 12, 2023): pp. 1–8, www.
nature.com/articles/s41398-
023-02462-x.
Kearns, Jeff. “AI’s Reverberations
across Finance.” IMF, December
2023. www.imf.org/en/Publica-
tions/fandd/issues/2023/12/
AI-reverberations-across-fi-
nance-Kearns.
King-Smith, Emma, et al.
“Predictive Minisci Late Stage
Functionalization with Transfer
Learning.” Nature Communica-
tions, vol. 15, no. 1 (January 15,
2024): p. 426. www.nature.com/
articles/s41467-023-42145-1.
King-Smith, Emma, et al.
“Probing the Chemical ‘Reac-
tome’ with High-Throughput
Experimentation Data.” Nature
Chemistry, January 2, 2024: pp.
1–11. www.nature.com/articles/
s41557-023-01393-w.
SELECTED SOURCES

Kirillov, Alexander, Eric Mintun,
Nikhila Ravi, et al. “Segment
Anything.” ArXiv, April 5, 2023.
iv.2304.02643.
Koc, Vincent. “Navigating
the AI Landscape of 2024:
Trends, Predictions, and Pos-
sibilities.” Medium, January
4, 2024. towardsdatascience.
com/navigating-the-ai-land-
scape-of-2024-trends-pre-
dictions-and-possibili-
ties-41e0ac83d68f.
Koponen, Jonna, et al. “Work
Characteristics Needed by
Middle Managers When Leading
AI-Integrated Service Teams.”
Journal of Service Research,
December 13, 2023. https://doi.
org/10.1177/10946705231220462.
Liu, Xian, Xiaohang Zhan,
Jiaxiang Tang, et al. “HumanGa-
ussian: Text-Driven 3D Human
Generation with Gaussian Splat-
ting,” n.d.
Llama. “Llama 2.” https://llama.
meta.com/llama2/.
“LLaVA.” Accessed February 21,
2024. https://llava-vl.github.io/.
Lorach, Henri, et al. “Walking
Naturally after Spinal Cord Injury
Using a Brain–Spine Interface.”
Nature, vol. 618 (May 24, 2023):
pp. 1–8. www.nature.com/arti-
cles/s41586-023-06094-5.
Lucente, Adam. “Will US Pressure
on UAE-China Tech Cooperation
Pay Off?.” Al Monitor, December
11, 2023. www.al-monitor.com/
originals/2023/12/will-us-pres-
sure-uae-china-tech-coopera-
tion-pay.
Maples, Bethanie, et al. “Loneli-
ness and Suicide Mitigation for
Students Using GPT3-Enabled
Chatbots.” NPJ Mental Health Re-
search, vol. 3, no. 1 ( January 22,
2024): pp. 1–6. www.nature.com/
articles/s44184-023-00047-6.
“Mastering Diverse Domains
through World Models.” Ac-
cessed February 21, 2024.
https://danijar.com/project/
dreamerv3/.
Mehrotra, Sukriti. “DALL-E 3: A
Fusion of Imagination and Con-
versation.” Medium, September
21, 2023. medium.com/@sukri-
timehrotra/dall-e-3-a-fusion-of-
imagination-and-conversation-
4c8ec8930442.
Merchant, Amil, et al. “Scaling
Deep Learning for Materials Dis-
covery.” Nature, November 29,
2023: pp. 1–6. www.nature.com/
articles/s41586-023-06735-9.
Microsoft News Center. “G42 and
Microsoft Unlock New Oppor-
tunities for Digital Transforma-
tion with Joint Sovereign Cloud
and AI Offering.” September 5,
2023, news.microsoft.com/en-
xm/2023/09/05/g42-and-mi-
crosoft-unlock-new-oppor-
tunities-for-digital-trans-
formation-with-joint-sover-
eign-cloud-and-ai-offering/.
Mittal, Aayush. “The Role of Vec-
tor Databases in Modern Gen-
erative AI Applications.” Unite.
AI, October 11, 2023. www.unite.
ai/the-role-of-vector-databas-
es-in-modern-generative-ai-ap-
plications/.
Mulligan, Deirdre K. and Mina
Hsiang. “A Call to Service for
AI Talent in the Federal Gov-
ernment.” The White House,
January 29, 2024. www.
whitehouse.gov/ostp/news-
updates/2024/01/29/a-call-
to-service-for-ai-talent-in-the-
federal-government/.
Myers, Andrew. “AI’s Powers of
Political Persuasion.” Stanford
HAI, February 27, 2023. hai.stan-
ford.edu/news/ais-powers-polit-
ical-persuasion.
Nellis, Stephen. “Intel to Spin out
AI Software Firm with Outside
Investment.” Yahoo Finance,
January 3, 2024, finance.yahoo.
com/news/intel-spins-ai-soft-
ware-firm-133626026.html.
Nippa, David F., et al. “Enabling
Late-Stage Drug Diversification
by High-Throughput Experi-
mentation with Geometric Deep
Learning.” Nature Chemistry, vol.
16 (November 23, 2023). www.
nature.com/articles/s41557-023-
01360-5.
Nour, Matthew M, et al. “Trajecto-
ries through Semantic Spaces in
Schizophrenia and the Relation-
ship to Ripple Bursts.” Proceed-
ings of the National Academy of
Sciences of the United States
of America, vol. 120, no. 42
(October 10, 2023). https://doi.
org/10.1073/pnas.2305290120.
Pan, Xingang, Ayush Tewari,
Thomas Leimkühler, et al. “Drag
Your GAN: Interactive Point-
Based Manipulation on the
Generative Image Manifold.”
ArXiv, May 18, 2023. https://doi.
org/10.48550/arXiv.2305.10973.
Park, Joon Sung, Joseph C.
O’Brien, Carrie J. Cai, et al. “Gen-
erative Agents: Interactive Simu-
lacra of Human Behavior.” ArXiv,
August 5, 2023. https://doi.
org/10.48550/arXiv.2304.03442.
Peplow, Mark. “Google AI and
Robots Join Forces to Build New
Materials.” Nature, November 9,
2023. www.nature.com/articles/
d41586-023-03745-5.
Qureshi, Nabeel S. “Moore’s
Law for Intelligence.” Digi-
tal Spirits, February 5, 2024.
digitalspirits.substack.com/p/
moores-law-for-intelligence.
Ramesh, Aditya, Prafulla Dhari-
wal, Alex Nichol, Casey Chu,
and Mark Chen. “Hierarchical
Text-Conditional Image Gener-
ation with CLIP Latents.” ArXiv,
April 12, 2022. http://arxiv.org/
abs/2204.06125.
“Responsible AI Guidelines.”
https://www.diu.mil/responsi-
ble-ai-guidelines.
SELECTED SOURCES

Rodman, Adam, et al. “Artificial
Intelligence vs Clinician Perfor-
mance in Estimating Probabili-
ties of Diagnoses Before and Af-
ter Testing.” JAMA Network Open,
vol. 6, no. 12 (December 11, 2023):
p. e2347075. jamanetwork.com/
journals/jamanetworkopen/ful-
larticle/2812737.
Roser, Max, et al. “What Is
Moore’s Law?” Our World in Data,
March 28, 2023. ourworldindata.
org/moores-law.
Runway. “Gen-1 by Runway.”
https://research.runwayml.com/
gen1.
Ryan-Mosley, Tate. “AI Isn’t
Great at Decoding Human
Emotions. So Why Are Regula-
tors Targeting the Tech?” MIT
Technology Review, August 14,
2023. www.technologyreview.
com/2023/08/14/1077788/ai-de-
coding-human-emotions-tar-
get-for-regulators/.
Savage, Rashaud. “Welcoming
Mistral, Phi, Jais, Code Llama,
NVIDIA Nemotron, and More to
the Azure AI Model Catalog.”
Microsoft, November 15, 2023,
techcommunity.microsoft.com/
t5/ai-machine-learning-blog/
welcoming-mistral-phi-jais-
code-llama-nvidia-nemotron-
and-more/ba-p/3982699.
Savcisens, Germans, et al. “Us-
ing Sequences of Life-Events to
Predict Human Lives.” Nature
Computational Science, vol. 4, 18
(December 2023): pp. 1–14, www.
nature.com/articles/s43588-
023-00573-5.
Schmidt, Eric. “Eric Schmidt:
This Is How AI Will Transform
the Way Science Gets Done.”
MIT Technology Review, July 5,
com/2023/07/05/1075865/
eric-schmidt-ai-will-transform-
science/.
“Segment Anything | Meta AI.”
https://segment-anything.com/.
Sheth, Amit, et al. “Neurosym-
bolic Artificial Intelligence (Why,
What, and How).” IEEE Journals
Magazine vol. 38, no. 3 (May
2023): pp. 56–62, ieeexplore.ieee.
org/document/10148662.
Shwartz-Ziv, Ravid, and Ami-
tai Armon. “Tabular Data: Deep
Learning Is Not All You Need.”
ArXiv, November 23, 2021. http://
arxiv.org/abs/2106.03253.
Simbo.ai—Blog. “How AI Tech-
nologies Improve the Patient
Experience.” September 23,
2021. www.simbo.ai/blog/index.
php/2021/09/23/how-ai-tech-
nologies-improve-the-pa-
tient-experience/.
Stanford University. “Artificial
Intelligence Index Report 2023
Introduction to the AI Index
Report 2023.” 2023.
Technical University of Den-
mark. “Artificial Intelligence
Can Predict Events in People’s
Lives.” ScienceDaily, December
18, 2023. www.sciencedaily.com/
releases/2023/12/231218125850.
htm.
The Alan Turing Institute.
“Neruo-Symbolic AI.” www.turing.
ac.uk/research/interest-groups/
neuro-symbolic-ai.
The Isomorphic Labs team
and Google DeepMind Alpha-
Fold team. “A Glimpse of the
Next Generation of AlphaFold.”
Isomorphic Labs, October 31,
2023. www.isomorphiclabs.com/
articles/a-glimpse-of-the-next-
generation-of-alphafold.
The University of Montana.
“AI Tests into Top 1% for Orig-
inal Creative Thinking.” Sci-
enceDaily, July 5, 2023, www.
sciencedaily.com/releas-
es/2023/07/230705154051.htm.
Topics, European Parliament. “EU
AI Act: First Regulation on Artifi-
cial Intelligence,” June 8, 2023.
https://www.europarl.europa.eu/
topics/en/article/20230601S-
TO93804/eu-ai-act-first-regula-
tion-on-artificial-intelligence.
Trinh, Trieu, and Thang Luong.
“AlphaGeometry: An Olympi-
ad-Level AI System for Geome-
try.” Google DeepMind, January
17, 2024. deepmind.google/
discover/blog/alphageome-
try-an-olympiad-level-ai-sys-
tem-for-geometry/.
Trinh, Trieu H, et al. “Solving
Olympiad Geometry without Hu-
man Demonstrations.” Nature,
vol. 625, no. 7995, (January 17,
2024): pp. 476–482. https://doi.
org/10.1038/s41586-023-06747-5.
Tuli, Shreshth, Giuliano Casale,
and Nicholas R. Jennings. “Tra-
nAD: Deep Transformer Networks
for Anomaly Detection in Multi-
variate Time Series Data.” ArXiv,
May 14, 2022. http://arxiv.org/
abs/2201.07284.
Vesuvius Challenge. “Vesuvius
Challenge 2023 Grand Prize
Awarded: We Can Read the
Scrolls!” Scrollprize.org, Feb-
ruary 5, 2024, scrollprize.org/
grandprize.
Vora, Lalitkumar K, et al. “Ar-
tificial Intelligence in Phar-
maceutical Technology and
Drug Delivery Design.” Phar-
maceutics, vol. 15, no. 7 (July
10, 2023): pp. 1916–1916https://
doi.org/10.3390/pharmaceu-
tics15071916.
“Voxel.” https://www.voxelai.
com/.
Wang, Chengyi, Sanyuan Chen,
Yu Wu, et al. “Neural Codec Lan-
guage Models Are Zero-Shot Text
to Speech Synthesizers.” ArXiv,
January 5, 2023. https://doi.
org/10.48550/arXiv.2301.02111.
SELECTED SOURCES

Wornow, Michael, et al. “The
Shaky Foundations of Foun-
dation Models in Healthcare.”
Stanford University, February
27, 2023. hai.stanford.edu/news/
shaky-foundations-founda-
tion-models-healthcare.
Wu, Qingyun, Gagan Bansal,
Jieyu Zhang, et al. “AutoGen:
Enabling Next-Gen LLM Applica-
tions via Multi-Agent Conversa-
tion.” Microsoft, August 16, 2023.
https://www.microsoft.com/
en-us/research/publication/
autogen-enabling-next-gen-llm-
applications-via-multi-agent-
conversation-framework/.
Yadlowsky, Steve, et al. “Pretrain-
ing Data Mixtures Enable Narrow
Model Selection Capabilities
in Transformer Models.” ArXiv
(Cornell University), November 1,
arxiv.2311.00871.
Yim, Kristin, and Hema Man-
ickavasagam. “Turn Ideas into
Music with MusicLM.” Google,
May 10, 2023. blog.google/
technology/ai/musiclm-goo-
gle-ai-test-kitchen/.
Youyou, Wu, et al. “How AI Can
Help Researchers Navigate the
“Replication Crisis.“” Kellogg
Insight, July 1, 2023. insight.
kellogg.northwestern.edu/arti-
cle/how-ai-can-help-research-
ers-navigate-the-replication-cri-
sis.
Zewe, Adam. “Technique Enables
AI on Edge Devices to Keep
Learning over Time.” MIT News,
November 16, 2023. news.mit.
edu/2023/technique-enables-
ai-edge-devices-keep-learning-
over-time.
Zhang, Can, Tianyu Yang, Junwu
Weng, et al. “Unsupervised
Pre-Training for Temporal Action
Localization Tasks.” ArXiv,
March 25, 2022. http://arxiv.org/
abs/2203.13609.
Zhang, Jianjing, et al. “Neural
Rendering-Enabled 3D Modeling
for Rapid Digitization of In-Ser-
vice Products.” CIRP Annals,
vol. 72, no. 1 (January 1, 2023):
pp. 93–96. www.sciencedirect.
com/science/article/abs/pii/
S0007850623000409.
Zhang, Lvmin, Anyi Rao, and
Maneesh Agrawala. “Adding
Conditional Control to Text-to-
Image Diffusion Models.” ArXiv,
org/10.48550/arXiv.2302.05543.
Zhang, Susan, Stephen Roll-
er, Naman Goyal, et al. “OPT:
Open Pre-Trained Transform-
er Language Models.” ArXiv,
June 21, 2022. http://arxiv.org/
abs/2205.01068.
Zhao, Mingmin, et al. “Assess-
ment of Medication Self-Admin-
istration Using Artificial Intelli-
gence.” Nature Medicine, March
18, 2021. https://doi.org/10.1038/
s41591-021-01273-1.
Zhou, Tingtao, et al. “AI-Aid-
ed Geometric Design of An-
ti-Infection Catheters.” ArXiv.
org, April 27, 2023. arxiv.org/
abs/2304.14554.
Zhu, Luyang, Dawei Yang, Tyler
Zhu, et al. “TryOnDiffusion: A Tale
of Two UNets.” ArXiv, June 14,
arXiv.2306.08276.
SELECTED SOURCES

WEB3

TABLE OF CONTENTS
WEB3
167 Top Headlines
168 State of Play
169 Key Events
170 Likely Near Term
Developments
171 Why Web3 Trends Matter
to Your Organization
172 When Will Web3
Trends Impact Your
Organization?
To Consider
175 Central Themes
177 Ones To Watch
178 Important Terms
180 The Web3 Landscape
181 The Rising Regulation of Web3
182 Decentralizing Venture Funding
182 Security Risks Take Different
Shapes
192 Scenario: What If We
Established Systemically
Important Technology
Institutions (SITIs)?
193 Web2 Web3 Integration
194 Digital Content Provenance and
Authentication
194 NFTs: Beyond the JPEG
195 Self-Sovereign Identity
Solutions
195 Tokenization of TradFi, Digital
Assets Security Tokens
196 Credibility Scoring and
Anonymity
196 On-Chain Gaming: Play, Own,
Earn, Enjoy
197 Scenario: What If We Used
AI To Monetize Privacy?
198 Authors
182 Superpowering Blockchain with
AI and IoT
183 Build through the Bear
184 Scenario: What If SEC lawsuits
were successful?
185 Web3 Infrastructure
186 Proof of Stake Proves Its Worth
186 Emerging Forms of Consensus
Protocols
186 Blockchain Modularity
187 Zero-Knowledge Proofs
187 The Appchain Thesis on
Ethereum
188 Data-as-a-Problem
189 Decentralized Applications
190 DeFi Protocols Network Effects
190 Guerilla Marketing: Crypto
Airdrops
190 Hyperfinancialization
191 Personal “X-As-A-Service”
Earning Models
TECH

Web3 has had a
challenging year, but
protocols in progress
bode well for its future
on the other side.
TOP HEADLINES
WEB3
01
02
03
04
05
If You Build it They Will Come (Maybe)
Many of the technical restrictions that limited the blockchain functionality are being
solved; however, novel, adoption-driving applications have yet to be developed.
Bad News Sells Better, but Innovation Survives
Major media crypto headlines remain overtly negative, matching the landscape
of economic uncertainty, but development, project launches, and TradFi interest
remain elevated.
Just Don’t Call Them NFTs
Consumer-facing companies increasingly use blockchain-based digital collectibles
to create and foster online communities for their fans and customers.
Web2.5 the New Web3?
Web3 promised to replace Web2 tech, but developments and integrations on both
sides of the web divide blend the tech, better meeting users where they are.
Splinter Regulation of a Global Network
Blockchain networks are inherently global, but their nodes and users are not;
fragmented regulations across geographies shift businesses and restrict users.
TECH

STATE
OF PLAY
The crypto market has had a rough 12-18 months. Challenging economic
conditions and high-profile meltdowns have pushed market volume, pric-
es, and VC investment downward.
But bright spots exist beyond the headlines and price charts. Throughout
the bear market, developers have continued to ship code, releasing cut-
ting-edge projects and providing core scalability and reliability enhance-
ments to public networks like Ethereum and Solana. Aside from raw tech-
nology infrastructure advancements, there has never been an easier time
for new companies or traditional businesses to launch custom chains due
to the developing Rollup-as-a-Service business model and SDK from lead-
ing Ethereum Layer2 scaling solutions.
However, significant roadblocks in regulation, adoption, and security still
limit the industry’s growth past early adopters. Companies and projects
planning for future advancements will have a head start over competitors.
Companies are building private blockchain networks to enhance their
business, giving them the core benefits of blockchain while reducing scal-
ability and security concerns. Finally, there is a concerted effort on both
sides of the web divide to enhance the adoption of blockchain by integrat-
ing characteristics of Web2 and Web3 to provide users with a more effort-
less, more familiar experience and entry point into the industry.
Blockchain has proven
technological staying power, but
its product-market fit remains
uncertain, and near-term events
will significantly impact its
potential.
WEB3
TECH
168

APRIL 20, 2023
Markets in Crypto Assets (MiCA)
MiCA regulation receives formal
EU adoption as the first regulatory
crypto framework in the world.
JUNE 5 6, 2023
SEC Sues Binance and Coinbase
The regulator targets top companies
in the crypto exchange industry and
classifies many coins as securities.
AUGUST 9, 2023
Base Launch on OP Stack
Base, a Layer 2 scaling solution
for Ethereum developed by
Coinbase, launches its mainnet.
JUNE 23, 2023
BlackRock Bitcoin Spot ETF
The investment company’s filing for
a Bitcoin Spot ETF spurs a deluge of
applications from other institutions.
AUGUST 15, 2023
Zynga Announces “Sugartown”
The leading mobile game
developer teases its first
blockchain-integrated game.
KEY EVENTS
WEB3
TECH

BRIDGING THE WEB DIVIDE
Many of the technological limitations
of blockchain have been resolved
or reduced, but adoption is the next
hurdle for blockchain technology and
crypto markets. Many forces are driv-
ing and limiting adoption today, and
the speed of adoption will depend on
the intersection and final outcomes of
these forces. Even in the depth of the
bear market, the industry is seeing
very promising signs of interest and
adoption from traditional industry
players. However, major roadblocks—
like regulation—persist, and these
factors are largely out of the hands of
the crypto industry.
Enshrined Account Abstraction
Account abstraction is a proposed upgrade
to Ethereum that, when implemented, will
provide flexibility in account setup via smart
contracts. This will give users easier routes to
maintain self-custody of tokens, more akin to
account management in Web2.
Verifying AI Output
AI models are sowing online discord. Deep-
fakes and misinformation are major issues
for political elections and online interaction.
Zero-knowledge cryptography could enable
verifiable online content and remove distrust
behind content consumed online.
Globally Successful Web3 Game
The video game industry is years into the
creation of top titles that have blockchain
built into the core of the gameplay. The
similarities of these games to titles gamers
are familiar with, combined with new ways to
play, should attract crypto natives and Web2
gamers.
Tokenized Asset Network Adoption
Traditional global financial players are working
on private blockchains to facilitate the trans-
fer of tokenized financial assets, which are
quicker to transfer and settle, and which help
companies avoid the regulatory and security
concerns of public networks.
US Regulators Forced to Decide
SEC investigation and regulation of crypto
assets is causing pressure to mount. Major
decisions around the SEC’s regulatory scope,
classification of securities, and legality of de-
centralized finance are coming to a head, which
could shift the entire crypto market.
Crypto Double Down in Africa
In sub-Saharan Africa, crypto is more than a
“nice-to-have”; it’s a financial necessity. A mix
of financial instability and demographics in
this region have led to the quiet adoption of
crypto payments, which will lead to a greater
industry focus in the region.
11 MACRO SOURCES OF DISRUPTION
Technology Government
Media
Telecom
Public Health Infrastructure
Demographics Education Economy
Environment Geopolitics Wealth
Distribution
WEB3
TECH

The outside world widely
understands crypto as an
instrument for currency
exchange. However, novel
decentralized business
models in lending/
borrowing, sequencing
transactions, float, and
others have developed.
They are driving revenue
for projects and
stakeholders through
high levels of automation
and low overhead.
Blockchain has created
the ability to own and
transact online value.
Web3 users have adopted
this technology, shifting
their expectations of
ownership rights of digital
goods and services. As the
adoption of Web3 grows,
it will shift consumer
preferences, requiring
businesses to enhance
their digital operations.
Storing data in
cloud databases has
limitations and downfalls.
Data can easily be
manipulated maliciously
or accidentally, and
sharing data outside
the organization can be
slow and challenging.
Blockchain’s immutable
ledger can be used in
private networks to store
and share data, avoiding
scalability and security
concerns.
Web3, like many
technologies, has a
rapid pace of innovation
that causes it to evolve
quickly and often.
Paired with regulatory
uncertainty, businesses
contemplating entering
the space should focus
on the macro details,
application opportunities,
and risks while
continuing to consistently
track technical aspects
and nuances.
DeFi solutions may
not seem to threaten
traditional finance,
but they shouldn’t be
overlooked. Many projects
and protocols exist at a
scale that can compete for
liquidity, loans, trading,
and other services from
clients and users. As
investors become more
comfortable with Web3,
these solutions will pose
real competition.
As Web3 solutions
become more
sophisticated, legacy
businesses will benefit
from establishing early
partnerships to learn
about and implement
the technology. There
are many opportunities
to build relations with
leading Web3 projects
today that can provide
meaningful learnings
for near-term strategic
decision-making.
New Business Model
Opportunities
Shifting Consumer
Expectations
Strengthening Data
Management
Need for Technical
Sophistication
Tracking New Sources
of Competition
Integrations and
Partnerships
WHY WEB3 TRENDS MATTER TO YOUR ORGANIZATION
WEB3
TECH

TECH
172
WHEN WILL WEB3 IMPACT YOUR ORGANIZATION?
Forecasted Time of Impact
Financial services
Video games
Social media
Supply chain management
Customer relations
Real estate
Higher education credentialing
IoT
Artificial intelligence
Health care Interplanetary economies
0-4 YEARS 5-9 YEARS 10-14 YEARS 15+ YEARS
WEB3

Threats
The decentralized nature of blockchain can expose businesses to new forms
of cybersecurity risk, particularly for small and medium-size businesses that
lack the capacity or technical sophistication to audit their Web3 capabilities.
Many jurisdictions are still developing regulations for blockchain technology
and cryptocurrencies, posing risks for businesses in terms of compliance,
legal challenges, and sudden changes in the regulatory landscape.
As Web3 grows in popularity, issues like network congestion, high
transaction fees, and slow processing times may hinder the user experience
on Web3 services, impacting businesses that are early adopters.
The high volatility of the cryptocurrency market and immaturity of Web3
startups can pose risks to businesses that rely on cryptocurrency for
transactions, fundraising, or as part of their business model.
The complexity of today’s Web3 applications compared to traditional
Web2 solutions could be a barrier to widespread adoption and will burden
businesses implementing Web3 to ensure ease of use for consumers.
Opportunities
Web3 enables tokenizing real-world assets that can be exchanged via
blockchain marketplaces. Businesses can transform how they exchange goods
or develop related services for asset valuation or exchange platforms.
Advancements in security and cryptography will unlock the development of
decentralized identity solutions that allow consumers to control their personal
information and share it or authenticate it without revealing PII.
Web3 technologies like the InterPlanetary File System (IPFS) provide
opportunities for decentralized data storage that will enable businesses to
transform their data storage, management, and verification solutions.
As more companies integrate blockchain into their operations, there will be
a growing demand for advisory services that is domain-specific and helps
companies ensure they are implementing the technology responsibly and
correctly.
With the growth of virtual assets and worlds, companies can transform their
Web2 properties into increasingly immersive and interactive Web3 properties,
for consumers to interact with in novel ways.
WEB3
TECH

Partnerships have become
essential for two-way
knowledge transfer and
exposure to Web3. Compa-
nies should seek partners
across the web divide with
mutually beneficial goals
where technical com-
ponents in Web3 benefit
traditional businesses,
and traditional business-
es can provide adoption,
exposure, or reputational
benefits.
Businesses should begin
building or improving the
necessary structure for
Web3 applications. This
may include setting up
a more robust network,
building the foundation for
user-friendly interfaces,
or developing middleware
that facilitates the integra-
tion of traditional systems
with blockchain-based
systems.
Companies—especially
large or public institu-
tions—should begin en-
gaging with regulators
to stay up to date on the
development of related
regulations or to begin
shaping favorable regu-
lations for their industry.
Companies may also invest
in legal expertise to navi-
gate Web3-specific regula-
tory environments.
Businesses may seek to
improve their institutional
knowledge in Web3 by
working to develop new
blockchain protocols,
exploring use cases for
Web3 within their industry,
or experimenting with new
forms of digital assets. This
early experimentation can
help companies prioritize
starting points for Web3
engagement.
Educational initiatives and
training serve as a starting
point for companies that
may be affected by Web3.
Companies should provide
upskilling opportunities for
employees to grow their
knowledge of Web3 tech-
nologies from foundational
protocols to user-facing
applications, and all of the
technologies in between.
Internal technology teams
should revisit security
protocols, encryption
standards, and data
security approaches to
take advantage of Web3
capabilities such as
zero-knowledge proofs.
This review will also help
companies prepare for
new and increasingly
complex cyber risks that
will emerge from a more
interconnected economy.
1 4
2 5
3 6
WEB3
TECH

CENTRAL THEMES
175
WEB3
TECH
Web 2.5 as a Bridge to Web3
Web3 has long been touted as a replacement for Web2
technology, but many barriers limit the technology’s
widespread adoption by businesses and users. Com-
panies and projects on both sides of the web divide are
acutely aware of these issues, and strides are being
made to overcome the barriers. Many of these innova-
tions and developments are leading to an intermediate
Web2.5: a technology that gives users the convenience,
familiarity, or safety of Web2 as well as the ownership
and decentralized aspects of Web3. Web2.5 may just be
the training wheels the world needs to drive adoption,
or it could be a lasting infrastructure that lets users
operate across the spectrum of the web. While Web3
infrastructure continues to make progress with its
potential to support the entirety of the web, regulation,
consumer preferences, and business practices will
likely be limiting factors in the time to come.
Positive Innovation Externalities
Blockchain technology is extremely versatile, with
applications well beyond finance and the ability to
integrate with other rising technologies like artificial
intelligence and Internet of Things to solve business
problems. The rise in crypto market valuations and
popularity has also increased investment in adjacent
technologies that work within the ecosystem and will
play an important role in solving issues we face dai-
ly online. However, overly restrictive regulations that
do not account for the novel nuances of the technol-
ogy and assets built on top of it threaten to subdue
or eliminate the positive innovation occurring in the
space. Regulators need to work with industry experts
to develop new regulations that hold innovation and
investor protection in the highest regard to limit ne-
farious activity without eliminating the technology’s
positive externalities for economies and society.
Relentless Building
Crypto prices are way off all-time highs, media cover-
age is overtly negative, transaction volume is down,
and the fallout from FTX’s collapse still hangs over
the industry. And yet, late 2022 and 2023 have been
arguably the most productive period in Web3’s his-
tory. Major networks have had multiple significant
upgrades, the ecosystem of Layer 2 chain on Ethereum
has blossomed, and blockchain-based games continue
to launch and improve. That said, there are blemishes
in crypto’s rebound: The NFT market has lost signifi-
cant value, and security and scams remain a key issue
though they haven’t scared off the developers. This fo-
cus on building has even spread to traditional compa-
nies as major financial institutions, retailers, and video
game companies have used this market to focus on
experimentation with Web3 aspects that can enhance
their businesses. Companies that continue to overlook
Web3 because of headlines or comfort with previous
technologies may fall behind competitors that have
worked countercyclically to the hype cycle.

CENTRAL THEMES
176
WEB3
TECH
Blockchains as Infrastructure
Today, the terms “Web3,” “internet,” and “metaverse”
are often (incorrectly) used interchangeably. This re-
flects a broader trend in technology use: Most users en-
gage with technology on a functional level, rather than
with an understanding of its intricate mechanics. This
is particularly true, and important, for blockchain tech-
nology. Business leaders should focus more on iden-
tifying the real business impact of blockchain, which
will be realized through the applications built on it. As
the technology rapidly evolves, it has the potential to
support full-scale, innovative applications. A key indi-
cator of blockchain’s success will be when its complex
infrastructure becomes an unnoticed foundation, en-
abling powerful and successful applications to take the
forefront. This transition marks a significant milestone
in technology integration and user experience, high-
lighting the value of functionality over technical details
in driving business innovation and user adoption.
Hiring Robotic Staff
Each year, cobots become smarter, more autonomous,
and more prolific, and this year is no exception. In fact,
the first humanoid robot factory is set to open and
produce 10,000 robots a year. These robots that work
alongside human workers are being trained on more
data that allows them to adapt and work around their
human counterparts. Developers have focused on
improving cobot safety measures, so a cobot knows
what to do if it bumps into an unexpected obstacle or
person. These cobots mitigate potentially harmful work
for humans by either augmenting the human body or
replicating repetitive tasks that could cause future in-
juries. Some of the augmented wearables can also offer
predictive pathways through the warehouse to ensure
worker safety. As autonomy continues to grow in robots
and transportation, this trend will create newfound
efficiencies and productivity, particularly during peak
demand seasons.
The Intelligent Manufacturing Evolution
Manufacturing continues to transform from a tradi-
tional labor-intensive practice to a more sophisticated
and interconnected system. Recent advances intend
to create higher levels of productivity and efficiency,
but they’re also addressing sustainability require-
ments and enabling product personalization. The new
tools and technology can spot flaws in products before
they leave the floor, greatly increasing consistency for
goods. Along with quality control, sensors and digital
twins are getting companies to focus on predictive
maintenance by reducing downtime during large runs
or times of high demand. Additive manufacturing also
allows for the integration of new materials that are
themselves smarter and more connected. And it helps
reduce the number of parts needed for production of
a good, which can streamline production and reduce
waste as only the parts needed are produced.

Hayden Adams, CEO of Uniswap Labs and
developer of the Uniswap decentralized ex-
change protocol, for his thought leadership
in Web3 and DeFi protocol development.
Vitalik Buterin, co-founder of Ethereum and
other open-source projects, for his steward-
ship of Ethereum, blockchain thought leader-
ship, and vision for the technology’s future.
Evin Cheikosman, director at Blockchain
Law for Social Good Center and former leader
of the World Economic Forum’s Crypto Sus-
tainability Coalition, for her thought leader-
ship on blockchain’s benefits.
Anatoly Yakovenko, co-founder of Solana
and CEO of Solana Labs, for his ongoing lead-
ership in the Solana ecosystem and advocacy
for the unification of blockchain technologies.
Sergey Nazarov, co-founder of Chainlink, for
his ongoing contributions to Chainlink’s ora-
cle network and to interoperability protocols
driving DeFi and TradFi integration.
Dr. Balaji Srinivasan, former CTO of Coin-
base, general partner at Andreessen Horow-
itz, and author of “The Network State,” for his
thought leadership on Web3 for businesses
and governments.
Dr. Gavin Wood, co-founder of Ethereum and
creator of Polkadot and Kusama, for his devel-
opment of leading blockchain technology and
vision for Web3.
Brian Armstrong, founder and CEO of Coin-
base, for his unwavering commitment to the
US crypto market and thought leadership on
application-specific use cases of blockchain.
Changpeng Zhao, co-founder and CEO of
Binance, for his leadership of the world’s larg-
est cryptocurrency exchange amid increased
regulatory scrutiny across the globe.
Joseph Lubin, co-founder of Ethereum and
founder and CEO of ConsenSys, for leadership
and foundational innovations like MetaMask
and Linea zkRollup.
Prithvi Subburaj, a 15-year veteran of Google
and now COO of OP Labs, for his background
and opportunity to lead the expansion of the
Optimism Network.
Steven Goldfeder, co-founder and CEO of Off-
chain Labs (developer of Arbitrum) and author
of “Bitcoin and Cryptocurrency Technologies,”
for his vision for Etherum’s future.
Yoseph Ayele, founder of Borderless Africa
and community builder, for his dedication to
solving Africa’s economic insecurities with
Web3 technologies, education, and access.
Neel Somani, founder of Eclipse Labora-
tories, for pushing the limitations of the
modular blockchain thesis and encourage-
ment for the unification of blockchains and
communities.
Mustafa Al-Bassam, a hacktivist turned se-
rial entrepreneur, now co-founder and CEO of
Celestia Labs, for his contributions to block-
chain modularity and Celestia mainnet.
Fumio Kishida, prime minister of Japan, for
his support of Web3 initiatives and his work
to make Japan a more open environment for
Web3 businesses.
ZachXBT, a pseudonymous X (Twitter)
influencer and on-chain detective, for his
commitment to uncovering bad actors in the
crypto ecosystem tied to scams and crimes.
Yat Siu, co-founder of Animoca Brands and
investor, for leading a major blockchain gam-
ing company and pursuit of a decentralized
metaverse with secure digital property rights.
Daniel Alegre, Activision Blizzard and Google
veteran, now CEO of Yuga Labs, for his lead-
ership of one of the most innovative Web3
brands combining NFTs, gaming, and culture.
Avery Ching, co-founder and CTO of Aptos,
for his experience working with traditional
tech companies and leadership of a promis-
ing Layer 1 blockchain.
Rune Christensen, co-founder and CEO of
MakerDAO, for his contribution to DeFi and
decentralized stablecoins and his vision for
the integration of TradFi and DeFi.
Larry Fink, chairman and CEO of BlackRock,
for his support of a tokenized future for Trad-
Fi and his leadership of a company paving
the way for crypto asset ETFs proliferation.
Jose Fernandez da Ponte, general manager
of blockchain, crypto, and digital currencies
at PayPal, for his leadership in driving the
adoption of implementing blockchain tech-
nology and compatibility in TradFi.
Daniel Shorr, co-founder and CEO of Modu-
lus Labs, for his rapid and transparent exper-
iments and implementations of AI models
deployed on public blockchains.
ONES TO WATCH
WEB3
TECH

IMPORTANT TERMS
178
Airdrop
A marketing strategy where a project team distrib-
utes tokens to users for free in exchange for using
the protocol or other requirements. Airdrops are
often used as a guerrilla marketing technique to
stimulate interest and adoption.
AppChains (application specific chains)
Special-purpose blockchains serving a single
application. This gives developers total control of
software upgrades and gives users less competi-
tive block space of general-purpose blockchains.
Block space
The storage area on a blockchain for transaction
and data storage, including smart contracts. Block
space significantly impacts blockchain scalability
and decentralization, and therefore gas fees for
data inclusion.
Blockchain
A distributed ledger technology typically employed
for the transaction and storage of data. It utilizes
cryptography to provide an immutable and verifi-
able data source for participants in a network.
Decentralized Exchange (DEX)
Aa peer-to-peer marketplace for users to trade
crypto assets.
Decentralized finance (DeFi)
Financial services including banks, asset man-
agers, insurance companies, and other financial
services that leverage blockchain and smart
contracts for transactions, data sharing, and other
operations.
Exit scams (“rug pulls”)
A common type of fraud where a project team
deceives investors to garner their investments and
uses a purpose-built vulnerability to drain all funds
and abandon the project.
Fork
A term commonly used to describe the act of copy-
ing and/or modifying existing code to either up-
grade an existing system or launch a new product.
Forks are frequently necessary for blockchain-wide
software updates and are commonly observed in
the decentralized finance (DeFi) sector, where one
project replicates the code of another.
Layer 2 blockchain
A broad term that describes blockchains that
delegate core infrastructure to another blockchain.
Examples include Ethereum’s Rollups and Bitcoin’s
Lightning Network, which aim to enhance scalability.
Modular vs. monolithic
Blockchains can be modular, breaking core com-
ponents (execution, settlement, data availability,
consensus) into separate specialized networks
to address the blockchain trilemma. Monolithic
chains like Ethereum provide all core modules
within their infrastructure.
NFT (non-fungible token)
A digital token on a blockchain that contains
unique and indivisible data. It is frequently used in
digital art or when tokenizing real-world assets.
Nodes
Individual devices within a connected network of
computers that serve various functions such as
communication, transaction validation, and histor-
ical data storage within a blockchain network. Dif-
ferent nodes exist, each with functionality specific
to the network they support. Examples include full,
light, super, and archive nodes.
Blockchain trilemma
An optimization challenge faced by monolithic
blockchains, requiring trade-offs between decen-
tralization, scalability, and security; only two can
be maximized. Solana, known for high transactions
per second (TPS), prioritizes scalability and security
over decentralization.
Bridge
A tool to facilitate the transmission of information
and assets between distinct blockchains regard-
less of the interoperability of the networks.
Decentralization
The process of constructing architectural infra-
structure, system logic, and social systems without
the presence of a centralized authority that holds
decision-making power or exerts disproportionate
influence. Instead, control is distributed among the
stakeholders of the network.
Decentralized autonomous organization (DAO)
An internet-native organization formed by individ-
uals who agree to adhere to a specific set of rules
and goals without a central authority. DAOs employ
tokenized ownership and smart contracts to imple-
ment decisions.
WEB3
TECH

IMPORTANT TERMS
179
Oracle
A capability or service that gathers, collects, and
transmits data on- and off-chain to facilitate
real-time transactions and information transmis-
sions. Oracles are bridges between blockchains
and external off-chain information sources on the
internet.
Phygitals
This refers to the blending of physical and digital
assets into an NFT. Phygitals are commonly used
for tokenizing physical collectibles and art pieces;
they frequently include a burn and redeem func-
tionality where the NFT is destroyed for the owner to
receive the physical item.
Proof of stake (PoS)
A blockchain consensus mechanism that uses
stake tokens to secure the network. Validators
(nodes responsible for verifying blocks of transac-
tions) must stake their tokens (use them as collat-
eral) to participate in the block verification selec-
tion process. Malicious validators—those that fail
to validate or attempt to mislead the network—will
Sharding
A database partitioning technique that divides an
extensive database into more manageable parts
called shards. Ethereum’s roadmap plans to use
an adapted sharding methodology to improve the
scalability of the blockchain by partitioning the
chain and its validators into distinct but intercon-
nected shards, allowing for parallelized transaction
processing.
Smart contracts
A blockchain-based computer program that exe-
cutes autonomously when predetermined criteria
are met.
Stablecoin
Cryptocurrency assets whose value is referenced
(or pegged) to another financial instrument, often
a fiat currency. These assets are typically collateral-
ized by fiat currencies, cryptocurrencies, and liquid
assets.
Tokenomics
The economic framework of tokens, encompassing
elements such as consensus mechanisms, yields,
supply limits, and other monetary policies.
Traditional finance (TradFi)
Conventional means of money or asset manage-
ment where services are provided by traditional
banks, asset managers, insurance companies, etc.
Zero-knowledge proofs (ZKPs)
Mathematical techniques that allow users to prove
knowledge (the prover) of something without di-
vulging the private knowledge associated with it to
another user (the verifier). Zero-knowledge proofs
encompass two core principles important to block-
chain technology: succinctness, which means that
verifying the proof is significantly easier than pro-
ducing the computation itself, and privacy-preserv-
ing, which involves hiding portions of computation
while maintaining correctness during verification.
see their collateral value slashed, while benevolent
validators earn yields or other benefits for their
work. Ethereum successfully transitioned from
proof of work to PoS, resulting in a 99% reduction in
the blockchain’s energy consumption.
Rollups
A subcategory of Layer 2 blockchains with a
scalability focus that process and bundle transac-
tions to be submitted to Ethereum for settlement
and consensus. The most popular types include
optimistic rollups like Arbitrum and Optimism and
zkRollups like zkSync.
Scalability
A blockchain’s capacity to process and store data as
network demand grows, typically measured in TPS.
Security tokens (ST)
Digital assets representing ownership of off-chain
assets such as bonds, commodities, or real estate.
Off-chain assets are tokenized into STs to enable
trading on blockchain networks.
WEB3
TECH

THE WEB3
LANDSCAPE
WEB3
TECH

THE WEB3 LANDSCAPE
TECH
The world of cryptocurrency and Web3 tech-
nologies has witnessed exponential growth
in recent years, with North America, Europe,
and East Asia emerging as the three largest
markets in terms of cryptocurrency volume.
However, within these regions, three major
players—Japan, the European Union, and the
United States—are taking distinctly differ-
ent approaches to regulating this dynamic
industry. These regulatory decisions have
far-reaching impacts, influencing innovation,
business creation, and even the global power
dynamics within the crypto industry. Regula-
tion is a double-edged sword. On one hand, it
can provide stability and investor protection,
fostering trust in the market. On the other
hand, it has the potential to stifle innova-
tion and disrupt the balance of power in the
crypto industry. However, businesses in this
space are resilient, and innovations tend to
be “sticky,” often finding ways to thrive even
under regulatory constraints.
The European Union, in contrast, has po-
sitioned itself as a “fast follower” in the
realm of crypto regulation. In 2023, the EU
ratified the Markets in Crypto-Assets (MiCA)
framework, which is set to take effect in
2024. MiCA aims to strike a balance by
creating a regulatory framework that builds
on successful regulations in countries like
France. The objective is to foster a workable
environment that encourages innovation
without stifling it, all while ensuring the
protection of investors. This approach has
been warmly received by the crypto industry,
with positive public relations and marketing
efforts successfully attracting business-
es, such as Coinbase and Nexo, to relocate
to stable regulatory regions within the EU.
Notably, Nexo, a US-based business, moved
to Europe due to regulatory inaction in its
home country.
The United States presents a more chaotic
regulatory landscape for the crypto indus-
try. The lack of a clear partisan line for or
against crypto has resulted in many bills
floating around the House and Senate,
each with drastically different stances on the
future of crypto in the US. The Securities and
Exchange Commission has resorted to regu-
lation by enforcement, adding to the overall
uncertainty.
Despite this regulatory uncertainty, busi-
nesses in the TradFi and blockchain sectors
continue to launch crypto-focused products.
However, the lack of clarity has somewhat
tempered the full potential of these innova-
tions. It’s important to note that investors,
both current and prospective, seek regulatory
clarity. As crypto has become an ingrained
investment category in the US, its regulation
is crucial for attracting and retaining capital.
Regulation and innovation must be balanced
effectively in order for Web3 to succeed. These
countries must work closely with the crypto
industry to protect investors while fostering
the incredible potential blockchain technol-
ogy and Web3 innovations can bring to their
respective economies. The evolving landscape
of crypto regulation will continue to shape the
industry’s future on a global scale.
Japan was an early adopter of cryptocurren-
cy technology, with the country serving as
the epicenter of crypto activity in the early
2010s. However, the infamous Mt. Gox ex-
change debacle and other issues prompted
the Japanese government to introduce strin-
gent regulations and tax regimes to protect
investors. These measures, while well-inten-
tioned, had the unintended consequence of
restricting the viability of Web3 businesses
in the country. Businesses, however, recog-
nized the potential of blockchain technology
and cryptocurrencies and took the initiative
to create self-regulating organizations to
promote adoption, particularly in areas such
as the Security Token Market. Today, Japan
is gradually rolling back these strict regula-
tions, signaling a desire to encourage Web3
businesses to bring their innovation and
investment back to the country.
WEB3
The Rising Regulation of Web3

THE WEB3 LANDSCAPE
TECH
Decentralizing Venture Funding
Web3 VC funding has reached a three-year
low, driven by regulatory uncertainty and the
specter of major company collapses. Despite
this drought, developer activity (SDK down-
loads, smart contract deployments, test net
activity) is surging across various blockchain
ecosystems. In this bearish market, DAOs and
foundations are pivotal, enabling builders
to persevere. The Ethereum Foundation, with
assets nearing $2 billion, is a prime example
of decentralized funding supplanting ear-
ly-stage investment traditionally targeted by
seed-stage investors. The Foundation’s suc-
cess has birthed DAOs and foundation funds
for alternative Layer 1 blockchains, allowing
them to invest in their own growth flywheels.
Uniswap, a prominent DEX, has grown to the
extent that it now operates its own grant
program to fund DeFi protocol development.
Grants, often structured more favorably than
VC investments, offer a straightforward appli-
cation process and are typically non-dilutive.
While VC funds have played an indispens-
able role thus far and are expected to persist
in Web3, the proliferation of DAOs and the
AI for auditing, verification, and proactive
threat detection.
Social engineering exploits, on the oth-
er hand, rely on human error and include
phishing attacks and exit scams (commonly
known as “rug pulls”). Rug pulls are popu-
lar due to their ease of execution and were
responsible for $100 million in scams in the
first half of 2023 and just over $200 mil-
lion in 2022. Failure to safeguard investors
through education and industry efforts may
prompt regulators to intensify scrutiny and
impose tighter controls on existing plat-
forms.
These looming threats underscore the urgent
need for enhanced security measures and
regulatory oversight to protect participants
and promote trust within the ecosystem.
Superpowering Blockchain with AI and IoT
Blockchain is becoming even more powerful
by integrating with IoT and AI for specialized
applications.
While these technologies have their strengths,
they also have known weaknesses. IoT gathers
data but is insecure, and AI is great for ana-
lyzing data but needs clean data to learn. But
when they’re paired with blockchain, the tech-
nologies become even more powerful. Light-
ning Labs’ Bitcoin tool kit mixes blockchain
with AI, enabling AI agent-to-agent payment
and AI access to paywalled APIs autonomous-
ly. When blockchain is combined with IoT,
it secures IoT data generation and IoT-to-IoT
communications. And when all three technolo-
gies combine, the benefits feed off each other.
State Farm has filed a recent patent to use
blockchain, AI, and IoT to track autonomous
vehicles on the chain to provide immutable
records of incidents, assign liability, and pro-
cess claims.
These kinds of integrations will become in-
creasingly common as many more companies
file patent applications involving all three
technologies. Those include a wide array of in-
dustries, from financial services (Mastercard)
to information technology (IBM) and food
intelligence platforms (Innit). As AI and IoT
expansion of their investable capital (the top
10 DAOs, excluding foundations, collectively
hold nearly $16 billion) can furnish addition-
al liquidity during market downturns and
expansions. This dynamic may engender
heightened competition for VC investors.
Security Risks Take Different Shapes
Web3 security concerns are on the rise, with
a growing focus on hacks and on-chain
scams that result in the loss of millions of
dollars. These illegal activities fall into two
broad categories—code vulnerabilities and
social engineering exploits.
Critical targets for code vulnerabilities
include bridges, liquidity pools, and wallets
due to their concentration of value. Despite
protocols’ investment in secure smart con-
tracts, code audits, and bug bounty pro-
grams, the crypto space exceeded $1.5 billion
in losses in 2023. On-chain cybersecurity is
still in its nascency due to the incompatibil-
ity of Web2 security measures with open-
source networks, driving on-chain security
solutions companies to explore the use of
WEB3

THE WEB3 LANDSCAPE
TECH
become more ubiquitous, we can expect new
blockchain applications due to the synergis-
tic relationships with these technologies.
Build through the Bear
Building through the bear market is a recur-
ring theme in the crypto space: Even when
markets cool and attention shifts away from
the technology, protocol developers and
companies keep innovating. This trend has
persisted in the crypto world and is becom-
ing more common at traditional companies
building new crypto solutions for the next
economic upturn.
Despite the recent downturn and crypto’s
uncertain regulatory landscape, new compa-
nies continue to launch, operate, and devel-
op Web3 initiatives. Those include payment
companies like Visa and PayPal; investment
applications like Spot Bitcoin ETH Futures,
and ARK Invest; brokerage firms like Coin-
base’s L2 network; NFT loyalty programs like
that offered by Adidas ALTS; and games like
Zynga’s “Sugartown.” As traditional compa-
nies implement new business models and
cost reduction strategies in Web3, the tech-
nology is poised to have an impact as signif-
icant as cloud infrastructure in the 2010s.
This is the beginning of Web3’s era of in-
stitutional adoption, as major criticisms
of public blockchain technology around
energy utilization and scalability continue
to weaken. Expect to see continued devel-
opment and deployment of Web3 initiatives
from companies across every industry; the
companies experimenting with the technol-
ogy now will have a leg up on competition in
the future.
Adidas ALTS are NFTs that provide early access to virtual and physical products.
Image credit: Adidas
WEB3

SCENARIOS
SCENARIO YEAR 2027
What If SEC lawsuits were successful?
“US Falls Behind in Fintech as SEC Wins Major Crypto Lawsuits”
In a landmark development, the US Securities and Exchange Commission has successfully sued major crypto entities including Bi-
nance, Coinbase, DAOs, stablecoins, and NFT projects. This crackdown has driven crypto projects out of the US, severely limiting Amer-
ican access to the crypto markets and crippling related businesses. The SEC’s aggressive stance has effectively severed the on-ramps
for US citizens to crypto markets and reversed the country’s fortune in the crypto industry, once boosted by China’s crypto ban.
This shift has left the US lagging in fintech innovation. Now, regions like Japan, the EU, and Africa are emerging as new fintech lead-
ers, capitalizing on the exodus of blockchain expertise from the US. The departure of blockchain projects has led to a dearth of skilled
professionals in the field. US investors face minimal protection, as they can still access offshore protocols without oversight, echoing
the FTX collapse in the Bahamas. Japan, in particular, is rising as an economic powerhouse due to its alignment with the Web3 indus-
try and supportive regulatory environment. Meanwhile, the growth of DeFi and other blockchain projects is expected to slow, especially
with US users facing access barriers.
The upcoming Supreme Court decision on defining investment contracts could further shape the SEC’s reach over stablecoins and
DeFi. As the global blockchain business adapts, the US’ once-dominant position in fintech innovation is now challenged, marking a
significant shift in the landscape of financial technology.
TECH WEB3

WEB3
INFRASTRUCTURE
WEB3
TECH

WEB3 INFRASTRUCTURE
TECH
Proof of Stake Proves Its Worth
Ethereum’s 2022 transition from electric-
ity-guzzling proof of work (PoW) to a more
energy-efficient and decentralized network
secured by proof of stake was an incredible
feat. But it’s only a stepping stone in the
progression toward a hyper-scalable, fully de-
centralized, highly secure, and easily usable
platform for the internet.
Now, the Ethereum community has a list
of improvement proposals, and research is
ongoing to improve each area. They include
adding temporary data storage to enhance
the scalability of Layer 2 chains, increasing
decentralization by separating block propos-
er and block builder capabilities to strip out
MEV bots’ capabilities to censor transactions,
and improving account abstraction.
While none of the proposals or areas of re-
search have hard implementation timelines,
the Ethereum community has a proven track
record of delivering high-quality and thor-
oughly tested protocol updates. Companies
still on the Web3 sidelines because they lack
creasingly important for specialized chains,
which may need to change underlying in-
frastructure to bring new capabilities to the
blockchain ecosystem.
Another recurring theme is new alternative
Layer 1 blockchains. Most recently Aptos and
Sui, two heavily venture-backed L1s, deployed
their mainnets and are attempting to wres-
tle network usage away from incumbents
like Ethereum and Solana with tech stack
upgrades. They both use PoS but have very
different algorithms under the hood: They’re
now using a two-pronged approach to trans-
action consensus that allows for high scal-
ability through parallelizing transactions.
These protocols weren’t possible in Ethere-
um’s PoS based on how the blocks are struc-
tured and could present a red flag for the
company. Ethereum’s technology has lagged
behind the industry for many years, but its
massive network effects keep it relevant.
Time will tell if a younger blockchain with
more innovative consensus mechanisms
and infrastructure designs can outcompete
the current platform of choice.
Blockchain Modularity
The “blockchain trilemma” highlights a key
challenge in Layer 1 blockchains: Optimizing
for scalability, decentralization, and security
simultaneously is difficult. Ethereum excels
in decentralization and security but lags in
scalability, while Solana offers scalability and
security but compromises on decentraliza-
tion. To address this, the blockchain sector
is turning to modularity, separating Layer 1
blockchain functions—execution, settlement,
consensus, and data availability—into dis-
tinct, specialized chains.
This approach is evident in Ethereum’s Layer 2
solutions like Optimistic and Zero Knowledge
Rollups, which enhance transaction speed
and cost-efficiency. More broadly, various
Layer 2s and blockchains are experimenting
with different module combinations, aim-
ing to improve blockchain performance and
interconnect ecosystems. A notable example
is Eclipse, a new Layer 2 architecture that in-
tegrates Ethereum, Celestia, Solana, and RISC
Zero for different functionalities, showcasing
advanced modularity.
technical capability or fear usability issues
should observe these behind-the-scenes
moves; these new developments will contin-
ue to improve the capabilities of the plat-
form for all types of applications.
Emerging Forms of Consensus Protocols
By many measures, Ethereum’s merge to
proof of stake (PoS) was a great success:
The transition went smoothly, energy con-
sumption dropped by 99%, staked ETH that
secures the network has increased every
month since the merge, and many more
Ethereum users can participate in securing
the network. But while PoS is a mainstay con-
sensus mechanism in Web3, Ethereum’s PoS
is not a one-size-fits-all approach and still
has downsides, such as limited scalability.
Other networks have chosen different forms
of consensus protocols. Filecoin, a decentral-
ized storage blockchain network, uses two
different types of consensus mechanisms
that allow the nodes in the network to verify
data has been stored and continues to be
stored in the network. This may become in-
WEB3

WEB3 INFRASTRUCTURE
TECH
Although still in its early stages, the modular
blockchain concept is gaining traction. The
uptake of existing Layer 2 solutions and on-
going experiments in specialized chains sug-
gest a future where modular strategies could
be vital in solving the blockchain trilemma
for diverse applications and business needs.
Zero-Knowledge Proofs
Zero-knowledge proofs (ZKPs) aren’t new:
They were theorized back in the 1980s by
researchers at New York University and were
first deployed at scale with the 2016 launch
of ZCash, a privacy-focused cryptocurrency
based on bitcoin’s codebase. More recently,
improvements to computational capabilities
and the cryptography itself have allowed for
wider-scale deployment of ZKPs in the crypto
space, most notably with ZK-Rollups.
ZKPs have gained significant adoption due to
their two core characteristics: succinctness
and privacy. The succinctness characteristic
is the basis for the scalability enhancements
of ZK-Rollups, because nodes can compute
proofs of transactions off-chain and submit
Ethereum’s L2 chains like Arbitrum, Opti-
mism, and zkSync are embracing this ap-
proach, allowing easier creation of L2 and L3
chains. These solutions enhance deployment
ease, and upgrade flexibility, interoperabil-
ity, and scalability, although Cosmos was
the first to introduce this model. Coinbase’s
Base is another example of this trend, built
on the Optimism OP Stack. Its success may
prompt other companies to explore similar
solutions.
This shift is crucial as it enhances scalabil-
ity and reduces costs, making blockchain
more user-friendly, especially for high-trans-
action applications. Application-specific
chains offer more specialization and the
possibility of private chains within a decen-
tralized network, believed to be key for future
interoperability and simplified user experi-
ence. However, not all in the crypto commu-
nity agree with the multichain approach of
Appchain. Some, like Solana, focus on a sin-
gle chain for all applications. The effective-
ness of these diverse strategies will become
clearer over time.
the proofs to Layer 1 for verification. This
shifts the heavy computation off-chain while
still verifying the validity of every transaction.
However, ZKP characteristics allow them to
be applied far beyond blockchain scalabil-
ity solutions: Researchers are studying the
application of ZKPs to AI training, interaction,
and verification of model outputs. As com-
putation capacity continues to increase and
ZKP technology improves, expect to see it
used throughout digital interactions to pro-
vide great control over private data and de-
crease unverified or fraudulent information.
The Appchain Thesis on Ethereum
The Appchain concept, originated by Cos-
mos, has been adopted in Ethereum’s Layer
2 (L2) ecosystem. This idea advocates for
protocols to create their own independent
chains when they grow significantly, en-
suring full control and reducing reliance on
their original blockchain. RollApps, a key L2
provider, facilitates this by offering tools for
easy deployment of new chains.
WEB3

WEB3 INFRASTRUCTURE
TECH
Data-as-a-Problem
Data storage, now cheaper and faster in the
traditional economy, is a problem for block-
chains. Their unique design, which limits how
much data can be stored in blocks, makes
data storage costly and challenging to opti-
mize.
A key issue for blockchain scalability is
the block size and the hardware needed to
process and validate data. Bigger blocks
are more scalable but require more compu-
tational power, making them expensive for
validators and potentially reducing decentral-
ization. Solutions like sharding or modularity
can help blockchains circumvent these hard-
ware demands, but they face the data avail-
ability problem, where malicious nodes could
hide crucial transaction data. So, light nodes
need methods to verify data availability with-
out downloading everything, maintaining the
efficiency of sharding.
Developers are seeking solutions to these
challenges. PayPal sought a patent to prune
blockchain blocks and store them with a data
storage provider. Ethereum is set to intro-
duce EIP-4844 (proto-danksharding), adding
a new, less resource-intensive data section
to blocks, called “blobs,” to ease the compu-
tational load on nodes and improve scalabil-
ity. Other projects are exploring approaches
like data hosting consortiums and modular
blockchain designs. Further progress from
various crypto market segments should
enhance scalability without compromising
decentralization or security.
WEB3

DECENTRALIZED
APPLICATIONS
WEB3
TECH

DECENTRALIZED APPLICATIONS
TECH
DeFi Protocols Network Effects
DeFi protocol business models are traditional:
Protocols provide products and services and
generate fees distributed to stakeholders. But
these protocols operate in a unique environ-
ment: The business’s “secret sauce” is public
knowledge because nearly all DeFi protocols
are open source.
Open-source protocols can generate compe-
tition—Uniswap has forked nearly 500 times,
spurring continuous innovation—but they can
also splinter the market and provide an open-
ing for fraudulent activity. This was the case
with Compounder Finance, a fork of Yearn
Finance that scammed investors out of more
than $10 million with minimal development
effort.
Fraudulent projects are still a concern, but
the bear market has washed out smaller
competitors as investors leave the market or
flee to more reputable service providers. As
a result, large DeFi protocols continue to ex-
pand their influence, both vertically to provide
more DeFi services (such as Maker starting
they can miss the mark of getting real user
adoption. Users have learned to game air-
drop programs, known as “airdrop farming,”
where users set up dozens of automated
wallets programmed to maximize their
chances and total rewards from an airdrop.
Incentive structure and qualifications for
airdrops can limit this behavior, but many
platforms have struggled. Arbitrum, a lead-
ing Ethereum L2, had a significant issue
with airdrop farming that left many retail
users dissatisfied, and to this day, 72 million
ARB tokens are unclaimed.
While Arbitrum has become extremely suc-
cessful anyway, the viability of these pro-
grams is murky. Most, if not all, tokens with
significant airdrops see massive sell-offs
at token distribution, as many users dump
tokens to lock in value. It’s difficult to tell if
these users eventually return to the plat-
form. But even if the outcomes are unclear,
airdrops have become a mainstay in the
ecosystem as a significant hype generator.
Hyperfinancialization
Financialization has been a hot-button topic
in the crypto market as goods typically not
considered financial assets—such as art,
video games, and social media—are commod-
itized, recasting their value from enjoyment to
investment returns.
In art, advanced trading techniques like mass
NFT buys and sells commoditized art and
shifts the focus away from culture and com-
munity. At the same time, token incentives
spurred the gamification of NFT trades and
manipulated volumes. The financialization
of video games, GameFi (the combination of
decentralized finance and video games) is
often criticized for its lack of engaging game-
play and a hyper-focus on financialization. The
latest development in social media financial-
ization—Friend.tech—is also controversial. It
allows users to tokenize social connections
on X by buying and selling a sort of “key” in
public profiles that gives access to private
chat rooms and is criticized for commodifying
people through their social media accounts.
as a stablecoin provider but expanding into
lending as Spark Protocol) and horizontally
to provide the same services across multi-
ple chains (such as Aave’s lending product
being available across nine different block-
chains).
The average user can’t verify code and will
likely gravitate toward protocols with the
best reputations. At the same time, busi-
nesses will have robust and tested code-
bases to use and adapt to their own chain
services.
Guerilla Marketing: Crypto Airdrops
Airdrops raise awareness, drive user adop-
tion, stress test platforms, and reward active
or long-time users. They can be extremely lu-
crative: NFT platform Blur distributed tokens
as part of its multiseason airdrop program,
and in the first season, 23 users earned
more than $1 million in tokens.
Still, platforms should weigh benefits
against long-term goals. If airdrops are too
short term or have misaligned incentives,
WEB3

DECENTRALIZED APPLICATIONS
TECH
In each case, the implementation and degree
of financialization seems to be extremely
important in the outcomes of the applica-
tion. Despite the criticisms, financialization
remains a core tenet of the crypto ecosystem.
Expect to see new applications of financial-
ization continue to be developed with ensuing
volatility and uncertainty.
Personal “X-As-A-Service” Earning Models
Web3 has opened the door for consumers to
earn through “X-as-a-service” models that al-
low for the automation of personal hardware
and digital assets to generate returns—partic-
ularly in computation, security, and utility.
As mining is now monopolized by profession-
al outfits, users seeking to provide compu-
tation-as-a-service are pivoting to running
millions of iterations of training data for AI
models—and getting paid handsomely for it.
In Security-as-a-Service, blockchain users
can stake their tokens to participate in trans-
action and block validation, helping secure
the network. In the same area, restaking is
a new blockchain primitive where staked
token balances on Ethereum can be used to
validate Ethereum and other chains simulta-
neously for juicier yields. In NFT gaming proj-
ects, users can provide utility-as-a-service
by providing digital services to player bases
and earning portions of the transactions on
their digital properties.
There are many more as-a-service earning
opportunities for crypto adopters, such
as providing liquidity as a service on DeFi
platforms or providing cloud storage on Fi-
lecoin’s IPFS network. Many of these oppor-
tunities are limited to those with the knowl-
edge and technical abilities to navigate the
crypto ecosystem. Still, as barriers to adop-
tion come down, these earning models will
be available to anyone with internet access.
Crypto miner Hut 8 invested in five data centers which can now be used for other purposes, including
training AI.
Image credit: Coindesk
WEB3

SCENARIOS
SCENARIO YEAR 2038
What If We Established Systemically Important Technology Institutions (SITIs)?
The global financial system has undergone a transformation, with technology companies at its core. Over the past two decades, as digital assets
have taken over, technology companies have invested heavily, by developing digital wallets, cryptocurrency exchanges, and NFT marketplaces,
making these services integral to their digital ecosystems.
As a result, following a recent meeting, the US Financial Stability Oversight Council (FSOC) has announced plans to designate certain technology
companies as being systematically important: a significant departure from the definition of “systemically important institutions” portrayed in
the 2010 Dodd-Frank Act.
This oversight regulates technology organizations that now steward most of the country’s digital assets, which have become the underpinning
of economic value. Digital assets have ballooned in value due to the financialization and monetization of people’s digital identities, capabili-
ties, and online followings, as well as the tokenization of physical assets such as homes and vehicles. While blockchain technology spurred the
creation of digital assets and ease of transacting online, adoption hurdled around technical requirements, leaving oversight, self-custody, and
security concerns to be solved.
Systemically Important Technology Institutions (SITIs) will be required to adhere to strict standards in financial stability and risk management.
They must also comply with advanced cybersecurity measures to protect against potential threats. This increased burden of compliance will
weigh heavily on some technology companies but will at last secure the new economy of digital assets that many have come to rely on.
TECH WEB3

WEB2 WEB3
INTEGRATION
WEB3
TECH

WEB2 WEB3 INTEGRATION
TECH
Digital Content Provenance and Authentication
As more and more platforms adopt AI, it be-
comes increasingly important for users to be
able to verify the origin and authenticity of dig-
ital content. C2PA, the coalition of major tech
companies founded in 2020 looking to develop
an open-source infrastructure to protect digital
content and consumers from fake news, has
seen its membership increase by 60% over the
past year due to interest in AI.
Digital watermarking can help human content
creators by marking human content that’s
used in AI training to make sure it doesn’t
run afoul of copyright laws. In contrast, cryp-
tographic hash functions that track and store
the data manipulation history of assets can
help human consumers. Zero-knowledge
proofs can also help authenticate AI output
by providing a model with certain conditions
without revealing how the model created the
output. Of course, in the future, as AIs start to
be involved in higher-stakes decisions, there
will be more incentives to tamper with or re-
place the model. Modulus Labs is experiment-
ing with chain AI deployments for high visibili-
a sneaker release that are convertible into a
physical part of shoes once they’re produced.
ALTS by Adidas offer NFTs with perks for
holders, like exclusive access to drops.
Web2 and Web3 native NFT projects are
also converging. Blue-chip NFT projects like
Doodles and Pudgy Penguins have launched
clothing and toy lines, and Nouns DAO
funded a full-length movie featuring popular
NFT characters. Even if NFT trading markets
never return to the highs they reached in
2021 and 2022, the technology is helping
businesses and projects acquire, develop,
collaborate with, and monetize a deeper con-
nection with communities.
ty and easy verification of model outputs.
While progress has been swift, larger strides
are necessary to get these technologies
deployed at scale to protect consumers from
malicious and false information and content
creators’ ownership of their digital content.
NFTs: Beyond the JPEG
When NFTs hit the mainstream, they focused
on JPEG art. While this made NFTs a house-
hold term, the technology’s ease of transfer,
easy verification, historical ownership track-
ing, and immutability have benefits beyond
pictures-for-profiles. Now, NFTs are being de-
ployed to digitize ownership of assets in the
real estate industry, make event ticketing
more secure, and augment retailers’ loyalty
programs.
Consumer-facing companies are entering
Web3 with NFTs at the core of their strat-
egy: Nike, Starbucks, and Ducati have NFT
programs that bridge the gap between their
physical and digital products. Puma and NBA
player LaMelo Ball partnered with Open Sea
and Gutter Cat Gang to sell NFT versions of
WEB3
Starbucks launched an NFT project in Korea called
‘Starbucks Starlight’ which encourages customers
to use their own cups.
Image Credit: Starbucks

TECH
Self-Sovereign Identity Solutions
Growing online interactions dramatically
increase the data people share with third par-
ties. While centralized providers like Google
and Facebook make it easy to sign in with one
click, this enables tracking and puts con-
sumer data at significant risk. In response,
organizations like The Linux Foundation, W3C,
and the Decentralized Identity Foundation
have developed software standards that allow
users to manage access to their data.
Self-sovereign identity (SSI) digitizes iden-
tity, giving data owners control of their
digital identities, enabling access to digital
ecosystems, and giving users control over
what details of their identity are shared with
different parties. While SSI has no set stan-
dard, it typically combines distributed ledger
technology and cryptography with verifiable
credentials and decentralized identifiers. SSI
could reduce data breaches, as companies no
longer need to store personal information.
Players like J.P. Morgan, Workday, and Micro-
soft have decentralized ID projects, but adop-
tion has been slow, likely because of the lack
private blockchain networks and that which
is underway in public blockchain networks.
On private blockchains, companies like Citi-
group and Goldman Sachs are experiment-
ing with tokenized digital securities mar-
kets to facilitate faster and more efficient
financial services. On the public blockchain,
US government bonds have been tokenized—
Maker, Tether, and USDC use US treasuries
as collateral.
Another area of development is in the tokeni-
zation of physical collectibles. In September,
Arcade.xyz, a protocol specializing in peer-
to-peer loans, facilitated a $1.1 million loan
where an NFT of a Supreme T-shirt collection
was used as collateral. Many prominent fig-
ures and consulting firms see the tokeniza-
tion of assets as a potential multitrillion-dol-
lar market for blockchain technology.
of profit incentives and consumer visibility.
However, US and EU government agencies
are offering grants and installing mandates
to allow decentralized identity solutions to
be built into consumer-facing wallets. In the
EU, decentralized identity wallets are ex-
pected to launch as part of its 2030 Digital
Decade program next year.
Tokenization of TradFi, Digital Assets
Security Tokens
The tokenization of real-world assets—se-
curity tokens—has been a bright spot in
blockchain development. So far, security
tokens have had limited implementation.
One exception is Japan, which already has a
thriving security token market focused on
tokenizing corporate bonds or real estate–
backed securities.
Growth is on the horizon: More organizations
are adopting security tokens, more juris-
dictions offer regulatory clarity, and bond
market yields are increasing. The industry
is experiencing two trajectories of security
tokens: tokenization that occurs in closed or
WEB3
Microsoft launched Entra Verified ID in early 2024
to include Face Check, a facial matching feature.
Image credit: Microsoft

TECH
Credibility Scoring and Anonymity
DeFi lending is not capital efficient: Most
DeFi loans are overcollateralized. To fix that
balance, lending companies are turning to
conventional means of credit risk scoring us-
ing on-chain data and nontraditional metrics
focusing on reputational scoring of addresses
based on DeFi activity.
Companies like Spectral pull on-chain data
for accounts and use AI algorithms to an-
alyze lending, borrowing, and other history
and output a credit score. Other protocols
like Taraxa and Cred Protocol use a mixture of
on-chain and off-chain reputational metrics
to enhance score outputs. DeFi is also using
off-chain credit scores, like those reported by
TransUnion. The company’s partnership with
Web3 companies Quadrata and Spring Labs
allows users to port their credit scores direct-
ly into DeFi applications.
While credit scores improve capital efficiency,
attaching scores to blockchain reduces user
anonymity, a core tenant of the technology.
The range of impact depends on the solution:
for their lack of fun. These days, blockchain
advancements in scalability and usability
have enabled developers to create AA and
AAA titles. Game studios such as Star Atlas,
Gala Games, Bright Star, and others are deep
into developing such games.
Traditional gaming companies have hesi-
tated to adopt blockchain. Still, companies
like Zynga and Ubisoft are experimenting
with on-chain games—most notably Zynga’s
“Sugartown.” It’s built on Ethereum and just
dropped an NFT collection tied to the project.
Zynga has a history of producing high-qual-
ity hits, and the game’s success could pave
the way for users’ increased familiarity with
on-chain assets, tokens, and staking. This
could in turn lead to increased Web3 integra-
tion and game development in the future.
On-chain credit calculations have a lower
risk, but blending on- and off-chain data will
at a minimum result in pseudonyms. Off-
chain credit scores will tie directly to indi-
vidual identities. This could be a significant
deterrent for crypto natives for whom ano-
nymity is crucial. Still, for the next wave of
adopters who are used to standard identity
verification, it may have less impact.
On-Chain Gaming: Play, Own, Earn Enjoy
On-chain games incorporate blockchain tech-
nology, ranging from fully on-chain games to
those that only have digital assets on-chain.
Fully on-chain games have all the benefits
of blockchain, including alternative fund-
ing, community development, decentralized
serverless development, player ownership,
composability, and player-driven economies.
While the benefits of on-chain gaming are
attractive, game deployments are subject
to current capacity constraints of public
blockchains. This has resulted in the release
of initially slow turn-based games involving
trading cards or battles that were criticized
WEB3
Sugartown is a Web3 gaming platform created
by Zynga.
Image Credit: Zynga

SCENARIOS
SCENARIO YEAR 2032
What If We Used AI To Monetize Privacy?
Are you tired of feeling like your personal data is out of your control? Well now you can rest easy with DataSentinel. This game-changing protec-
tion in digital privacy and data control is powered by AI, and purpose-built to keep you safe. Imagine a world where you are the master of your
data. With DataSentinel, that world is now a reality. DataSentinel’s cutting-edge self-sovereign identity solutions keep your private data stored
securely and decentralized—meaning you, and only you, have the keys to your digital kingdom.
Here’s the best part: With DataSentinel, you can monetize your data. That’s right, turn your data into dollars! DataSentinel’s AI-powered system
optimizes your data, making it valuable for companies willing to pay for your insights. It’s time your data started working for you!
The security on DataSentinel is top-notch. Thanks to blockchain technology, your personal information is safe and sound. Plus, you have the
power to grant or revoke access as you see fit. It’s like having a digital bodyguard. Gone are the days when companies could exploit your data
without your consent. With DataSentinel, you forge a new path in the digital world, one where privacy isn’t just a feature—it’s a right.
Let’s not forget about convenience! DataSentinel models are like personal assistants for your digital life, handling everything with the utmost
confidentiality and efficiency. In compliance with the strictest data privacy regulations, DataSentinel ensures you’re always in the driver’s seat
of your digital identity. So, if you’re ready to take control of your digital life, sign up for DataSentinel and join the digital revolution. Your data will
thank you!
TECH WEB3

AUTHORS
CONTRIBUTORS
WEB3
TECH

WEB3
TECH
MELANIE SUBIN
Managing Director
Melanie Subin is Managing
Director of Future Today Insti-
tute, where she serves on our
management committee and
leads our consulting division.
Renowned for her pragmatic, forward-thinking approach, Melanie
has successfully steered numerous clients towards future-ready
strategies, harnessing emerging trends and technologies to identify
risk and opportunity early enough for action. Her leadership has
significantly impacted how industries envision and execute their
long-term strategies.
Melanie specializes in strategic transformation, quantitative and
qualitative research, and scenario development. With deep exper-
tise in the development and establishment of foresight capabilities
within large organizations, Melanie regularly counsels C-staff on
strategy and execution. She has spent years assessing the impact
of major external forces such as increasing technological sophis-
tication, changing consumer and business preferences, and rising
connectivity on the evolution and transformation of industries and
markets across the globe.
Melanie is a recognized expert in fostering psychological safe-
ty within teams, a crucial element for operationalizing strategic
foresight effectively. Her work emphasizes creating an environment
where open dialogue and innovative thinking are encouraged,
enabling organizations to embrace change and navigate future
uncertainties with confidence.
Melanie serves in the World Economic Forum’s Metaverse Work-
ing Group and is a founding member of the Dubai Future Forum’s
advisory group. She serves as a coach in the strategic foresight MBA
course at the NYU Stern School of Business. Melanie holds a BS in
Finance from Central Connecticut State University and a Fintech
Certification from the Massachusetts Institute of Technology.

GRANT TINKER
Associate
Grant Tinker served as a Strategic Foresight Consultant at Future
Today Institute following his graduation from NYU Stern School of
Business, where he earned an MBA with specializations in strategy,
entrepreneurship innovation, and marketing. He brings a strong
financial background as a Chartered Financial Analyst (CFA) and
his multiple years of corporate finance experience from SaaS companies in Silicon Valley to his work.
In addition to his MBA, Grant holds a Bachelor of Science in Economics from California Polytechnic
State University.
WEB3
TECH
AMY WEBB
Creative Director
EMILY CAUFIELD
Editor
ERICA PETERSON
Copy Editor
SARAH JOHNSON
CHERYL COONEY

SELECTED
SOURCES
TECH WEB3

“A Gentle Introduction: Orbit
Chains.” Arbitrum Docs, Decem-
ber 7, 2023, docs.arbitrum.io/
launch-orbit-chain/orbit-gen-
tle-introduction.
Amadán, Dylan. “Why Ai Needs
Zero-Knowledge Proofs.” Me-
dium, Verum Capital Insights,
April 4, 2023. medium.com/
verum-capital-insights/why-ai-
needs-zero-knowledge-proofs-
957ec72627b9.
Asher, Mattison. “Zero-Knowl-
edge Proofs: Starks vs Snarks.”
Consensys, May 18, 2021. con-
sensys.io/blog/zero-knowledge-
proofs-starks-vs-snarks.
Awosika, Emmanuel. “Account
Abstraction: Past, Present, Fu-
ture.” MetaMask, April 25, 2023.
metamask.io/news/latest/
account-abstraction-past-pres-
ent-future.
Banon, Justin. “The Trillion
Dollar Crypto Opportunity: Real
World Asset Tokenization.”
CoinDesk, September 6, 2023.
www.coindesk.com/consen-
sus-magazine/2023/08/31/
the-trillion-dollar-crypto-oppor-
tunity-real-world-asset-tokeni-
zation/.
Beckett, Alex. “Modular vs. Mono-
lithic: A Beginner’s Guide.” Celes-
tia Blog, November 9, 2022. blog.
celestia.org/modular-vs-mono-
lithic-a-beginners-guide/.
Bedawala, Mustafa. “Monolithic
vs. Modular Blockchain.” Visa,
usa.visa.com/solutions/crypto/
monolithic-vs-modular-block-
chain.html.
Behnke, Rob. “Blockchain’s
Storage Problem Is Growing. Are
There Any Solutions?” Web3 and
Blockchain Security Solutions,
November 3, 2022. www.halborn.
com/blog/post/blockchains-
storage-problem-is-growing-are-
there-any-solutions.
Berkowitz, Bram. “Visa and Mas-
terCard Have Embraced Block-
chain Technology. It’s a Smart
Long-Term Strategy.” The Motley
Fool, February 16, 2023. www.
fool.com/investing/2023/02/16/
visa-and-mastercard-have-em-
braced-blockchain/.
Bochan, Toby. “Pudgy Penguins
NFT Project, Once Endangered,
Proves Web3 Turnaround Is
Possible.” CoinDesk, 23 May
2023, www.coindesk.com/
web3/2023/05/25/pudgy-pen-
guins-nft-project-once-endan-
gered-proves-web3-turnaround-
is-possible.
Brooks, Miles. “The 10 Best Cryp-
to Learn and Earn Platforms in
2023.” CoinLedger, September
15, 2023. coinledger.io/tools/
learn-and-earn-crypto.
Butterfield, Sean. “What Is Ac-
count Abstraction and Why Is It
Important?” Crypto Council for
Innovation, December 13, 2023.
cryptoforinnovation.org/what-is-
account-abstraction-and-why-
is-it-important/.
Chen, Alec. “Modular Block-
chains: A Deep Dive.” Volt
Capital, July 7, 2023. volt.capital/
blog/modular-blockchains.
Chow, Katelyn. “Tokenizing
Identity: Friend.Tech’s Explo-
sive Rise and Controversies.”
Metatalks, August 28, 2023.
www.metatalks.ai/tokeniz-
ing-identity-friend-techs-explo-
sive-rise-and-controversies/.
Copeland, Tim. “‘Airdrop Farmers’
Tell Us How They Make Millions.”
The Block, April 6, 2023. www.
theblock.co/post/225215/we-
made-close-to-1-million-inside-
the-murky-world-of-airdrop-
farming.
Cryptopedia Staff. “Consen-
sus Algorithm/Consensus
Mechanism Innovations.”
Gemini, March 10, 2023. www.
gemini.com/cryptopedia/
blockchain-consensus-mecha-
nism-types-of-algorithm.
“DAOs and DeFi.” Deepdao, deep-
dao.io/organizations. Accessed
September 7, 2023.
“Decentralizing Base with the
OP Stack and Optimism.” Base,
Coinbase, February 23, 2023.
base.mirror.xyz/H_KPwV31M7O-
JT-THUnU7wYjOF16Sy7aWvaEr-
5cgHi8I.
EigenLayer. “Intro to Eigenda:
Hyperscale Data Availability
for Rollups.” EigenLayer Blog,
October 19, 2023. www.blog.
eigenlayer.xyz/intro-to-eigen-
da-hyperscale-data-availabili-
ty-for-rollups/.
Fay, Mike. “Base: Assessing
Coinbase’s New L2 Network.”
Seeking Alpha, August 22,
2023. seekingalpha.com/
article/4630199-base-assess-
ing-coinbases-new-l2-network.
Floyd, Matthew L., et al. Systems
and Methods for Maintaining a
Distributed Ledger Pertaining to
Autonomous Vehicles. November
21, 2023.
Forta Network. “The Future of
Threat Prevention in WEB3.”
June 5, 2023, forta.org/blog/
the-future-of-threat-prevention-
in-web3/.
Gehmlich, Ben. “Ethereum
Blockchain Security; Pros and
Cons of Pos Solutions.” Gigster,
September 27, 023, gigster.com/
blog/pros-and-cons-of-pos-for-
ethereum-security/.
Gkritsi, Eliza. “Microsoft, Aptos
Labs Team up on New Block-
chain AI Tools.” CoinDesk,
August 11, 2023. www.coindesk.
com/web3/2023/08/09/micro-
soft-and-aptos-labs-partner-on-
new-ai-blockchain-solutions/.
“HACK3D The Web3 Security
Quarterly Report—Q1 2023.” Cer-
tik.com, April 6, 2023. https://
www.certik.com/resources/
blog/3BaCA6ytR6uLFc1JV-
vt313-hack3d-the-web3-securi-
ty-quarterly-report-q1-2023.
“HACK3D The Web3 Security
Quarterly Report—Q2 2023.”
Certik.com, July 5, 2023.
www.certik.com/resources/
blog/6KJQKBrVJEwHR6Tc4PAv-
Jo-hack3d-the-web3-securi-
ty-quarterly-report-q2-2023.
Hall, Phil. “Zynga Announces
Web3 Game Sugartown.” Play
to Earn, August 18, 2023. www.
playtoearn.online/2023/08/18/
zynga-announc-
es-web3-game-sugartown/.
Hoffman, David. “David’s
Takeaways from Ethcc 2023 on
SELECTED SOURCES
TECH WEB3

Bankless.” Bankless, July 26,
2023. www.bankless.com/da-
vid-ethcc-2023.
Hunt, James. “November saw
$343 million lost to crypto hacks
and fraud cases: Immunefi.”
November 30, 2023. The Block.
https://www.theblock.co/
post/265437/november-saw-
343-million-lost-to-crypto-hacks-
and-fraud-cases-immunefi
Inabinet, Jack. “Tokenized Trea-
suries and an RWA Revolution
on Bankless.” Bankless, Sep-
tember 6, 2023. www.bank-
less.com/tokenized-treasur-
ies-and-an-rwa-revolution.
Irrera, Anna. “PayPal Launch-
es US Dollar Stablecoin PY-
USD for Payments, Transfers.”
Bloomberg, August 7, 2023.
www.bloomberg.com/news/arti-
cles/2023-08-07/paypal-launch-
es-us-dollar-stablecoin-py-
usd-for-payments-transfers.
Isaacson, Thomas M. “Pat-
ents at the Trivergence Com-
bining Blockchain, Artificial
Intelligence, and the Internet of
Things.” Blockchain Research
Institute, 2022. https://www.
blockchainresearchinstitute.
org/project/patents-at-the-triv-
ergence/te
John, Joel. “Grants WEB3.”
April 7, 2022. www.decentralised.
co/p/grants-and-web3.
Joshi, Shashank, and Arhan
Choudhury. “Tokenization of Real
Estate Assets Using Block-
chain.” IJIIT vol.18, no.3 2022:
pp.1-12. http://doi.org/10.4018/
IJIIT.309588.
Khun, Eric. “NFT Misconception:
JPEG Aren’t on the Blockchain.”
Eric Khun, October 31, 2021. er-
ickhun.com/posts/nft-miscon-
ception-image-arent-on-block-
chains/.
Kufuor, Kofi. “The State of Crypto
Security.” Medium, The Control,
October 7, 2022. thecontrol.co/
the-state-of-crypto-security-
d628ac5b609d.
Kumar, Ronak. “Lido Chooses
Axelar and Neutron for WSTETH
Launch on Cosmos.” The Crypto
Times, September 15, 2023.
www.cryptotimes.io/lido-choos-
es-axelar-and-neutron-for-
wsteth-launch-on-cosmos.
Ledesma, Lyllah. “SEC’s Binance,
Coinbase Suits Create Uncer-
tain Future for Listed Tokens:
Legal Experts.” CoinDesk, June
13, 2023. www.coindesk.com/
markets/2023/06/12/secs-bi-
nance-coinbase-suits-create-
uncertain-future-for-listed-to-
kens-legal-experts/.
Levin, Michael, and Olaoluwa
Osuntokun. “AI for All: Powering
Apis and Large Language Models
with Lightning.” Lightning Labs,
July 6, 2023. lightning.engineer-
ing/posts/2023-07-05-l402-
langchain/.
Lindrea, Brayden. “CTFC Cracks
Down on DEFI Protocols Opyn,
Zeroex and Deridex.” Cointele-
graph, September 7, 2023.
cointelegraph.com/news/ctfc-is-
sues-order-against-defi-proto-
cols-opyn-zeroex-deridex.
Liu, Bessie. “Why Data Availability
Sampling Matters for Blockchain
Scaling.” Blockworks, 23 August
2023. blockworks.co/news/da-
ta-availability-ethereum.
Mirpuri, Jash. “Unraveling the
Mantle Network: A Deep Dive into
the New Layer-2 Scaling Solution
for Ethereum.” Flagship.FYI, July
2, 2023. flagship.fyi/outposts/
blockchains/mantle-network-
deep-dive-into-the-new-layer-2-
scaling-solution-for-ethereum/.
Mitchelhill, Tom. “AAVE Proto-
col Launches Stablecoin Gho
on Ethereum Mainnet, $2M
Minted.” Cointelegraph, July 17,
2023. cointelegraph.com/news/
aave-protocol-launch-stable-
coin-gho-ethereum-mainnet.
Mondello, Antonino, and Alberto
Troia. “Cryptography-based
Platooning Mechanism for Au-
tonomous Vehicle Fleet Manage-
ment.” US20210319702A1.
Nagarajan, Shalini. “Borrowing
Crypto with Defi? Transunion
Shares Real-World Credit Scores.”
Blockworks, April 21, 2023. block-
works.co/news/transunion-cryp-
to-defi-credit-scores.
“OP Stack Components.” Opti-
mism Docs, December 11, 2023.
docs.optimism.io/stack/compo-
nents.
Overgang, Alexandra. “What Is
Filecoin? The Leading Decentral-
ized Storage Network.” Cointele-
graph. cointelegraph.com/learn/
what-is-filecoin-the-leading-de-
centralized-storage-network.
Owie, Best. “Study Says You’re Bet-
ter off Dumping Your Airdropped
Tokens.” Bitcoinist.com, March 11,
2022. bitcoinist.com/study-says-
youre-better-off-dumping-your-
airdropped-tokens/.
Parkhouse, Arthur. “A Look at
LVMH’s Blockchain Consortium.”
Hypebeast, August 17, 2022. hy-
pebeast.com/2022/8/lvmh-au-
ra-blockchain-luxury-fashion.
Peterson, Macauley. “Maker to
Spark ‘New Explosion of Dai’
with Custom Lending Mar-
ket.” Blockworks, May 8, 2023.
blockworks.co/news/mak-
ers-spark-protocol-dai.
Richards, Daniel. “The Rising
Efficiency—and Falling Costs—of
Data Storage.” Impax Asset Man-
agement, August 17, 2023. im-
paxam.com/insights-and-news/
blog/the-rising-efficiency-and-
falling-costs-of-data-storage.
Sandor, Krisztian. “Tokenized
U.S. Treasurys Surpass $600M
as Crypto Investors Capture
Tradfi Yield.” CoinDesk, July 11,
2023. www.coindesk.com/mar-
kets/2023/07/11/tokenized-us-
treasurys-surpass-600m-as-cryp-
to-investors-capture-tradfi-yield/.
“Securitize Japan Announces
Business Alliance with Sony Bank
and Sumitomo Mitsui Trust Bank
on Landmark Tokenized Joint
Money Trust.” Securitize, July 3,
2023. securitize.io/press-releas-
es/securitize-japan-announc-
es-business-alliance-with-so-
ny-bank-and-sumitomo-mit-
sui-trust-bank-on-landmark-to-
kenized-joint-money-trust.
SELECTED SOURCES
TECH WEB3

Sensei, Degen. “The Case for and
Against the Appchain Thesis.”
The Aspiring Degen, October 16,
2022. degensensei.substack.
com/p/the-case-for-and-
against-the-appchain.
Staff, Decrypt. “These Are the
High-Profile Spot Bitcoin ETF
Applications Currently in Play.”
Decrypt, December 7, 2023. de-
crypt.co/62912/high-profile-bit-
coin-etf-applications.
Strack, Ben. “Race for a Spot
Ether ETF Begins with Bid from
Ark Invest, 21Shares.” Block-
works, September 6, 2023.
blockworks.co/news/21shares-
ark-filing-for-spot-ether-etf.
Strack, Ben. “TRADFI, DEFI Con-
vergence Continues Through
Tokenizing Real-World As-
sets.” Blockworks, August 31,
2023. blockworks.co/news/
real-world-assets-tradfi-defi-to-
kenization.
Team, Chainalysis. “Curve
Finance Pools Exploited Due to
Code Vulnerabilities.” Chainaly-
sis, September 19, 2023.
www.chainalysis.com/
blog/curve-finance-liquidi-
ty-pool-hack/.
Thapa, Ronal. “Aptos vs Sui Block-
chain: Similarities and Differenc-
es.” Boxmining, February 21, 2023.
boxmining.com/aptos-vs-sui/.
“The Crunchbase Web3 Track-
er.” Crunchbase News, April 28,
2023. news.crunchbase.com/
web3-startups-investors/.
“The Rug Pull Report.” Certik,
March 14, 2023. www.certik.com/
resources/blog/5Zai85KkvIgW-
0wBJPQxl3i-the-rug-pull-report.
Thomas, Langston. “NFT Tickets
Are the Future of Live Events.
Here’s Why.”NFT Now, March 30,
2023. nftnow.com/features/nft-
tickets-are-the-future-of-live-
music/.
Thomas, Stanley. “The Role of
Liquidity Providers in Decentral-
ized Finance (DEFI)!” Medium,
DataDrivenInvestor, April 24,
2023. medium.datadrivenin-
vestor.com/the-role-of-liquidi-
ty-providers-in-decentralized-fi-
nance-defi-d5af679aac2f.
Time, Big. “Big Time: Airdrop
Program Explained.” Medium,
PlayBigTime, January 19, 2023.
medium.com/playbigtime/
big-time-airdrop-program-ex-
plained-ed92fb3dd7d2.
“Tokenized Treasuries.” Rwa.Xyz.
Accessed November 1, 2023. app.
rwa.xyz/treasuries.
Tropee Team. “8 Best Benefits of
Nfts in 2023.” Tropee, February 23,
2023. www.tropee.com/blog/8-
best-benefits-of-nfts-in-2023.
“Understanding Tokenized US
Treasury Bills.” Conduit Guides.
Accessed December 7, 2023.
www.conduit.fi/guides/under-
standing-tokenized-us-trea-
sury-bills.
Vishu. “What Is Crypto Airdrop
and How You Can Earn Money
through It.” Binance, April 16,
2023. www.binance.com/en/
feed/post/426221.
Wagner, Casey. “Stoner Cats
Agrees to Pay $1M Fine to Settle
SEC Charges.” Blockworks, Sep-
tember 13, 2023.
blockworks.co/news/ston-
er-cats-sec-unregistered-securi-
ties-nfts-hollywood.
Weiss, Ben. “A Brief History of Ze-
ro-Knowledge Proofs, the Buzzy
Mathematical Technique That’s
Taken Crypto by Storm.” For-
tune, June 6, 2023. fortune.com/
crypto/2023/06/05/zero-knowl-
edge-proofs-history-zk-rol-
lups-cryptography-zcash/.
“What Is Zora? How Does It
Work?” OpeanSea, August 22,
2023. opensea.io/learn/block-
chain/what-is-the-zora-network.
Wodehouse, Carey. “What Will
Blockchain Mean for Data Stor-
age?” Pure Storage Blog, August
18, 2023. blog.purestorage.com/
perspectives/what-will-block-
chain-mean-for-data-storage/.
Wood, Catherine, et al. “Why
Bitcoin Is the Currency for AIt.”
Ark Invest, September 7, 2023.
ark-invest.com/podcast/bitcoin-
brainstorm-02-why-bitcoin-is-
the-currency-for-ai/.
Wood, Chris. “Publicis Groupe
Joins Human vs. Ai Content
Verification Initiative C2pa.”
MarTech, June 5, 2023. martech.
org/publicis-groupe-joins-hu-
man-vs-ai-content-verifica-
tion-initiative-c2pa/.
Woodmansee, Benjamin. “AI
and Blockchain Infrastructure
Joyrides: Can AI and Block-
chain Germinate?.” Medium,
April 28, 2022. medium.com/@
benwoodmansee/ai-blockchain-
infrastructure-joyrides-can-
ai-and-blockchain-germinate-
april-2021-d937a1dd21fc.
Worldcoin Team, and dcbuilder.
eth. “An Introduction to Ze-
ro-Knowledge Machine Learning
(ZKML).” Worldcoin Blog, Febru-
ary 22, 2023. worldcoin.org/blog/
engineering/intro-to-zkml.
Wu, Eva. “The Financialization
of Fun: Crypto Gaming Thesis.”
Mechanism Capital, January 14,
2022. www.mechanism.capital/
crypto-gaming-thesis/.
Xu, Tammy. “AI Watermarking
Locks down Unauthorized Use of
Datasets.” IEEE Spectrum, May
1, 2023. spectrum.ieee.org/water-
mark-ai.
Y. Li, M. Zhu, X. Yang, Y. Jiang,
T. Wei and S.-T. Xia, “Black-Box
Dataset Ownership Verification
via Backdoor Watermarking,”
in IEEE Transactions on Infor-
mation Forensics and Security,
vol. 18, pp. 2318-2332, 2023, doi:
10.1109/TIFS.2023.3265535.
zkSync. “Introducing the ZK
Stack.” Medium, Matter Labs,
September 6, 2023. blog.mat-
ter-labs.io/introducing-the-zk-
stack-c24240c2532a.
SELECTED SOURCES
TECH WEB3

METAVERSE • NEW REALITIES

TECH
TABLE OF CONTENTS
208 Top Headlines
209 State of Play
210 Key Events
Developments
212 Why Metaverse New
Realities Trends Matter
213 When Will Metaverse New
Realities Trends Impact Your
Organization?
To Consider
216 Central Themes
218 Ones To Watch
219 Important Terms
221 Metaverse Form Factor
222 Headsets
222 Smart Glasses
223 Haptic Wearables
232 Connected Well-being and VR
Assisted Therapy
232 Medical Metaverse
233 Scenario: Exploring Mars with
Man’s Best Friend
234 Education in the Metaverse
234 Synthetic Media in Hollywood
234 Forensic AR / VR
235 Metaverse-Enhanced Science
236 Scenario: Experiential
Equations
237 Psychosocial Dynamics and
Inclusivity in the Metaverse
238 Situated VR
238 The Panopticon
238 XR Accessibility
239 Diminished Sensory Overload
239 Cybersickness
240 Scenario: “Visiting” Pregnancy
241 Experiencing Immersive
Worlds
242 World Building on Blockchain
242 Events in the Metaverse
242 AR Lenses and Filters
243 Holograms
243 Real Estate in the Metaverse
244 Worlds for Purpose
244 Worlds for the Enterprise
245 Play-to-Earn and Virtual
Marketplaces
246 Scenario: Cosplay Category
Announcement
247 Metaverse Infrastructure
248 Interoperability
248 Government Investment
248 Developer Tools and
Application Building Blocks
249 Interdevice Synchronization
249 5G for the Metaverse
250 Authors
223 Voice, Gesture, and Neural
Interfaces
224 Senses in the Metaverse
224 Movement in the Metaverse
225 Scenario: A House Divided
226 Digital Identity
227 Avatars
227 Avatar Portability
227 Hyperrealistic Avatars
228 Fragmentation of Virtual
Identity
228 Leasing identity
229 Synthetic Speech
229 Synthetic Personalities
230 Applications
231 Virtual Training for Real
World Jobs
231 Industrial True-to-Reality
Simulations and Digital Twins
231 Human Digital Twins

The metaverse
sees a cooldown
from its initial hype,
marking a phase
of recalibration
and more realistic
expectations.
TOP HEADLINES 01
02
03
04
05
Meta and Apple Impress with New VR Hand Tracking
and Gesture Recognition
The tech giants’ latest headsets showcase hand tracking and gesture recognition
capabilities offering users a more intuitive way to interact with virtual environments.
Events in the Metaverse Flop
Metaverse Fashion Week saw major brands invest but drew few visitors, who found it
lonely and dull. Substantial investment and creativity are needed to boost engagement.
Metaverse Layoffs: Meta and Disney Scale Back Amid Cooling Hype
As the initial excitement around the metaverse wanes, Meta reduced its workforce, and
Disney shut down its division dedicated to exploring metaverse opportunities, signaling
a reset in the industry’s approach.
Deepfakes Emerge as a Concern Ahead of Election Cycle
While some are entertainment, the technology’s potential for fabricating hyperrealistic
news and political footage to deceive voters is alarming ahead of upcoming elections.
Experts worry viral deepfakes could spread misinformation across social media.
Early Adopters Explore Digital Twins for the Enterprise
Innovative early adopters in sectors like autonomous vehicles and smart cities
are looking into the potential of digital twins.
TECH METAVERSE NEW REALITIES

STATE
OF PLAY
When Facebook rebranded as Meta in 2021, it bet big on leading the next computing platform—the
metaverse. However, early criticisms emerged, given the rudimentary state of virtual and augment-
ed reality (AR) technology at the time, with many clunky headsets and cartoonish avatars. The spot-
light soon shifted when OpenAI unveiled ChatGPT in 2022, captivating public interest in AI. Unlike
past AI systems, ChatGPT’s accessibility enabled mainstream user interaction for the first time,
representing a key inflection point. This overshadowed enthusiasm for the metaverse, which was
having adoption challenges. Disney retreated on its own metaverse goals amid disillusionment,
and Meta’s Reality Labs posted a staggering $13.7 billion loss in 2022.
Yet in 2023, major tech players recognized the potential in fusing AI and metaverse technologies
to take immersive simulations to the next level. Meta CEO Mark Zuckerberg unveiled strikingly
realistic virtual avatars, enabled by AI—a major leap in simulation quality. Microsoft shared plans to
blend its AI Copilot with HoloLens 2, delivering an AR experience for workers. And Apple introduced
Vision Pro, a spatial computing headset employing machine learning for more natural user interac-
tions. Rather than competing trends, AI and the metaverse now appear poised to complement each
other. The future points to AI-enabled metaverse experiences that feel increasingly personalized by
learning user preferences.
Meta also unveiled more subtle technologies like smart glasses, signaling a shift in consumer
preferences for extended reality (XR) technologies that integrate digital experiences with physical
reality. In doing so, these technologies could enhance human connection, as the technical interface
fades into the background, replacing screen-centric interactions with more natural, human-cen-
tered ones. Spaces and the nature of work could be transformed, reducing our reliance on tradition-
al screens and keyboards, and potentially revolutionizing the built environment and our interac-
tions with computers. The form factor of XR technology could evolve to where our natural actions
and environments become the primary means of interaction, ushering in a new era of computing
that is more integrated with daily life. This evolution mirrors AI advancements in natural language,
enabling more intuitive metaverse interactions.
In many ways, the trajectories of AI and the metaverse have come full circle. What began as sepa-
rate trends with muted enthusiasm has given way to recognition of their interdependence in creat-
ing more immersive digital interactions. Their futures are now fundamentally linked, with advances
in one enabling progress in the other.
Overshadowed by the rise
of artificial intelligence, the
metaverse is searching for new
potential through AI integration,
evolving from initial hype to a
measured and mature future.
INDUSTRY
209

FEBRUARY 1, 2023
A Long-Term Vision is Challenged
Meta’s Reality Labs division, specializing
in AR and VR technologies, reports
significant operating losses of $13.7
billion for 2022.
SEPTEMBER 28, 2023
Meta Town Hall Highlights AR
Mark Zuckerberg showcases Ray-Ban smart
glasses as key to Meta’s subtle, stylish AR shift.
Shortly after, Meta Quest 3 debuts, offering
advanced mixed reality experiences but
garners mixed reviews and slow adoption.
NOVEMBER 15, 2023
Metaverse Medical Market
Ocutrx’s OcuLenz AR headset, designed for
macular degeneration, highlights medical
applications as a key path for AR technology
adoption beyond entertainment.
OCTOBER 23, 2023
Saudi Cities Embrace Digital Twins
South Korea’s Naver wins a $100 million
deal to create digital twins for five Saudi
cities, enhancing urban planning and flood
management.
FEBRUARY 2, 2024
Apple Vision Pro is Released
Apple released its newest mixed
reality headset with over 600 apps,
video passthrough, and spatial audio.
KEY EVENTS

THE FOUNDATIONS OF THE METAVERSE
As virtual experiences become more
embedded into daily life, we can expect
growing pains. Regulators must balance
guidelines that ensure integrity without
restricting an nascent industry. Tapid
advances precipitate more seamless,
intuitive user experiences. Hands-free
environments feel within reach as inno-
vations like gesture control tech remove
friction. Past the hype, enterprise use cas-
es will gain traction, and efficiency gains
will drive adoption beyond novelty appeal.
Generative AI will further accelerate pro-
liferation by enabling amateur creators to
construct fully realized 3D worlds without
coding skills. With increased adoption,
interoperability will rise as a priority.
Metaverse technology has perpetually
searched for a wide audience—if it fails to
reach the mainstream consumer, indus-
trial use cases may take over. Medical,
industrial, or civic metaverses may be
the first to find product-market fit. Their
focused nature could lend itself more
readily to creating utility and value.
Completely Hands-Free Experience
As delivery routes and last-mile deliveries
continue to increase in speed and complexity,
automation will let logistics companies cre-
ate an intricate web of delivery offerings that
can be unique to each customer.
Extended Reality Gets Serious
With workers continuing to grow in scarcity,
virtual agents will take over back-of-house
work in the warehouse. These virtual agents
will soon be able to oversee themselves and
their cobot workers, reducing the need for
human intervention.
Metaverse-Tailored Offerings
As climate change continues to be a disrup-
tion, logistics providers will explore how they
can insure against extreme weather events.
These costs could be passed along to con-
sumers who choose goods from more volatile
regions.
Metaverse World Building for All
Increased automation and use of virtual
agents raise the need for workers to know how
to manage new tech-enabled work and tasks.
This upskilling could be done through remote
learning and working opportunities.
Regulators Target Virtual Trading
With manufacturing locations moving closer
to the consumer and e-commerce increasing,
manufacturers need to consider how to create
products in the exact spot as their consumers.
Consumer Push for Interoperability
Global conflicts, combined with consumers
and businesses wary of supporting govern-
ments with values antithetical to their own
will increase demand for verification of supply
chains. Granular data collection and transmis-
sion will enable this shift in transparency.
Media
Telecom
Distribution

Hyperrealistic avatars will
enable more personalized
brand experiences, but
consumer identity may
fragment across multiple
avatars tailored to
different contexts. This
complicates marketing
analysis, as brands must
derive insights from
fragmented consumer
journeys and understand
core motivations
persisting across digital
identities.
Digital twins will
revolutionize enterprise
operations by enabling
virtual prototyping,
testing, and optimizing.
Companies can digitally
iterate designs and
stress test ideas for
better quality, gaining
a competitive edge in
market speed while
minimizing downtime
and disruption. By saving
money and managing
risk, this shift to virtual
environments will drive
innovation.
Hyperrealistic avatars
offer brands detailed
customer insights but
risk misuse. They enable
lifelike interactions
and richer data while
supporting a deeper
understanding of
target demographics.
However, the potential for
deepfakes makes identity
validation a challenge and
puts brand reputations
at risk. The ability to trust
the accuracy of avatar
identities will be critical
for ensuring messaging
reaches intended
audiences.
Immersive virtual worlds
offer a new frontier for
brand marketing. In
the metaverse, brands
can engage customers
through lifelike,
personalized interactions,
blending products and
messages into the virtual
cultural fabric. This
narrative-based approach
boosts virality and peer
sharing, while rich data
informs strategy and
product development.
Substantial investment
in creative talent and
technology is essential
for establishing top-tier
experiences.
XR in enterprise
training rapidly upskills
employees’ safely, even
in high-risk areas.
Realistic simulations
for tasks like operating
machinery or handling
emergencies build skills
and confidence without
real-world risks. This
leads to a workforce
adept at complex jobs,
reducing mistakes. XR
training offers quicker
proficiency, lower risk,
and no need for physical
practice, optimizing
talent development and
shortening onboarding.
Augmented reality could
reduce reliance on screens,
minimizing screen fatigue
and enabling more natural
human-computer collabora-
tion. Spatial overlays could
also optimize workflows by
aligning virtual models with
physical spaces. As XR ad-
vances, traditional seated
desk work may transition
to immersive environments
centered on voice and ges-
tures rather than mouse
clicks. Built spaces could
be reimagined, reducing
fixed real estate costs and
promising more ergonom-
ic, satisfying, and dynamic
work routines with less
dependence on screens.
Avatars Enable
Personalization But
Fragment Identity
Digital Twins Save
Time, Money, and
Reduce Risk
The Dual Edge of
Hyperrealistic Avatars
Metaverse Unlocks
Immersive Marketing
Avenues
Safely Mastering
High-Risk Skills
Early XR Adoption
Boosts Workforce
Dynamics
WHY METAVERSE NEW REALITIES TRENDS MATTER TO YOUR ORGANIZATION

TECH
213
WHEN WILL THE METAVERSE NEW REALITIES DISRUPT YOUR ORGANIZATION?
Digital twins in manufacturing
Hands-free virtual
Photo-realistic avatars
Multifactor biometric sensing
Holograms
Enterprise XR collaboration
Spatial computing
Digital twins in health
Normalized use of avatars
in collaboration platforms
Decentralized identity
Spatial mapping
6G-enabled metaverse
Full interoperability

Threats
The lack of asset and avatar portability across platforms like Meta Horizon
or Microsoft Mesh limits the metaverse’s potential and user experience.
The inability to transfer digital assets across different metaverse platforms
creates a fragmented and restrictive environment.
There is significant risk in prioritizing proprietary control, and those seeking
walled-garden dominance risk stunted growth. If standardized protocols are
not developed as they were for the internet, the metaverse’s vision may not be
actualized and it will be a flop.
The advancement of synthetic media has given rise to substantial ethical
challenges, particularly around the issue of consent. Both public figures and
private individuals face the prospect of their likenesses being replicated
without permission, prompting serious concerns about autonomy and
control over digital identities.
The creation of digital replicas raises philosophical questions about the
authenticity of human expression and the ethical implications of creating
and interacting with digital beings that mimic real individuals.
The metaverse could exacerbate isolation and detachment from reality. The
lack of physical human interaction could diminish the quality of real-world
relationships, leading to feelings of loneliness and disconnection.
Opportunities
Digital twins can serve as powerful tools for testing, optimizing, and making
more informed decisions in a controlled setting. They help with proactive
maintenance, predicting failures before they occur, and reducing downtime.
Metaverse marketing could bring brands closer to consumers and influencers
closer to their followers. It enables interactive spaces for community building
so users can interact not only with the brand but also with each other,
enhancing loyalty and engagement.
Virtual training offers efficient skill development in low-risk environments,
improving employee competencies and safety. Trainees can experience realistic
scenarios without the dangers associated with real-life training, and it reduces
the need for physical resources, travel, and on-site training facilities.
With companies integrating metaverse capabilities into platforms like Teams,
remote collaboration is becoming more effective. Enhanced AI translations and
the metaverse’s elimination of geographical constraints means global teams
can overcome language and distance barriers to access talent.
The companies that contribute most meaningfully to open standards
and standardized protocols will emerge as long-term winners. By building
compatibility across platforms, these companies can expand their reach,
enabling their products to seamlessly integrate with a range of metaverse
environments and tools.

Pursue novel brand interac-
tions. Rather than mirroring
real-world marketing in vir-
tual spaces, explore novel
modalities unique to these
mediums. Provide value
via exclusive metaverse
experiences that offer in-
timacy at scale. Stay atop
innovations in synthetic
media and immersive
technologies to remain
cutting-edge.
Get ahead of unintended
consequences. Monitor
early metaverse adopters
across functions for poten-
tial psycho-social impacts,
and implement support
systems proactively, like VR-
based counseling or hybrid
work policies. Navigate this
responsibility wisely.
Spearhead industry in-
teroperability standards.
Get ahead of fragmenta-
tion by convening an indus-
try consortium to align on
metaverse interoperability
standards early. Prioritize
user portability and plat-
form connectivity. Lead
here before ecosystems
splinter and momentum
makes alignment difficult.
Use immersive training for
risk mitigation. Leverage
XR for disaster prepared-
ness, hazardous environ-
ment rehearsals, and other
high-risk training needs.
Repeatable virtual drills
identify gaps efficiently at
lower risk and cost than live
exercises.
Differentiate with digital
previews. Build 3D virtual
models of spaces to allow
remote previews and walk-
throughs. For real estate,
this enables global proper-
ty tours. Hotels and venues
can offer virtual site visits
to event planners. Unique
visibility and accessibili-
ty become a competitive
edge.
Accelerate development
cycles. Construct digital
prototypes of products
early in the design process
for rapid iteration and
user testing in VR. Refine
designs, materials, and in-
teractions before physical
production. Democratize
innovation by extending
tools to customers.
1 4
2 5
3 6

CENTRAL THEMES
216
TECH
The Quest for Connection
There is a nuanced interplay between digital immer-
sion and the desire to maintain a connection with the
physical world. Consumers increasingly show a pref-
erence for augmented reality over virtual reality, indi-
cating a broader inclination toward digital experiences
that complement, rather than replace, the physical
environment. The social dimension of the metaverse
is equally critical. For the metaverse to be genuinely
transformative, it needs to enhance social connections
rather than create isolation. Its success will largely
depend on its ability to foster genuine human interac-
tion and community building. However, this new digital
frontier is also reshaping social dynamics in profound
ways. For instance, the increasing use of devices like
smart glasses, which can record interactions, intro-
duces a new dynamic in social behavior. People may
modify their behavior, knowing they could be recorded
at any time, affecting the authenticity of social ex-
changes. In essence, the metaverse is not just a tech-
nological space but a social one, heavily influenced by
human behavior and societal norms.
The Technological Symphony of the Metaverse
The metaverse stands as a technological apex, relying
on the culmination and integration of various ad-
vanced technologies still in development. Its realiza-
tion hinges on the progress of numerous fields, includ-
ing powerful battery technology, enhanced connectivity
and the widespread deployment of 5G, as well as the
evolution of wearable devices. This convergence sug-
gests that the idealize vision of the metaverse is not
just an isolated development but rather the endpoint
of numerous technological trajectories. These include
significant advancements in computing power, so-
phisticated AI algorithms, and innovations in the en-
tertainment sector, all of which are crucial in shaping
an immersive, seamless, and interactive metaverse ex-
perience. As such, the journey toward the fully realized
metaverse is as much about the progression of these
individual technologies as it is about their harmonious
integration.
AI: Not Parallel But Key to Actualizing
the Metaverse
AI is not just an accompanying technology in the devel-
opment of the metaverse; it is a critical driver making
the metaverse’s existence and functionality possible.
For instance, a key contribution of AI in the metaverse
is the enhancement of user interactions through ad-
vanced natural language processing. This technology
allows for more natural and intuitive communication
within virtual environments, making interactions with
AI-driven avatars and interfaces more fluid and life-
like. Moreover, AI facilitates the creation of expansive,
complex virtual worlds. It enables the rendering of
highly detailed and dynamic environments, which can
adapt and respond to user interactions in real time.
Additionally, AI contributes to the personalization of
experiences within the metaverse. By analyzing user
data and behavior, AI can tailor experiences to individ-
ual preferences. This customization capability will be
necessary for making the metaverse less like a generic
virtual space and one that feels uniquely relevant to
each user.

CENTRAL THEMES
217
TECH
The Invisible Interface
An overarching drive toward more natural interactions
underscores a key vision emerging for the metaverse.
For input, companies like Meta, Apple, and Snap are
developing innovations in voice control, hand ges-
tures, and even neural signals that interpret eye and
facial muscle movements. These updates remove the
need for traditional physical controls, aligning with
a frontier of silent, seamless interaction powered by
AI. Gesture technology is advancing rapidly, pointing
to hands-free experiences. Neural signals take this
even further by tapping directly into biological cues
to enable control through cognition alone. For output,
advances in haptics and multisensory technologies
point to a future beyond visuals and audio, where vir-
tual experiences feel increasingly tactile and lifelike.
The emergence of these subtle, intuitive technologies
reveals we are slowly starting to actualize one vision
of the metaverse: for the interface between humans
and technology to fade into the background, becoming
almost invisible.
Fragmented Realities
While individual companies are making strides devel-
oping their own metaverse platforms and capabilities,
true interoperability across these disparate ecosys-
tems remains elusive. Challenges persist around asset
portability and avatar interoperability across differ-
ent virtual platforms. A custom-built avatar remains
confined to a single walled metaverse garden like Meta
Horizon or Microsoft Mesh. Purchasing a virtual Gucci
bag in one world doesn’t automatically transfer digital-
ly to other environments. This fragmentation severely
limits the scope and potential of the metaverse. Users
wish to move fluidly between experiences, carrying
their digital possessions and personas with them.
Creators want to build once and deploy everywhere,
not redo work for each platform. For the metaverse
to mirror the connected nature of the physical world,
technical infrastructure enabling this cross-platform
continuity is critical. While metaverse pioneers are
acknowledging this necessity, competitive pressures
and commercial interests of tech giants currently take
precedence over open collaboration.
The Serious Business of the Metaverse
While consumer applications have driven much of
the initial enthusiasm for metaverse technologies,
substantial enterprise use cases are emerging across
domains like training, digital twins, and remote col-
laboration. These practical business applications have
the potential to push adoption further into the main-
stream. Immersive training simulations are already
being deployed by companies like Walmart, BMW, and
Lufthansa to develop employee skills and knowledge in
low-risk virtual environments. Industries from avia-
tion to medicine are utilizing digital twins of complex
systems to enable virtual testing and optimization. For
remote work, Microsoft is integrating metaverse capa-
bilities directly into Teams, which has over 270 million
users. This allows colleagues to collaborate via lifelike
avatars and spatial layouts in virtual meeting envi-
ronments. While startups are innovating, Microsoft’s
massive reach gives it potential to make VR meetings
a workplace norm. As these use cases demonstrate
concrete ROI in areas from employee productivity and
safety to time and cost savings, enterprise adoption
can help further validate and destigmatize metaverse
technologies.

Andrew “Boz” Bosworth, CFO and head of
Meta’s Reality Labs, for being Mark Zucker-
berg’s right-hand man and overseeing more
than 20,400 people in realizing Meta’s vision
for the metaverse.
Tamir Berliner and Tomer Kahan, co-founders
of Sightful Spacetop, for innovation in spatial
computing interfaces.
Tara Boroushaki, Laura Dodds, Aline Eid,
and Maisy Lam, researchers at MIT, for contri-
butions to augmented reality with non-line-of-
sight perception.
Dr. Tim Bunnell, director at the Nemours Cen-
ter for Pediatric Auditory and Speech Scienc-
es, for work on AI-generated video clips that
mimic accents and speech patterns.
Professor Garuda Fujii, researcher at Shin-
shu University’s Institute of Engineering and
ELab2, for innovative approaches to designing
source-shifter structures.
Grimes, musician and artist, for pioneering
new business models in AI-generated voice
synthesis.
Im Doo Jung, a professor in the Department of
Mechanical Engineering at UNIST, for work on
smart contact lenses for AR-based navigation.
Henry Liu, professor of civil engineering and
director of Mcity and the Center for Connected
and Automated Transportation at the Univer-
sity of Michigan, for advancements in simu-
lated driving environments.
Akash Nigam, CEO of Genies, for develop-
ment of decentralized avatar systems in the
metaverse.
Dr. Seung-Kwon Seol, researcher at the
Smart 3D Printing Research Team at Korea
Electrotechnology Research Institute, for work
on smart contact lenses for AR-based naviga-
tion.
Chat Steelberg, CEO of Veritone, for advance-
ments in synthetic Voice as a Service (VaaS)
solutions.
Dr. Yu Xinge, associate professor in the
Department of Biomedical Engineering at
City University of Hong Kong, for co-leading a
study on wireless olfactory feedback systems
in VR.
Dr. Yon Visell, associate professor of biolog-
ical engineering at University of California,
Santa Barbara, for work on haptic holography.
Joshua Xu, CEO and co-founder of HeyGen, for
innovations in translation services.
Dr. Mikael Benson, researcher at the De-
partment of Clinical Science, Intervention,
and Technology at Karolinska Institutet, for
research on digital twins for treatment of
inflammatory diseases.
Pouya Hamadanian, electrical engineering
and computer science graduate student and
lead author at MIT, for work on Ekho, which
synchronizes audio and visual streams
transmitting to two devices.
Elizabeth Haas, an adjunct professor at NYU
School of Professional Studies, founding
director at NYU SPS Emerging Technologies
Collaborative, and partner at New York Con-
sulting Partners, for writing about cities and
technologies.
Dr. Sarah E. MacPherson, head of psycholo-
gy and professor at the School of Philosophy,
Psychology, and Language Sciences, Univer-
sity of Edinburgh, for her work on cybersick-
ness in immersive digital reality.
Michael Barnett-Cowan, professor at the
Department of Kinesiology and Health Sci-
ences, University of Waterloo, for his work on
motion sickness in VR games.
Jose Fuertes, founder and CEO of OWO, for
developing haptic vests for virtual gaming.
Jake Rubin, founder, chairman, and CEO at
HaptX, for working to bring virtual worlds to
life through realistic touch.
Jensen Huang, CEO and president of Nvidia,
for envisioning the Omniverse platform and
developer ecosystem to build the industrial
metaverse and a clear path to metaverse
revenue.
Soo-yeon Choi, CEO of Naver, for instru-
mental work in developing the digital twin
cities project between South Korea and Saudi
Arabia, which will revolutionize real-time de-
cision-making, prediction, and optimization
of urban infrastructure.
Yacine Achiakh, CEO and founder of Wisear,
for pioneering the development of the first
earphones with a neural interface, enabling
hands-free, voice-free control of XR devices.
ONES TO WATCH

IMPORTANT TERMS
219
Augmented reality (AR)
A technology that overlays digital information,
images, and objects onto the real-world environ-
ment. Users see virtual elements mixed into their
actual surroundings through a device screen or AR
glasses/headset.
Avatar
A digital representation of a user, often in the form
of a 3D model or illustration. Avatars serve as a
user’s persona in online/virtual environments.
Cybersickness
Nausea or motion sickness experienced by some VR
users due to proprioception disorientation. It arises
from the mismatch between perceived and actual
spatial positions in VR, with research suggesting
that factors like vertical orientation perception and
inclusion of music can influence its severity.
Data portability
The ability for users to transfer their digital iden-
tities, including avatars, and associated data
between platforms and services.
Decentralization
A core principle shared by the metaverse and block-
chain technology, emphasizing an open network
Haptics
Technology related to tactile sensations and feed-
back. Can include vibration, motion, pressure, and
temperature changes.
Holography
A technique for creating three-dimensional
projections; it’s becoming key in populating the
metaverse with realistic avatars and environments,
and merging with technologies like deepfake for
various applications.
Human-machine interfaces
The components and methods through which hu-
mans interact with and control machines, like key-
boards, mice, touchscreens, and voice commands.
Hyperrealistic avatars
Highly detailed avatars that closely mimic a
person’s real facial features, expressions, and
movements through advanced 3D modeling and
scanning.
Interoperability
Blockchain’s capability allowing assets and infor-
mation to seamlessly transfer between different
worlds and platforms within the metaverse.
Mixed reality (MR)
A hybrid form of reality that merges the real and
virtual worlds to produce new environments and
visualizations where physical and digital objects
coexist and interact in real time.
Neural interfaces
Technologies that connect directly to the user’s
neural activity, like brain waves or facial muscle
signals, to enable hands-free and silent control.
Non-fungible tokens (NFTs)
Unique digital assets representing ownership of vir-
tual items like land and avatars in the metaverse,
made credible and secure through blockchain
technology.
Olfactory feedback
Technology that generates smells and aromas dig-
itally, allowing smells to be simulated in a virtual
environment.
Panopticon
A system of control where individuals are aware
they might be watched at any time, leading to
self-regulation of behavior. In the context of smart
glasses, it refers to the heightened sense of being
observed and changing behavior because of it.
of interconnected virtual worlds, as opposed to
closed, proprietary platforms.
Deepfakes
Manipulated video/audio that uses AI to realis-
tically substitute someone’s likeness and voice
in existing content without their consent, raising
ethical concerns.
Digital twins
Virtual replications of physical systems used for
simulation and optimization.
Experiential artifacts
Lingering sensory and cognitive effects in VR users,
blurring the lines between virtual and real-world
experiences. These artifacts result from the dis-
sonance between virtual and physical realities,
leading to feelings of disembodiment or altered
physical world perceptions.
Extended reality (XR)
An umbrella term that encompasses virtual reality,
augmented reality, and mixed reality. XR provides
immersive digital experiences that blend the
physical and virtual worlds across a spectrum of
realities. It enhances interactions with the environ-
ment and digital elements.

IMPORTANT TERMS
220
Passthrough
A feature in some headsets that uses outward-fac-
ing cameras to display the physical environment
to the user while wearing the headset. Provides
awareness of surroundings.
Play-to-earn games
Virtual environments in the metaverse where play-
ers can earn real-world value through gameplay,
with blockchain technology enabling the collection,
breeding, and trading of digital assets as NFTs.
Situated virtual reality (situated VR)
A concept proposed to align the physical and virtual
worlds, minimizing experiential artifacts. It focuses
on syncing physical actions with virtual feedback
to create a congruent reality, including mirroring
body language and emotional expressions in virtual
and real worlds.
Synthetic personalities
Fully artificial digital influencers and identities
generated through AI training, not tied to any spe-
cific human individual.
Synthetic speech
AI-generated simulated speech that clones a
person’s vocal characteristics to create natural
sounding vocalizations. Enables voice banking,
which benefits people who may lose their ability to
speak later in life.
Virtual reality (VR)
An artificial digital environment that is fully im-
mersive and isolates users from the physical world.
Users typically wear a headset with stereoscopic
displays and head tracking to look around the
virtual world.

METAVERSE
FORM FACTOR

METAVERSE FORM FACTOR
TECH
Headsets
While early virtual reality headsets offered
consumers an escapist diversion, the tech-
nology is maturing, and developers are now
targeting more pragmatic industries and
experiences. The Meta Quest 3, for instance,
brings a 30% improvement in screen reso-
lution, faster processing speed, and a sleek-
er design. Most notably, outward-facing
passthrough cameras allow wearers to view
both physical and virtual surroundings. This
“mixed reality” mitigates the isolating feel
of previous models. Apple’s forthcoming
Vision Pro is also set to incorporate similar
passthrough technology, highlighting an in-
dustry-wide acknowledgment of the need for
more interactive and less isolating VR expe-
riences. The Vision Pro will launch with an ex-
pansive app library spanning entertainment
to productivity. Major streaming services like
Disney+, ESPN, and Amazon Prime Video will
be available alongside work apps like Micro-
soft 365, Slack and Zoom.
Smart Glasses
Developments in smart glasses aim to
make spatial computing technology sub-
tle yet powerful in daily life. In a landmark
September 2023 announcement, Meta
collaborated with EssilorLuxottica to unveil
Ray-Ban glasses with built-in AI capabilities.
Resembling traditional Ray-Bans in design,
these glasses integrate multimodal sensors
to interpret the user’s gaze and voice com-
mands. Whether the wearer is curious about
a building or needing a sign translated, Me-
ta’s assistant can provide answers without
hand gestures. Snap plans similar AI inte-
gration for contextual recommendations and
on-lens edits. Apple likewise has plans to
introduce augmented reality glasses, though
details are still forthcoming. Meanwhile, Mi-
crosoft recently patented swappable batter-
ies for extended power, potentially enabling
comfortable all-day wear by reducing weight.
Offloading processing functions to con-
nected accessories like backpacks presents
another option for portable use.
Meta’s substantial price hike compared to
its predecessor indicates a strategic piv-
ot from targeting general consumers to
focusing on enterprise applications. This
repositioning places the headset in direct
competition with established enterprise-fo-
cused devices like Microsoft’s HoloLens 2
and Magic Leap 2. Sony’s version, the SonyXR
Headset is primarily tailored for industrial
applications, aiming to integrate various
production stages, such as design and
prototyping, into the metaverse. By allowing
users to construct 3D design models, the
headset helps spot and fix problems while
plans are still digital, saving money previ-
ously spent on faulty physical prototypes.
Rather than pure escapism, developers seem
concentrated on increasing VR/AR function-
ality across specialized fields moving for-
ward. Allowing real environment interaction
reflects acknowledgment of earlier issues
in consumer adoption, while enterprise and
sensory enhancement applications point to
an evolving market for the technology.
Though slimmer smart glasses would blend into
social settings better than bulky virtual reality
headsets, battery limitations pose ongoing chal-
lenges to unlocking the full capabilities of the
sleeker wearable technology.

TECH
In early 2024, BMW introduced AR technol-
ogy in their cars with the help of XrealAir 2
AR glasses. These glasses show navigation,
entertainment, and electric car charging
information directly to the driver. The intro-
duction of smart glasses in vehicles and into
other parts of our daily life is more than just
a technological advance; it’s a societal shift.
Mobile phones revolutionized the way we in-
teract with the world—pausing to record mo-
ments or look up information. Smart glasses
promise the same but with an added layer of
immersion: You can live stream your expe-
riences while remaining fully present. How-
ever, this comes with questions about the
implications for interpersonal interactions.
Will behavior change in a world where people
know they could be continually recorded or
analyzed by AI? In essence, smart glasses
aren’t merely a new gadget; they represent a
significant leap toward pervasive computing
and could fundamentally alter our relation-
ship with technology and each other.
tracking systems like Sony’s full body suit
loaded with sensors, the building blocks are
falling into place for the creation of deeply
immersive and tactile responsive spaces
within virtual worlds. The end goal is ambi-
tious yet attainable: to produce a metaverse
experience that fully engages users across
visual, auditory and, critically, tactile do-
mains to enable suspension of disbelief and
flow state immersion.
Voice, Gesture, and Neural Interfaces
Human-computer interaction continues
progressing beyond phones and screens,
leveraging modalities like voice and ges-
ture. Virtual assistants have normalized
conversational AI, powered by robust speech
recognition models. Startups are unveiling
experimental devices prioritizing intuitive
interactions, like Humane’s screenless
wearable pin, which understands natural
language requests. Meta’s Ray-Ban Stories
glasses allow hands-free voice control, while
their VR headsets track hand motions to
manipulate virtual objects. Apple’s Vision
Pro will combine subtle finger gesture rec-
Haptic Wearables
Virtual worlds to date have focused primarily
on visual and auditory immersion, but repli-
cating tactile sensations presents the next
frontier for technologies like metaverse plat-
forms to conquer. The ability to not just see a
virtual object but reach out and feel textures,
weights, and movements would provide
unprecedented realism. To enable this,
developers are experimenting with haptic
wearables—gloves, vests, or suits equipped
with actuators to simulate different sensa-
tions through vibration, electrical stimula-
tion, or even tiny inflatable balloons. Simple
vibrational alerts via eccentric rotating
mass motors are already featured in smart-
phones and controllers, priming adaptation
for games and 3D virtual environments.
Companies like HaptX are leveraging more
advanced pneumatic glove actuators to rep-
licate lifelike textures and shapes. Spanish
startup OWO recently unveiled a haptic vest
using electrical signals to induce sensations
including bullet impacts or stabbings for
gaming and live entertainment applications.
As these devices advance alongside motion
Just as touchscreens became widespread in the
last decade, voice and gesture control are poised
to be the next major methods for interacting with
computers.

TECH
ognition with eye tracking, enabling users to
simply look at and point to items they want to
select in an augmented environment. Wisear,
a French startup, is pushing the boundaries
even further with experimental smart ear-
buds that detect facial muscle movements
and bioelectrical brain signals. By interpret-
ing these neural signals, Wisear envisions
completely silent, hands-free control of devic-
es through minute gestures like tightening
the jaw. Just as touchscreens have become
ubiquitous over the past decade, emerg-
ing modalities like voice, eye tracking, and
brain-computer interfaces could fundamen-
tally transform how we engage with technolo-
gy in the next 10 years.
Senses in the Metaverse
Recent advancements in human-machine
interfaces are ushering in a new era of mul-
tisensory experiences in VR and AR. On the
haptic front, researchers at the University of
California, Santa Barbara have focused ul-
trasound waves to induce tactile sensations
from afar. This noncontact “haptic hologra-
phy” allows users to perceive and manipulate
Movement in the Metaverse
Disney has developed a new flooring tech-
nology called HoloTile that allows users to
walk freely in any direction without actually
moving from their fixed location. Described
as the “world’s first multi-person, omnidi-
rectional, modular, expandable treadmill
floor,” HoloTile uses advanced sensors and
motors to detect a person’s movement and
seamlessly shift modular floor sections to
keep them centered in place. Multiple people
can use the floor simultaneously without
the risk of colliding. In a demonstration
video, inventor and imagineer Larry Smoot
walks through a virtual environment using
a headset while the HoloTile floor adjusts
dynamically beneath him. This technology
opens up new possibilities for virtual reality
and augmented reality, as well as applica-
tions like interactive theater where actors
could explore a scene without constraints.
Disney isn’t alone in its effort to enable more
natural movement in the metaverse. Other
companies are exploring similar technolo-
gies for natural movement in virtual spaces.
Virtuix has developed a VR treadmill for indi-
vidual users, featuring a concave design and
special shoe covers for a realistic walking sen-
sation. Additionally, Freedom Technologies is
working on specialized shoes with AI-enabled
motorized treadmills in the soles, allowing
for unlimited virtual movement in a confined
physical space. These innovations represent
different approaches to enhancing virtual
mobility, and it’s still unclear which will dom-
inate the market. The choice may depend on
specific use cases and user preferences in VR
and AR experiences.
virtual objects. To deepen sensory immer-
sion without physical temperature changes,
University of Tsukuba researchers simulate
persistent cold sensations through gentle
air blasts. This exploits the body’s sensitivity
to rapid cooling for virtual weather experi-
ences. In odor transmission, Dr. Yu Xinge’s
team at Beihang University developed wire-
less olfactory feedback systems with minia-
turized odor generators. Integrating face-
masks and skin patches, these can simulate
environmental scents ranging from flowers
to smoke. Startup OVR makes headsets that
diffuse cartridge fragrances for personalized
aroma experiences. Advances also continue
in spatialized audio for lifelike acoustics.
Professor Garuda Fujii of Shinshu University
engineered structures that make sounds
that seem to originate from different loca-
tions. By tricking the ears’ spatial perception,
more convincing augmented soundscapes
become achievable. As these technologies
mature, they pave the way for unified multi-
sensory environments—where users not only
see and hear but feel, smell, and taste simu-
lated worlds for unprecedented presence.

SCENARIOS
SCENARIO YEAR 2030
A House Divided
The Komanduri family finds themselves in a dilemma: a “house divided” not by sports teams or smartphone prefer-
ences, but by their choice of AR/VR ecosystems. The eldest child, Aditya, is an avid fan of Apple’s AR/VR ecosystem,
boasting the latest Apple AR headset and an array of digital accessories. Meanwhile, the younger sibling, Priya, is
deeply immersed in the Meta universe, equipped with the newest Meta Quest and a collection of Meta-exclusive digi-
tal games and experiences. This division has led to more than just friendly sibling rivalry; it’s become a logistical and
financial headache for the parents. Planning family activities in the virtual realm is nearly impossible, as each child is
locked into their respective ecosystems, unable to interact or join the same digital spaces.
The financial strain is palpable. The family has to subscribe to two different “family plans” to accommodate both
ecosystems. What’s more, the digital goods and games purchased for Aditya cannot be passed down to Priya, negating
the possibility of “hand-me-downs” that would have been a cost-saving grace. This means doubling up on purchases
for similar experiences or content, a redundancy that’s both frustrating and expensive.
Birthday and holiday gifts have also become a challenge. A game or digital accessory that delights Aditya is incompat-
ible with Priya’s Meta setup, and vice versa. This has led to careful, sometimes stressful planning to ensure equity and
satisfaction for both children, further adding to the family’s expenses.

DIGITAL
IDENTITY

TECH
Avatars
Avatars have matured beyond static profile
images into multifaceted digital represen-
tations in virtual spaces. Initially serving
as basic profile pictures, modern avatars
manifest as interactive 3D entities capable
of conversation, environmental navigation,
and even simulated physical mannerisms.
As online activities shift from websites and
apps toward immersive extended reality (XR)
metaverse experiences, avatars are becoming
our primary digital personas—used for social
connections, commerce, and professional
meetings. Microsoft enables Teams users to
utilize 3D avatars for calls, animated by voice
cues to function sans webcams. And a 2023
podcast hosted in the metaverse demonstrat-
ed major upgrades in avatar realism, utilizing
lifelike models of Mark Zuckerberg and Lex
Fridman. This noticeable leap from previous
cartoonish renditions drew significant public
and investor interest by making broader
metaverse goals appear more attainable.
Companies like Genies are further expanding
the scope and utility of avatars by working
on open, decentralized systems that let users
es and actions. As such, the issue of avatar
portability is inextricably linked to debates
about data ownership. When an avatar can
cross platforms, there’s a tacit understand-
ing that its underlying data should also be
portable. But who owns that data, especially
as avatars become increasingly sophisticat-
ed and personalized, is a point of contention.
The issue is even more critical when consid-
ering that some metaverse companies aim
to become comprehensive platforms for a
wide range of activities, from reading news
to socializing to shopping. In such a scenar-
io, the platform that hosts your avatar could
have significant power and control over a
large swath of your personal data. So, while
Ready Player Me’s tools for avatar creation
and portability offer users the ability to
maintain a consistent digital identity across
the metaverse, they also open up broader
discussions about data ownership and por-
tability in virtual spaces.
Hyperrealistic Avatars
Hyperrealistic avatars leverage 3D graphics
and AI to produce nearly identical digital
clones of individuals—capturing intricate
facial details, expressions, and motions. Two
methods exist for generating these sophis-
ticated models. Companies like Doob utilize
full body scans in studios, comprehensively
recording the user’s physical form. Alterna-
tively, apps like Avatar SDK and itSeez3D
enable DIY facial scanning directly through
smartphone cameras. By taking a 360 degree
image of their face and shoulders, users can
craft impressively realistic avatar likenesses.
Integrating natural language processing and
vocal mimicry, these avatars graduate beyond
appearances to also simulate voices, produc-
ing holistic digital surrogates. One pioneering
example was during the recent taping of Lex
Fridman’s podcast with Mark Zuckerberg in
the metaverse. Utilizing Meta’s advanced “co-
dec avatars,” their uncannily realistic digital
doubles exhibited nuanced mannerisms and
conversations.
As solutions like HeyGen’s AI-powered Avatar
Clones push fidelity even further, lines blur
between actual and simulated realities. While
constructive applications await in gaming, en-
create not only their virtual selves but also
the worlds they inhabit. This development
opens up new opportunities for user engage-
ment and potentially impacts how busi-
nesses operate in virtual environments. As
avatars grow increasingly sophisticated in
mimicking human appearance, motion, and
interaction modalities, they reinforce their
status as our digital surrogates across the
internet’s burgeoning virtual frontiers.
Avatar Portability
Avatar portability in the metaverse, as facil-
itated by companies like Ready Player Me,
intersects with larger discussions about
data portability and ownership. Ready Player
Me and its partner, Koji, offer users the abili-
ty to create a unified digital identity that can
migrate across over 200 games and virtual
experiences. While this offers convenience
and customization, it raises questions about
who truly owns these digital identities. Your
avatar, after all, is a form of data. Whether
it’s a visual approximation of you based on
uploaded photos or a purely imaginative cre-
ation, the avatar is shaped by your preferenc-
DIGITAL IDENTITY

DIGITAL IDENTITY
TECH
terprise metaverses, and beyond, deep ethical
questions emerge around authentic digital
identity and behavior. Ultimately though, as
avatar technology continues maturing to
deliver hyperrealism, it will profoundly trans-
form how we represent ourselves and interact
in online spaces.
Fragmentation of Virtual Identity
Without a standardized, universal avatar
system, users on various digital platforms
create multiple online personas, each rep-
resenting distinct facets of the self or even
entirely fabricated alter-egos. This emerging
paradigm signals a seismic shift in concep-
tualizations of identity in the digital era. We
now face not just divergence between our
physical and virtual selves but the fragmen-
tation of singular legible personalities across
ever-proliferating online spheres. A glimpse of
this reality already manifests professionally.
An individual may use one avatar, precisely
modeled after their real-world appearance,
for Microsoft Teams meetings. But the same
person could adopt a fanciful, anonymized
persona for recreational gaming universes.
playing circumscribed roles on WhatsApp,
Messenger, and Instagram. And deepfake
technology already allows for CGI actor sub-
stitutions in film or ads without a physically
present cast. When Bruce Willis retired for
health reasons, his likeness still appeared in
a Russian commercial via a deepfake gener-
ated by an AI company. To address this issue,
new contract language around “simulation
rights” is beginning to appear, allowing for
the legal use of an actor’s synthetic likeness
in future productions. The music industry
is also tapping into this trend; artists can
now create songs using an AI-generated
voiceprint of musician Grimes, splitting the
royalties with her if she approves the collab-
oration. These early examples highlight the
emerging potential to monetize synthetic
celebrity beyond entertainment—anyone
could perhaps license out digital persona
rights for conversational AI, branding deals,
or experiential metaverse content. But thorny
questions around likeness consent and fair
compensation remain open, especially as
technology blurs lines between virtual repli-
cation and individual autonomy.
For marketers reliant on data analytics, such
compartmentalized self-representation
poses challenges. When consumers have
multiple discrete avatars, extracting useful
signals becomes far more complex. Which
identity should companies target for person-
alized advertising? How to accurately track
preferences when individuals act differently
across contexts? Ultimately, the fracturing
of singular legible identities into special-
ized avatars requires businesses to funda-
mentally rethink behavioral analysis and
personalization methodologies. As personas
multiply across the metaverse, understand-
ing users grows increasingly nuanced.
Leasing identity
AI is enabling digital replications of celeb-
rities without their active participation or
consent—an ethically murky development
as likeness rights remain undefined. The
2023 Hollywood writers’ strike spotlit con-
cerns around studios exploiting synthetic
acting indefinitely sans compensation. Meta
recently launched celebrity chatbots like vir-
tual Paris Hilton and Snoop Dogg alter egos,
AI is enabling the creation of digital replicas of indi-
viduals’ appearances and voices, opening avenues
for monetizing identities.

DIGITAL IDENTITY
TECH
Synthetic Speech
Synthetic speech leverages AI to digitally
mimic human voices, enabling myriad appli-
cations from accessibility tools to creative
media. With enough training data, models
can precisely replicate the unique vocal sig-
nature of any individual. Startups like HeyGen
offer translation services that render person-
al video recordings into foreign languages,
while retaining the original speaker’s voice.
Meta is also developing real-time speech
translation to break language barriers using
natural voice cloning. And new models need
as few as 50 sentences to build vocal profiles,
making synthesis more efficient and accessi-
ble. Samsung demoed a feature for its Bixby
assistant to verbally respond to calls in a
user’s synthesized voice if they are unable to
speak. Other applications of speech synthesis
include “voice banking,” where people at risk
of losing their ability to speak due to disease
can record samples that AI uses to synthe-
size their voice. This allows them to preserve
their unique vocal identity for text-to-speech
systems if they do need it later in life. Previ-
ously expensive and time-consuming, voice
supplement to human influencers. In 2023,
the Federal Trade Commission indicated that
virtual influencers must still disclose brand
sponsorships like their human counterparts.
As immersive spaces like the metaverse de-
velop, AI-driven synthetic personas could en-
able personalized interactive brand experi-
ences exceeding static posts. The technology
remains nascent—but rapid improvements
in AI personality exhibition point toward
virtual influencers, celebrities, and beyond
materializing as bona fide social presenc-
es, engineered from data to serve economic
aims over authentic self-expression.
banking has become more efficient and
affordable through AI, with some companies
only needing 50 sentences to create a digital
voice.
Synthetic Personalities
Beyond mimicking existing personas, large
language models (LLMs) can invent com-
pletely synthetic yet persuasive media iden-
tities from scratch. After ingesting enough
human data patterns, AI can reliably simu-
late varied personality dimensions within
text or embodied conversational agents. Re-
searchers deliberately shape these traits to
craft AI virtual influencers—novel social me-
dia personas like Lil Miquela, Noonoouri, and
Imma, each boasting 400,000+ Instagram
followers. Instead of cloning celebrities, their
personalities and backstories emerge fully
formed from algorithms to fulfill commercial
roles. Brands like Coinbase, Maje, and Tiffany
Co. have partnered with these AI-driven
virtual influencers for endorsements and
promotions. The control and presumed
brand safety offered by synthetic influenc-
ers makes them an attractive alternative or
Synthetic personalities are entirely artificial digital
influencers and identities created by AI, not linked
to any real human.

APPLICATIONS

TECH
Virtual Training for Real World Jobs
Virtual reality enables immersive job training
simulations across industries from retail to
medicine, with measurable improvements in
information retention and role comprehen-
sion. Companies like Walmart have incorpo-
rated VR into training as early as 2017, report-
ing 5%-10% testing gains versus classical
training. Simulations run the gamut from
customer scenarios like Black Friday to op-
erational skills like spill cleanups. Maryland
nonprofit Vehicles for Change likewise has
trainees first pick up VR goggles instead of
physical tools to observe demonstrations and
practice procedures before touching actual
cars. Aviation leaders including Lufthansa
also apply extended reality across domains
from product design to flight crew certifica-
tion. The technology provides a low-risk yet re-
alistic environment to build muscle memory
and mastery of complex tasks, from retail soft
skills to technical maintenance procedures.
Hands-on learning through lifelike simulation
before real-world performance allows workers
to avoid consequences as they develop confi-
dence. With customizable training exportable
domains from transportation to emergency
response. Numerous metropoles worldwide
have embarked on urban twin projects—from
Shanghai to Singapore to Chattanooga. The
city of Chattanooga itself collaborates with
research institutions on specialized twins
examining factors like energy-efficient mo-
bility infrastructure and pedestrian-vehicle
intersection patterns to inform planning.
As cloud computing power scales, digital
urban replicas grow increasingly high-fidel-
ity to run simulations and extract insights
unfeasible in the real world. The technology
enables observers to holistically visualize,
quantify, and optimize the intricate orches-
tration of modern cities.
Human Digital Twins
Research institutions are pioneering medi-
cal digital twins—detailed physiological sim-
ulations of individual patients for person-
alized care and quantitative analysis. These
complex computational models incorporate
genetic, molecular, and environmental fac-
tors to replicate disease mechanisms with
high precision. In conditions like rheumatoid
arthritis, Crohn’s, and ulcerative colitis, health
care teams can leverage twins to run clinical
scenarios—predicting outcomes of adjusted
drug regimens to tailor optimal treatments.
The simulations also enable deeper study
of biological drivers and responses in silico.
Medical twins diverge from simplistic avatars
by encapsulating dynamic physical detail
beyond just identities. Researchers ultimately
envision diagnostic, prognostic, and even two-
way communication functions as integration
with sensor data and AI intensifies.
However, as computing power expands, digital
twins may mature beyond niche medical uses
alone. More advanced systems could maintain
dynamic multifaceted models of individuals—
incorporating both medical and psychological
factors to mimic personality, knowledge, and
behaviors. Researchers envision such sophis-
tication may eventually enable twins to serve
as persistent, autonomous virtual agents for
their human counterparts. These highly faith-
ful digital doppelgängers could seamlessly
interact with people and other twins in virtual
spaces or the metaverse. They could collabo-
across geography, VR promises scalable and
measurable skill-building superior to static
manuals or lectures. Trainees engage more
senses in contextualized scenarios, enabling
organizations to elevate talent development.
Industrial True-to-Reality Simulations and
Digital Twins
True-to-reality digital twins are revolutioniz-
ing industries from autonomous vehicles to
smart cities by enabling virtual testing and
optimization of complex real-world systems.
The University of Michigan developed a
statistically accurate simulated roundabout
to rigorously refine self-driving algorithms
without physical risk. BMW likewise models
exact factory conditions in Nvidia’s Om-
niverse platform years before producing a
new car to optimize layouts and processes.
On a macroscale, digital urban twins facili-
tate data-driven planning and governance.
South Korean company Naver partnered with
Saudi Arabia to craft cloud-based digital
clones of Riyadh and other municipalities.
These foundations centralize infrastructure
data to assist long-term development across
APPLICATIONS

APPLICATIONS
TECH
rate in ways impossible for remote humans,
collectively analyzing problems through
continuous data sharing exceeding biological
cognition limits.
Connected Well-being and VR Assisted
Therapy
VR is increasingly being recognized as a pow-
erful tool in the field of psychological therapy
and mental health treatment. VR’s origins in
mental health date back to 1997, when it was
first invested in for treating PTSD in military
populations. One major application is virtual
reality exposure therapy (VRET): leveraging
simulated environments to gradually con-
front patients with anxiety disorders. VRET
allows for a controlled and personalized
intervention where patients can confront and
become accustomed to the sources of their
anxieties in a virtual setting. The treatment
has proven to be as effective as traditional
in-person exposure therapy for conditions like
specific phobia and agoraphobia with panic
disorder. Interventions for post-traumatic
stress disorder likewise show VR match-
ing traditional psychotherapy techniques
where students identify hazards and dis-
cuss observations with professors, allowing
for more effective training than traditional
methods. However, while AI tutors in VR
can provide tailored feedback like a human
instructor, human oversight is still needed.
A McGill University study found VR trainees
with additional human instruction caused
less tissue damage and were more precise
than those trained by AI tutoring alone. Plat-
forms like Fundamental Surgery additionally
provide sophisticated haptic feedback for
practicing complex manual tasks from pal-
pating tissue to maneuvering instruments.
Its patented HapticVR technology accurately
simulates the tactile sensations experienced
during surgery, from bone textures to muscle
and soft tissue interactions. FundamentalVR
also recently launched its Fundamental Core
SDK, a toolkit empowering developers to
create diverse medical training scenarios,
including multiuser VR experiences.
Beyond training, AR and VR are being used
for patient care and surgical planning. In
Cambridge, England, medical students use
for symptom relief. Beyond exposure, the
immersive medium also aids in develop-
ing coping mechanisms. Recent studies,
such as a pilot project by Stanford Medicine
researchers, have also started using VR to
address hoarding disorders. Study partici-
pants rehearsed giving up possessions in a
simulation of their own homes, an exercise
that helped them practice organizational
and decision-making skills while also de-
sensitizing them to the emotional distress
associated with discarding items. As soft-
ware and analytical dashboards improve, VR
appears poised to mainstream into mental
health—blending digital solutions with clin-
ical wisdom for more agile, quantified, and
personalized interventions.
Medical Metaverse
The metaverse enables transformative
innovations in medical education and care
delivery via immersive simulation. Medical
students can now perform virtual neurosur-
gery to improve their technical skills before
operating on actual patients. The University
of Texas uses a VR “patient safety room”
AR and VR are already utilized in surgical planning
and training, enabling surgeons and patients to
preview expected outcomes. At least one hospital
has used AR headsets during complex surgeries to
overlay patient data in real-time.

SCENARIOS
SCENARIO YEAR 2037
Exploring Mars with Man’s Best Friend
After strapping on the haptic suit and headset, I’m immersed in a stark reddish-orange alien landscape that can only be Mars. As I “step” forward, I feel the
crunch of strange soil beneath metal feet that are not my own. This body is an extension of mine—its sensors connected directly to my nervous system via a neu-
ral link system. When I lift my foot, the robot lifts its tread. When I reach out my hand, its claw extends. My consciousness inhabits this machine shell millions of
miles away, and I’m seeing Mars directly through my avatar’s camera eyes. Every sensation is mapped from its tactile sensors to my own synapses, blurred with
just enough latency to make the illusion complete. This robot is me, and through it I have stepped onto the surface of another world for the first time.
But what truly anchors me in this extraordinary experience is Moe, my dog. She’s not just lying next to me in the physical world; she’s also here with me on Mars,
in a way. When I adopted Moe, she came with something special: a digital twin. This isn’t a virtual pet; it’s a precise digital clone, created from detailed scans and
biometrics, designed to mimic every physical detail of Moe. Initially, these digital twins were meant for health monitoring, a technological advancement in pet
care. But soon, people realized their comforting potential in strange, digital realms. Here on Mars, as I navigate through vast, sweeping vistas, Moe’s digital twin
is right beside my avatar. As I explore, I occasionally reach down to pet the real Moe, feeling her warm fur, and then I see her digital doppelgänger reacting similar-
ly beside my avatar. It’s a bizarre yet heartwarming experience to have both versions of my best friend with me as I explore this alien world.
As night falls on Olympus Mons, I gaze out transfixed with my loyal dog clone by my side at a view no earthling has witnessed firsthand. Her presence, just like
the real Moe, comforts me—two versions of man’s best friend, one analog, one digital, both equally enthralled by this alien world we get to explore together.

TECH
mixed reality headsets to train on “hologram
patients,” while Swiss company Arbrea Labs
employs AR and 3D simulation to allow both
surgeons and patients to preview expected
outcomes of plastic surgeries, from nose jobs
to breast augmentations. Surgeons at insti-
tutions like Houston’s MD Anderson Cancer
Center utilize AR headsets to overlay patient
data seamlessly during complex procedures—
establishing a new paradigm of in-situ surgi-
cal guidance.
Education in the Metaverse
Research has indicated the efficacy of VR
in enriching learning experiences. Recent
findings suggest VR contributes to improved
understanding, heightened attention spans,
and inclusive access to digitized global cur-
riculum exceeding geographical constraints.
Real-world implementations demonstrate
scalability too. Japan’s N and S high schools
immerse over 6,000 students in collaborative
virtual classes using Meta Quest headsets.
Morehouse College’s VR chemistry labs yield
higher average test scores than conventional
modalities. Beyond supplemental content,
without restriction. This move clearly brings
into focus the ongoing debate about public-
ity rights, the legal concept that grants indi-
viduals the right to control the commercial
use of their name and likeness.
At the same time, examples of synthetic me-
dia featuring well-known actors are growing.
For instance, James Earl Jones authorized
the use of his iconic Darth Vader voice so AI
could generate it for future “Star Wars” films.
Companies like Metaphysic AI are employ-
ing de-aging technology, allowing stars like
Harrison Ford, Tom Hanks, and Robin Wright
to appear as younger versions of themselves
on screen. In 2022, after a 40-year hiatus,
ABBA made a comeback with fresh music,
accompanied by their de-aged 3D avatars,
known as ABBAtars. In 2023, Kiss bid fare-
well to the stage at their final performance
in Madison Square Garden—only to reemerge
minutes later as digital avatar versions of
themselves, suggesting virtual immortality
for these personas powered. Platforms like
Weverse give K-pop artists opportunities
to directly engage followers through virtual
meetups while monetizing exclusive digital
content. As immersive media expands, ex-
isting regulations around reasonable com-
pensation, consent, and protections demand
modernization to address ethical dimensions
related to identity replication via emerging
technologies.
Forensic AR / VR
AR and VR show great potential in advancing
forensics. Sophisticated AR/VR crime scene
simulations facilitate analysis even when the
physical location remains inaccessible after
the fact. Some systems use machine learning
and medical imaging to enhance AR-assisted
autopsies, leading to more accurate victim
identification and determinations of cause
of death. A major advantage of AR/VR is the
ability to visualize complex forensic data in
new ways. AR overlays digital information
directly onto physical crime scenes, allowing
for dynamic interaction like real-time tagging
of evidence and voice-recorded annotations. In
the legal system, AR and VR are modernizing
evidence presentation in courtrooms through
3D modeling and mapping to provide an
interactive platforms like RoybiVerse and
Nanome reconstruct traditional models
for more intuitive handling. Learners can
manipulate molecular structures with their
hands or traverse inside dinosaur anatomy.
By blending immersion with interactivity, VR
edtech unlocks multisensory comprehen-
sion of abstract or ephemeral concepts at
individual scale.
Synthetic Media in Hollywood
The rise of synthetic media in Hollywood, ac-
celerated by advancements in AI, is bringing
both opportunity and ethical dilemmas to
the industry. The 2023 Screen Actors Guild
strike revolved around usage of talent like-
nesses without clear restrictions. Ironically,
during the strikes, Meta and a company
called Realeyes capitalized on the abun-
dance of out-of-work actors by hiring them
for an “emotion study” aimed at making
AI-generated avatars appear more human.
These actors signed away extensive rights
“in perpetuity,” allowing their facial expres-
sions and other characteristics to be used
by Realeyes, Meta, and third parties almost
APPLICATIONS

TECH
immersive crime scene experience for judg-
es and jurors. This boosts understanding of
complex evidence sequences. These tech-
nologies also facilitate remote testimony for
improved accessibility. On the training side,
companies like CBF Forensics use realistic VR
simulations to deliver cost-effective forensic
skills training for law enforcement. While still
emerging, AR and VR integration in forensics
promises to enhance remote capabilities,
evidence presentation, and training.
Metaverse-Enhanced Science
The metaverse could significantly enhance
the effectiveness and reach of science. Its
interactive, three-dimensional spaces pro-
vide opportunities for unprecedented levels
of collaboration and accessibility in research.
For example, digital replicas of physical labs
can enable scientists around the globe to get
together and discuss and advance projects,
bypassing geographic and logistical con-
straints. Experiments conducted in virtual
environments have the added benefit of being
precisely replicated, improving the reproduc-
ibility of research findings. Agencies like the
Centers for Disease Control and Prevention
are already leveraging the metaverse to
train scientists in various locations, allow-
ing them to engage in experiential learning
without the real-world risks. Moreover, the
metaverse allows for the creation of entire-
ly new kinds of experimental spaces. For
instance, scientists could utilize existing
data and images to develop virtual models
of far-off places, such as Mars, and remotely
engage with these environments. Howev-
er, realizing the metaverse’s full promise
requires surmounting adoption obstacles
around equipment costs and centralized
tech giant control.
APPLICATIONS
Digital lab replicas could allow scientists to collaborate and perform virtual experiments mirroring
real-world conditions, reducing costs and increasing accessibility to experimentation.

SCENARIOS
SCENARIO YEAR 2027
Experiential Equations
Daniel slips on a headset, eager to explore the immersive calculus simulator his teacher introduced in class. As the
virtual world loads, Daniel finds himself standing on an abstract grid landscape, with colorful curves undulating in all
directions. He reaches for a squiggly purple function in front of him, knowing that in this virtual world he can physical-
ly interact with the fundamental building blocks of mathematics.
“Today I want you to explore the concept of curvature,” the instructor’s voice emanates from the air. “Take this control-
ler and manipulate the curve in front of you. Get an intuition for how bending changes along the arc.” Daniel reaches
out and takes hold of the squiggly purple function. As he moves his hands, the curve reshapes itself like virtual taffy.
Daniel instantly understands the curve better than equations on paper could convey—he feels how tapered regions
have lower curvature than tight curls.
“Now, let’s step into a derivative’s shoes,” the instructor says. The world blinks, and Daniel’s perspective shrinks. He
has become the orange derivative denoting the curve’s slope! As he glides along the ripples of the function, Daniel
physically experiences each peak and valley in its slope and their full spatial relationship clicks intuitively. The im-
mersive manipulations unlocked conceptual knowledge allowing math to finally make sense. He had felt what those
cryptic formulas tried explaining: that curvature captures the rate of direction change. VR let Daniel enter into mathe-
matics, instead of just staring formulaically. Equations had become experience.

PSYCHOSOCIAL
DYNAMICS
INCLUSIVITY IN
THE METAVERSE

TECH
Situated VR
In the metaverse, the dissonance between
virtual and real-world experiences often leads
to “experiential artifacts,” a phenomenon
where VR users experience lingering senso-
ry and cognitive effects that blur the lines
between these two realities. This issue arises
from the noncongruent realities that users
encounter when transitioning between digital
and physical spaces. Users may feel a surreal
detachment from their bodies or environ-
ment, reporting feelings of disembodiment or
altered perceptions of the physical world after
VR sessions. These experiential artifacts are
a result of the complex interplay between our
sensory inputs and past interactions with the
world, evolving as we are exposed to various
contexts, including virtual environments. Rec-
ognizing the psychological impacts of these
artifacts, which can range from mild curi-
osity to disorientations affecting daily life.
Researchers at MIT have proposed “situated
virtual reality” as a solution. This concept,
currently under review for publication in IEEE,
aims to align the physical and virtual worlds,
minimizing experiential artifacts. Situated
interacting with someone wearing smart
glasses, people may act differently when
they experience a heightened sense of being
observed. This phenomenon is similar to the
observer effect in psychology, where indi-
viduals modify their behavior in response to
their awareness of being watched. In social
contexts, this can lead to increased self-con-
sciousness, anxiety, and potentially altered
social dynamics.
The introduction of “trust lights” on some
smart glasses, designed to indicate when re-
cording is taking place, attempts to mitigate
this effect by providing a visual cue to others
about their privacy status. On the Ray-Ban
Meta smart glasses for instance, the cam-
era will not record if the LED light is covered.
However, the effectiveness of these trust
lights in alleviating concerns about privacy
and constant surveillance is not clear-cut.
While they might offer some reassurance,
the underlying tension and psychological
impact of potential constant recording re-
main.
XR Accessibility
Accessibility XR is a growing area of focus,
aiming to ensure these immersive technolo-
gies are inclusive for all users, including those
with disabilities. Key areas of accessibility
in XR include sensory, physical, cognitive,
and universal design aspects. For example,
Google’s ARCore improves environmental
understanding, crucial for users with visual
impairments. The design of XR interfaces is
evolving to cater to diverse physical abilities,
incorporating features like voice commands
and eye tracking. Microsoft’s inclusive design
approach in VR and AR, featuring voice recog-
nition and adaptable controller settings, ex-
emplifies this advancement. Similarly, many
companies now offer voiceover support in AR
experiences, aiding visually impaired users
with audio descriptions. Cognitive accessibil-
ity is also a priority, particularly for users with
conditions like autism, necessitating simpli-
fied interfaces and controlled sensory inputs.
Meta Quest demonstrates this by providing
customizable controls and head tracking op-
tions for users with limited mobility.
VR focuses on syncing physical actions with
virtual feedback, such as matching natural
movement or object interactions, to create
a more congruent reality. The approach also
extends to social interactions, ensuring that
the body language and emotional expres-
sions in virtual environments mirror those
in the real world. The goal is to establish a
hybrid reality that is free from experiential
artifacts, prioritizing not just technologi-
cal innovation but also the psychosocial
well-being of users in the metaverse.
The Panopticon
The psychological effects of being constantly
filmed, particularly when interacting with
individuals wearing smart glasses, can be
significant and multifaceted. Smart glasses,
which may be always recording or perceived
as such, introduce a dynamic similar to the
concept of a panopticon in social interac-
tions. The panopticon refers to a system of
control where individuals are aware that
they might be watched at any time, leading
to self-regulation of behavior due to the
perception of constant surveillance. When
PSYCHOSOCIAL DYNAMICS AND INCLUSIVITY IN THE METAVERSE

TECH
Many efforts in the XR accessibility space are
underpinned by the principles of universal de-
sign, aiming to create XR experiences that are
inclusive and enjoyable for everyone, regard-
less of their abilities. Leading this charge are
initiatives like the XR Association’s Accessi-
bility Working Group and the XR Access Initia-
tive, which bring together industry, academia,
and advocacy groups to research, develop,
and educate about accessible XR technol-
ogies, ensuring that the immersive digital
world is open and welcoming to all.
Diminished Sensory Overload
People with autism spectrum disorder (ASD)
often struggle to process sensory informa-
tion, finding stimuli like bright lights or loud
noises particularly overwhelming. This senso-
ry sensitivity can turn routine activities into
significant challenges. Augmented reality
presents a promising tool to mitigate these
sensory integration issues. AR’s ability to cre-
ate controlled and customized environments
can gently introduce sensory inputs to those
with ASD, allowing them to adapt at a comfort-
able pace. Among the innovations in this field
Cybersickness
Motion sickness and nausea remain signifi-
cant challenges in the mass adoption of vir-
tual reality, both for consumer and industrial
applications. One key issue is proprioception
disorientation, which arises when there’s
a mismatch between where you perceive
your limbs to be and their actual spatial
position in the virtual environment. Recent
research led by the University of Waterloo
sheds light on why some individuals are
more susceptible to VR-induced “cybersick-
ness” than others. The study found that the
perception of vertical orientation could shift
after engaging in high-intensity VR experi-
ences. This sensory adjustment significantly
influences the severity of cybersickness. The
findings could help VR developers create
more comfortable and adaptable experi-
ences by understanding the relationship
between sensory reweighting and cyber-
sickness susceptibility. In another study,
researchers found that incorporating music
into VR experiences can mitigate symptoms
of cybersickness. Both joyful and calming
music were shown to reduce the intensity of
nausea-related symptoms, with joyful music
having a particularly significant impact on re-
ducing the overall intensity of cybersickness.
These research developments offer promising
avenues for reducing barriers to VR adoption,
potentially leading to more personalized and
enjoyable virtual experiences for users.
are “Unfear” and “Floreo,” which use technol-
ogy to create supportive spaces for individ-
uals with ASD. Samsung’s Unfear is an app
that uses real-time selective noise filtering
to reduce auditory stress. It targets specific
sounds that are distressing to the user, of-
fering a more relaxing and fear-free auditory
experience. Unfear also extends its function-
ality to digital content, like mobile games, fil-
tering out loud sounds to make such media
more accessible for those sensitive to audio
intensity. Though not an AR application per
se, Unfear embodies principles that could be
integrated into AR platforms. Floreo, on the
other hand, directly utilizes AR to replicate
real-life situations in a controlled, less over-
whelming manner. It’s particularly beneficial
for teaching social, communicative, and
practical life skills in a secure environment.
Floreo’s simulated interactions offer a prac-
tice platform for users with autism, helping
them develop skills transferable to everyday
life. Both Unfear and Floreo exemplify how ad-
justing sensory inputs and offering flexible
learning environments can be revolutionary
for those with ASD.
PSYCHOSOCIAL DYNAMICS AND INCLUSIVITY IN THE METAVERSE

SCENARIOS
SCENARIO YEAR 2037
“Visiting” Pregnancy
Due to her age, Lauren was unable to safely carry her pregnancy herself. Instead, her daughter was growing in an artificial womb at a specialized facility. The high-
tech womb was calibrated to Lauren’s own heartbeat and voice via a monitor that she wore at all times, surrounding the growing baby with the familiar, comforting
sounds of its mother. The artificial amniotic fluid was even infused with Lauren’s natural scent through a filtering process.
Today, Lauren is preparing to “visit” her pregnancy. She dons a state-of-the-art haptic suit, designed to simulate the physical sensations of pregnancy. The suit gen-
tly expands around her abdomen, mimicking the gentle pressure and movements of a growing baby. As the virtual nursery flickers into view, Lauren gasps. It was a
line-for-line re-creation of the room her husband Eli was assembling at home, down to the plush rainbow rug and decaled quotes from her favorite children’s books.
As Lauren settles into the metaverse experience, the suit adjusts snugly around her midsection, gently applying pressure to simulate the swell of pregnancy. She
runs her hands along the warm, taut skin, marveling at how real it feels. Lauren feels a sudden thump and lays her hand on her belly. Her daughter is kicking up a
storm today.
The haptic suit is just one part of the experience. The headset she wears is pivotal, not just visually, but also in altering Lauren’s sense of smell to mirror the acute
olfactory sensitivity typical in pregnancy. Scents are intensified, creating a vivid and authentic sensory experience. Furthermore, the headset induces psychologi-
cal states akin to those of pregnancy, fostering a deep emotional bond between Lauren and her unborn child.
As Lauren walks through this virtual world, she speaks softly, knowing that her baby, though miles away in its artificial womb, can hear her. She sings lullabies,
tells stories, and shares her hopes and dreams for their future together. Each word, each note of her song, is transmitted back to the womb, enveloping her baby in
a warm embrace of mother’s love.

EXPERIENCING
IMMERSIVE
WORLDS

TECH
World Building on Blockchain
Advocates of the metaverse and blockchain
share a core principle: decentralization. In
the metaverse, this manifests as an open
network of interconnected virtual worlds as
opposed to closed, proprietary platforms.
Some advocates argue that when built on
blockchain, the metaverse can offer true
digital ownership through NFTs, by allowing
users to own, trade, and monetize virtual
assets like land and avatars. Blockchain’s
transparent and immutable record-keeping
lends credibility to digital ownership claims,
while its interoperability allows assets and
user profiles to carry seamlessly across
metaverse environments. Examples like The
Sandbox demonstrate this in practice—users
can build, own, and sell their game creations
through a dynamic, collective ecosystem.
Other platforms like Hyperfy and Voxels, both
running on the Ethereum blockchain, provide
tools for users to design, construct, and trade
entire worlds they own. This fusion of block-
chain and the metaverse not only democra-
tizes virtual space but also enables a more
trustworthy and interconnected digital reality.
to feelings of solitude, the complexity of
navigation, and a deficiency in compelling
content—revealing a stark contrast between
the high ambitions of brands and the actual
user experience. The concept of a metaverse
festival has also been embraced by Coachel-
la 2023, in partnership with “Fortnite.” This
collaboration introduced a new dimension to
music festivals, combining iconic Coachella
elements with the interactive and boundless
nature of the “Fortnite” universe. Weverse
has become a central hub for virtual K-pop
events, where fans can engage with their
favorite idols’ avatars in a digital concert
setting. This virtual space not only hosts
concerts but also fosters fan interactions,
proving that the metaverse can indeed
re-create the vibrancy of live events in a
digital format. As these digital spaces evolve,
the focus must shift toward user experience
to ensure the metaverse can fully realize its
potential as a platform for communal and
interactive events. Converting isolated soli-
tary activities into bonded social occasions
remains contingent on platform maturation.
AR Lenses and Filters
AR lenses overlay real-time digital informa-
tion onto the physical world. Simple AR filters
and lenses have long been a staple of social
media, enabling users to add digital enhance-
ments to their faces in real time. But these
playful features are just the beginning; AR
technology is becoming more sophisticat-
ed by leveraging generative AI. Snapchat’s
Cosmic Lens feature is a case in point, for
using generative AI to transform the user’s
environment into an animated cosmic back-
drop. Filters like TikTok’s “Bold Glamour” echo
this trend, offering hyperrealistic effects with
the help of AI. Furthering the integration of AR
into daily life, Snapchat is now streamlining
the user experience by introducing intelli-
gent recommendations for lenses based on
environmental context—such as the weather
or the time of day—using advanced visual
recognition and API integration. Snap has also
announced plans to introduce AR mirrors in
retail stores, allowing customers to see how
clothes would look on them in Nike stores and
in Men’s Wearhouse locations without the
While the idealized vision of the metaverse
and blockchain both emphasize decentral-
ization, it’s worth noting that many plat-
forms, like Meta’s proposed metaverse, may
operate as closed ecosystems.
Events in the Metaverse
The metaverse shows promise as the next
frontier for virtual experiences by creat-
ing immersive 3D environments mirroring
real-world settings. Gaming, an early adopt-
er of shared digital spaces, has expanded
its virtual horizons beyond gameplay into
events and social interactions. Games like
“Roblox” and “Fortnite” have been pioneers,
transforming their platforms into concert
venues for top-tier artists, creating a new
paradigm for live performances. In 2023,
Decentraland hosted its second Metaverse
Fashion Week, featuring over 60 fashion
brands, including heavyweights like Balen-
ciaga, Adidas, and Coach. While this event
showcased the allure and possibilities
within the metaverse, it also cast light on its
current developmental infancy, evidenced
by the participant feedback that spoke
EXPERIENCING IMMERSIVE WORLDS

TECH
need to change outfits. This effort is part of a
broader move to integrate AR technology into
the physical world, including at music festi-
vals and even in vending machines. Brands
like Ikea are also leveraging filters to enhance
the customer’s shopping journey at home.
Using smartphones’ lidar sensors, Ikea’s AR
tools allow customers to replace their exist-
ing furniture with detailed 3D models of new
items. Similarly, retail giants Walmart and
Amazon have successfully adopted AR to en-
able customers to visualize products in their
own space before making a purchase, wheth-
er it’s placing furniture or trying on glasses.
Holograms
Holography is a technique that records and
displays objects in three dimensions, mak-
ing them either stand still or move just like
real objects. It’s more than just a 3D picture;
it can show every detail of a person’s face
or body in motion. This capability is becom-
ing increasingly important for the future of
augmented reality and virtual reality, partic-
ularly as we look to populate the metaverse
with realistic avatars and environments. The
ARHT Media has also introduced Capsule,
a versatile holographic display designed
to shine in any lighting condition, perfect
for captivating audiences in cinema lob-
bies. This technology allows live hologram
presenters to interact with people across
distances, and it can showcase prerecorded
content that engages viewers through inter-
active touchscreens. The potential applica-
tions are vast, ranging from advertising and
immersive trailers to live talent meet-and-
greets and product demonstrations. In the
pursuit of more lifelike holographic projec-
tions, researchers have developed three-di-
mensional scattering-assisted dynamic
holography (3D-SDH), which greatly enhanc-
es the depth resolution of 3D images. This
cutting-edge method could revolutionize
how we interact with virtual environments by
offering a much richer, more detailed expe-
rience.
Real Estate in the Metaverse
Real estate in the metaverse is emerging as
a new digital frontier where people can buy,
develop, and experience virtual land and
fusion of holograms with deepfake technol-
ogy and synthetic media is paving the way
for their use in everyday settings. Already,
holography has enabled the creation of vir-
tual concerts featuring past celebrities, and
it holds the potential for production compa-
nies to bring popular synthetic characters
and celebrities into our physical world as
interactive entities. For instance, Ukrainian
President Volodymyr Zelenskyy harnessed
this technology for remote addresses across
European cities using Canadian firm ARHT
Media’s state-of-the-art holographic tech-
nology, speaking from within his country
amid conflict. The Hologram Zoo in Brisbane,
Australia, presents a new way for people to
experience wildlife. Here, holograms cre-
ate lifelike depictions of animals, such as
a herd of elephants charging toward and
then seemingly through the spectators. This
attraction makes holographic technology
more accessible by significantly reducing
costs, thus avoiding the ethical dilemmas
of captive wildlife exhibits and offering an
immersive educational experience.
Metaverse real estate offers long-term prospects for
virtual land as spaces for community, commerce,
and creativity, free from physical limitations.

TECH
properties. Though still nascent, metaverse
real estate is drawing interest from individual
investors, corporations, and even real world
real estate companies. In metaverses like
The Sandbox and Decentraland, virtual land
is sold as NFTs on blockchain. Parcels near
popular virtual spaces or owned by celebrities
can sell for hundreds of thousands of dollars
or more. The overall metaverse real estate
market is estimated to be worth over $1 bil-
lion. Corporations like Adidas and Atari, and
celebrities like Snoop Dogg have purchased
virtual land to host branded experiences, dig-
ital HQs, stores, and spaces. One buyer paid
$450,000 to become Snoop’s neighbor in the
virtual world.
Some companies are taking it further by
developing conceptual virtual homes and
structures with architectural firms. Every-
realm, a metaverse technology and infra-
structure company, has partnered with
artists like Misha Kahn and Daniel Arsham
to create an imaginative collection of digital
homes called The Row. These futuristic virtual
abodes feature melting, Salvador Dali-esque
gible prototype that exemplifies the vision
of a global community coming together in
a digital realm. The village is structured to
facilitate impactful interactions, with areas
such as a Virtual Congress Centre akin to a
digital town hall for hosting various future
meetings and sessions. It also compris-
es immersive collaborative centers that
serve as hubs for engaging storytelling and
spreading the Forum’s insights on critical
issues, fostering a collaborative spirit aimed
at real-world change. Partners of the Forum
have the opportunity to carve out their vir-
tual campuses, allowing them to rally their
stakeholders and drive forward initiatives to
solve worldwide problems. Simultaneously,
The Sandbox’s Wistaverse has emerged as a
pioneering virtual protest platform, offering
a secure space for worldwide activism and
education. Its goal is to empower users to
engage in nonviolent protest and discourse,
removing the risks associated with physical
gatherings.
Worlds for the Enterprise
Major tech companies are racing to make
virtual collaboration a reality for enterprises.
Microsoft is integrating Mesh directly into
Teams to allow coworkers to join 3D meetings
as avatars, sans VR headset. This aligns Mesh
with Microsoft’s everyday work tools after an
initial failed launch as a standalone develop-
er platform. Mesh now focuses on enabling
Teams’ 320 million users to have more engag-
ing meetings using spatial audio, simulated
environments like a virtual lake house, and
AI features like virtual whiteboards. Microsoft
is betting its massive user base will propel
adoption despite strong startup competitors
like Jugo and Frame also offering immersive
meeting solutions. Jugo similarly provides
3D collaboration spaces for remote teams to
brainstorm ideas on virtual whiteboards or
conduct meetings as customizable avatars.
Frame uses volumetric video to generate
photorealistic avatars, targeting enterprises
seeking hyperrealism. These companies aim
to make remote collaboration more natural by
simulating the nuances of in-person interac-
tions. Though still early, some companies are
architectural angles and dreamlike floating
spheres. Mirroring real world tactics, tradi-
tional home builders like KB Home have also
established presences to showcase model
homes. They aim to reach new demograph-
ics and drum up interest in real-life offer-
ings. While speculative now, advocates see
long-term potential for virtual real estate as
immersive spaces for community building,
commerce, and creativity. The metaverse
provides freedom from physical constraints
to create imaginative environments.
Worlds for Purpose
The metaverse is being harnessed for signifi-
cant causes, transcending beyond a mere
digital escape into a platform for meaningful
global action. The World Economic Forum
has taken a pioneering step by introduc-
ing the Global Collaboration Village in the
metaverse, a collaborative effort with Ac-
centure and Microsoft. This virtual space
is dedicated to convening organizations
to address, brainstorm, and act on global
crises. It was revealed during the Annual
Meeting 2023 in Davos, showcasing a tan-

TECH
piloting these technologies to enable hybrid
teams to work together more intuitively. Mic-
rosoft’s integration into Teams gives it a dis-
tribution advantage, but the space remains
competitive as startups offer differentiated
features around customization and realism.
Play-to-Earn and Virtual Marketplaces
Play-to-earn games fuse gaming and econom-
ics in metaverse environments by enabling
players to generate real-world value through
virtual gameplay. Leveraging blockchain
technology, these games establish in-game
assets like fantasy creatures or virtual land
as tradable NFT commodities. This diverges
from traditional in-game items, where items
lack external utility or ownership portability
after usage. Instead, metaverse games facili-
tate exchange of digital goods for cryptocur-
rency or fiat money. “Axie Infinity’’ is a pioneer
in this space, where players breed, raise, and
battle fantasy creatures called Axies, with in-
game transactions involving Ethereum-based
cryptocurrencies. “My Neighbor Alice’” also
offers a blockchain twist to casual farming
simulators, where players can buy virtual
land and earn rewards through daily activi-
ties and can use their NFTs to decorate their
virtual space. The well-known game “Mine-
craft” is also considered a virtual economy,
allowing players to trade items, and this
concept of in-game value is amplified in the
metaverse.
Marketplaces like OpenSea act as trad-
ing hubs for these virtual goods, allowing
the digital assets from various metaverse
games to be bought and sold, underpin-
ning the economic potential of play-to-earn
games. Exponential growth potential looms
as financial institutions like JPMorgan
strategically support metaverse transaction
platforms like Tilia. Seamless fiat curren-
cy interchangeability can further expand
adoption and capital inflow. Just as physical
economies rely on agreeing on currency, unit
of account, and other standards, maturing
these fundamentals will shape the trajectory
of blockchain-based virtual marketplaces.
The fusion of play and profitability through
ownership of digital goods represents merely
the first stage.
Players can earn real-world value in virtual environments through gameplay, with blockchain allowing for
the acquisition and trading of digital assets such as NFTs.

SCENARIOS
SCENARIO YEAR 2027
Cosplay Category Announcement
Welcome to the 2027 Super Cosplay Con—the world’s premier cosplay convention where physical reality converges
with the virtual multiverse. This year’s Super Cosplay Con is enabled by augmented reality glasses that project 3D
avatars and effects, blurring the lines between the real and virtual worlds. The costume contest categories reflect the
multifaceted experiences of Super Cosplay Con attendees.
The Digital Division honors stunning costumes crafted entirely from interoperable assets across different metaverses.
Contestants proudly display their creativity in mixing and matching skins, accessories, and effects from platforms
like Cryptoverse, Mara, and VRealms.
The Physical Division shines the spotlight on traditional handcrafted costumes made with real-world materials. Cos-
players spent months bringing elaborate costumes to life with fabric, foam, paint, and painstaking detail. Their devo-
tion to practical craftsmanship is on full display.
The Mixed Reality Division highlights the best blended costumes—real-world apparel taken to the next level through
digital avatars and AR enhancements viewable through attendees’ smart glasses. A knight’s armor appears charred
from virtual dragon fire, while a wizard’s cloak billows from fabricated wind.
Hope to see you at Super Cosplay Con!

METAVERSE
INFRASTRUCTURE

TECH
Interoperability
Interoperability is fundamental for the envi-
sioned metaverse: Akin to the universal stan-
dards of today’s internet, it would allow seam-
less data exchange and interaction across
various systems and technologies. This func-
tionality is essential to enable asset transfer
and identity management across diverse dig-
ital and physical spaces, and its importance
is highlighted in the World Economic Forum’s
2023 briefing paper, “Interoperability in the
Metaverse.” As companies advance in inte-
grating within their platform silos, the chal-
lenge shifts to connecting these silos to meet
the growing demands for cross-platform 3D
assets. Facilitating this effort, the Metaverse
Standards Forum, formally incorporated in
April 2023, brings together a consortium of
organizations to foster interoperability, not
by creating standards themselves but by
producing technical reports, best practices,
and guidelines. This collaboration is crucial
for developing the necessary standards to
build a cohesive and functional metaverse.
Businesses have made significant progress
in integrating various elements within their
and Information Technology (MIIT) is looking
to set standards for the metaverse industry,
which they see as crucial for promoting a
healthy and orderly market development.
China’s stance is to address the challenges
in the metaverse sector—such as the lack
of clear definitions that can lead to market
speculation—by providing standardization
and guidance. South Korea has also made
significant strides by investing 24 billion
Korean won ($18.1 million) in a fund dedicat-
ed to metaverse initiatives, recognizing the
difficulties local companies face in securing
private investments due to the risks as-
sociated with emerging technologies. The
government’s support extends to helping
with mergers and acquisitions and ensur-
ing domestic firms can compete on a global
scale. In a tangible demonstration of their
commitment, Seoul launched a digital twin
in the metaverse with a substantial govern-
ment investment, underscoring the impor-
tance of the metaverse in urban and social
development.
Developer Tools and Application Building
Blocks
The creation of the metaverse is driven by a
diverse array of developer tools, many of which
are open source to align with the metaverse’s
principles of inclusivity and community-driv-
en development. This democratic approach to
building virtual spaces is crucial, as it allows
for a metaverse that is crafted by its users,
rather than solely by large corporations. This
method of construction by a passionate user
base contributes to the metaverse’s unique
and innovative nature. A prime example is
the “Roblox” platform where a majority of
creators are under 18 years old. These young
developers use Roblox Studio to craft their
own games, employing Luau—a version of
the programming language Lua. The plat-
form facilitates creativity and entrepreneur-
ship, allowing these young creators to earn
through game passes and microtransactions,
a testament to the economic potential within
the metaverse. Tools like the Ethereal Engine
(XREngine) support this creative explosion,
offering open-source capabilities for crafting
immersive and interactive 3D experiences.
individual platform silos. However, the forth-
coming challenge lies in interlinking these
silos themselves. This necessity will grow
more pronounced as enterprise applications
converge and as the demand for cross-plat-
form 3D assets increases.
Government Investment
Government investment in the metaverse
is emerging as a key strategy for foster-
ing economic growth and innovation, with
several nations recognizing the potential of
this digital space. By supporting metaverse
infrastructure, governments aim to create
new jobs, attract businesses, and enhance
the lives of their citizens. Dubai’s Metaverse
Strategy is a prime example, where the
city plans to become a global hub for the
blockchain and metaverse sectors by 2030.
This strategy aims to support over 40,000
virtual jobs and increase the number of
blockchain companies fivefold, as part of
the UAE’s broader vision to create one of the
smartest cities worldwide. Similarly, China
is focusing on the metaverse as part of its
technology strategy. The Ministry of Industry
METAVERSE INFRASTRUCTURE

METAVERSE INFRASTRUCTURE
TECH
Platforms like Webaverse extend the creative
frontier further into decentralization, allowing
developers to build and interact with dApps
within the metaverse. What’s unique about
the metaverse developer tools is how they
empower builders, many of whom are young
and passionate users, to take ownership of
their creations. This participatory culture is
fueling the metaverse’s growth into a rich,
vibrant digital ecosystem where innovation is
constant and opportunities are vast.
Interdevice Synchronization
Interdevice synchronization is a vital com-
ponent in constructing the infrastructure of
the metaverse, as it guarantees a unified and
real-time interaction across various devices
in AR and VR settings. This has been a signifi-
cant hurdle to overcome. In environments like
online gaming and virtual workspaces, where
multiple users engage in a shared virtual
space, synchronization is key to ensuring
a smooth and coherent experience for all
participants. It is also crucial for accurately
replicating user movements within the virtual
realm, thereby enhancing the sense of immer-
mand high bandwidth and reliable networks
to process and transmit extensive visual
data for immersive experiences. The roll-
out of 5G by major US carriers like ATT,
Verizon, T-Mobile, and Dish Network is a
game-changer, offering up to 10 times the
bandwidth of 4G, with speeds up to 10 giga-
bits per second, lower latency, and greater
reliability. These features are key to avoiding
disruptions in the metaverse, ensuring a
seamless alternate reality experience. China,
as an early adopter and implementer of 5G,
exemplifies the impact of this technology on
the development of virtual and augmented
reality applications. With a large user base
in VR and AR, China is quickly becoming a
leading market in this field. The symbiotic
relationship between 5G and the metaverse
is evident here; the metaverse benefits from
5G’s widespread access, reliable connectiv-
ity, and the ability for XR devices to offload
processing to the edge of the network. This
mutual enhancement, evidenced in China’s
rapid growth in VR, is paving the way for
more comprehensive and globally accessible
metaverse experiences.
sion and averting problems such as motion
sickness. To address this challenge, re-
searchers from MIT and Microsoft developed
the Ekho system, which synchronizes audio
streams across different devices. The system
works by embedding subtle white noise
sequences into the audio stream sent from
the cloud server, which are then picked up
by the audio sensors in the player’s control-
ler or other devices. Ekho’s core mechanism
involves continuously monitoring the time
lag between the sent and received audio
streams through these white noise markers.
When a delay is detected, Ekho promptly
adjusts the timing, aligning the streams to
achieve near-perfect synchronization. This
advancement by MIT and Microsoft research-
ers marks a significant leap forward in
enhancing the overall quality and realism of
virtual experiences in the burgeoning field of
the metaverse.
5G for the Metaverse
The advancement of 5G technology is
required for the metaverse to reach its full
potential, as metaverse applications de-
For a seamless metaverse with low latency, it de-
pends on network connectivity, making 5G a crucial
enabling technology.

AUTHORS
CONTRIBUTORS
TECH

SAM JORDAN
Manager
Sam Jordan is a Manager at Future Today Institute. She leads our Advanced Comput-
ing practice area, which includes technology, artificial intelligence, virtual realities,
networking, telecommunications, and space. She is a distinguished practice area
lead, where she enables organizations to navigate through uncertainty with inno-
vative strategies. With a proven track record across various sectors, Sam’s visionary
leadership has driven growth and resilience for Future Today Institute’s global clients and partners.
Before joining FTI, Sam was the CEO and co-founder of TrovBase, a secure data discovery and analysis-sharing plat-
form. Sam grew the company from idea to launch and executed the company’s transition from scientific replication to
its current focus. In parallel, Sam engaged with the open science community, advocating for better data management
practices to address challenges in scientific replication. Previously, she worked for IBM, where she helped large enter-
prises in the retail and distribution sector modernize their IT stack. Her expertise centered around mainframes, assisting
with the integration of new software and modern methodologies to legacy systems.
Sam is a coach in the strategic foresight MBA course at the NYU Stern School of Business. She holds a BS in Economics
and Data Analysis from George Mason University and an MBA from New York University’s Stern School of Business.
AMY WEBB
Managing Director
MELANIE SUBIN
Creative Director
EMILY CAUFIELD
Editor
ERICA PETERSON
Copy Editor
SARAH JOHNSON
CHERYL COONEY
TECH

SELECTED
SOURCES

Adelson, Eric. “Virtual Reality Has
Expanded into Job Training.” The
Washington Post, August 10, 2023.
www.washingtonpost.com/tech-
nology/interactive/2023/virtu-
al-reality-job-training/.
Ahvenainen, Jouko. “Metaverses
Are Coming, but Who Owns Your
Avatar?” Prifina, February 8 2022.
medium.com/prifina/metaverses-
are-coming-but-who-owns-your-
avatar-61ae9750f9c2.
Ali M. Fazlollahi, Recai Yilmaz,
Alexander Winkler-Schwartz, et al.
“AI in Surgical Curriculum Design
and Unintended Outcomes for
Technical Competencies in Sim-
ulation Training.” JAMA Network
Open, 2023; 6 (9): e2334658.
https://jamanetwork.com/jour-
nals/jamanetworkopen/fullarti-
cle/2809667.
ARHT Media. “President Zelensky
to the Final Frontier: How ARHT Me-
dia Is Using Hologram Technology
to Transform the In-Theatre Experi-
ence.” Screen, June 21, 2022. www.
screendaily.com/president-zelen-
sky-to-the-final-frontier-how-ar-
ht-media-is-using-hologram-tech-
nology-to-transform-the-in-the-
atre-experience/5171865.article.
Bailey, Ashley. “Understanding the
Difference Between Deepfakes and
3D Human Avatars.” Veritone, July
21, 2022. www.veritone.com/blog/
understanding-the-difference-be-
tween-deepfakes-and-3d-hu-
man-avatars/.
Bain, Katie. “Coachella 2023
Announces ‘Fortnite’ Metaverse
Activation Featuring Music from
Porter Robinson.” Billboard, April
13, 2023. www.billboard.com/
music/music-news/coachel-
la-2023-fortnite-metaverse-coach-
ella-island-1235302313/.
Basu, Tanya. “Snap Is Launching
Augmented-Reality Mirrors in
Stores.” MIT Technology Review,
April 19, 2023. www.technologyre-
view.com/2023/04/19/1071849/
snap-ar-mirrors/.
Burgos, Matthew. “Meta’s New AI
Chatbots Let Users Text and Ask
Advice from Paris Hilton, Snoop
Dog and More.” Designboom,
October 2, 2023. www.designboom.
com/technology/meta-new-ai-
chatbots-paris-hilton-snoop-dog-
kendall-jenner-10-02-2023/.
‌
City University of Hong Kong.
“Wireless Olfactory Feedback Sys-
tem to Let Users Smell in the VR
World.” ScienceDaily, June 6, 2023.
www.sciencedaily.com/releas-
es/2023/06/230606111647.htm.
Clegg, Nick. “How the Metaverse
Can Transform Education.” Meta,
April 12,. 2023.
about.fb.com/news/2023/04/how-
the-metaverse-can-transform-ed-
ucation/.
Danry, Valdemar. “Towards a
Congruent Hybrid Reality With-
out Experiential Artifacts.” MIT
Media Lab, 2023. www.media.
mit.edu/projects/situated-vr-to-
wards-a-congruent-hybrid-reality/
overview/.
“Designed for Privacy, Controlled
by You.” Meta, 2023. about.meta.
com/ray-ban-stories-privacy/.
Diego Gómez-Zará, Peter Schiffer,
Dashun Wang. “The Promise
and Pitfalls of the Metaverse for
Science. Nature Human Behaviour,
2023. www.nature.com/articles/
s41562-023-01599-5.
DigiSomni. “Creating Hyper-Real-
istic Virtual Reality Avatars with
3D Modeling.” Medium, August 9,
2020 medium.com/@DigiSomni/
creating-hyper-realistic-virtu-
al-reality-avatars-with-3d-model-
ing-14317a1bddc.
Dohler, Mischa. “MWC 2023
Chronicles: Navigating the AR/
VR Evolution and the Metaverse’s
Downturn.” Ericcson Blog, April
20, 2023. www.ericsson.com/
en/blog/2023/4/mwc-rise-of-vr-
metaverse.
“Exploring the Impact of Virtual
and Augmented Reality in Courts.”
Unesco.org, 2023.
www.unesco.org/en/articles/
exploring-impact-virtual-and-aug-
mented-reality-courts.
“For Those Using Metaverse Tech-
nologies, the Impact Is Real.” Meta,
September 7, 2023. about.fb.com/
news/2023/09/impact-of-vr-and-
ar/.
Garuda Fujii. “Camouflaging the
Location of a Sound Source Via
Topology-optimized Source-shift-
er. Journal of Sound and Vibration,
2023; 559. www.sciencedirect.
com/science/article/abs/pii/
S0022460X23002171.
Golomingi, Raffael, et al. “Aug-
mented Reality in Forensics and
Forensic Medicine—Current Status
and Future Prospects.” Science
Justice, vol. 63, no. 4, April 2023.
https://doi.org/10.1016/j.sci-
jus.2023.04.009.
Gray, Georgi. “University of Texas
Medical Branch Augments Curric-
ulum with Virtual Reality Simula-
tions.” Virti Insights—Get the Most
out of Your People.” September 14,
2023. insights.virti.com/universi-
ty-of-texas-medical-branch-aug-
ments-curriculum-with-virtual-re-
ality-simulations/.
Guo, Eileen. “How Meta and AI
Companies Recruited Striking
Actors to Train AI.” MIT Tech-
nology Review, October 19,
com/2023/10/19/1081974/
meta-realeyes-artificial-intelli-
gence-hollywood-actors-strike/.
Hannah Raila, Tatevik Avanesyan,
Keara E. Valentine, et al. “Aug-
menting Group Hoarding Disorder
Treatment with Virtual Reality
Discarding: A Pilot Study in Older
Adults.” Journal of Psychiatric
Research, 2023; 166. www.science-
direct.com/science/article/pii/
S0022395623003965.
“How the Metaverse Could Shape
Science.” Kellogg Insight, June 21,
2023. insight.kellogg.northwestern.
edu/article/how-the-metaverse-
could-shape-science.
Huet, Natalie. “The Tech, Mind-
set and Headsets Making the
Metaverse ‘Real.’” Euronews, July
1, 2023. www.euronews.com/
next/2023/01/07/ces-2023-from-
a-scented-headset-to-neural-ear-
buds-this-tech-is-making-the-
metaverse-more-r.
Jiayi Xu, Shoichi Hasegawa,
Kiyoshi Kiyokawa, et al. “Integra-
tion of Independent Heat Transfer
Mechanisms for Non-Contact
Cold Sensation Presentation
With Low Residual Heat.” IEEE
Transactions on Haptics, 2023.
https://ieeexplore.ieee.org/docu-
ment/10285873.
Lay-Flurrie, Jenny. “Microsoft’s
Annual Ability Summit: Exploring
the Technology, People, Partner-
ships and Policies Driving a More
Accessible Future.” Microsoft
on the Issues, March 8, 2023.
blogs.microsoft.com/on-the-is-
sues/2023/03/08/accessibili-
ty-annual-ability-summit-peo-
ple-partnership-policy/.
SELECTED SOURCES

Martin, Scott. “BMW Group
Starts Global Rollout of Nvidia
Omniverse.” Nvidia Blog, March
21, 2023. blogs.nvidia.com/
blog/2023/03/21/bmw-group-
nvidia-omniverse/.
Massachusetts Institute of Tech-
nology. “Augmented Reality Head-
set Enables Users to See Hidden
Objects.” ScienceDaily, February
releases/2023/02/230227132721.
htm.
McGill University. “Assessing
Unintended Consequences in
AI-based Neurosurgical Training.”
ScienceDaily, September 19, 2023.
es/2023/09/230919154827.htm.
Meta. “Introducing the New Ray-
Ban Meta Smart Glasses.” Sep-
tember 27, 2023. about.fb.com/
news/2023/09/new-ray-ban-me-
ta-smart-glasses/.
Mulligan, Amanda. “Walmart Cata-
log Lenses.” Forbusiness.snapchat.
com, 2023. forbusiness.snapchat.
com/inspiration/walmart-edlp-
catalog-shopping-lens.
Na-young, Kim. “Naver to Cre-
ate Digital Twin Platform for
5 Saudi Arabian Cities.” Yon-
hap News Agency, October
24, 2023. en.yna.co.kr/view/
AEN20231024003200320.
Optica. “Technology Advance
Paves Way to More Realistic 3D
Holograms for Virtual Reality and
ore.” ScienceDaily, April 6, 2023.
es/2023/04/230406114000.htm.
Paleja, Ameya. “Microsoft Pat-
ents Swappable Batteries for AR
Glasses.” Interestingengineering.
com, October 10, 2023. interestin-
gengineering.com/innovation/mi-
crosoft-patent-ar-glasses-swap-
pable-battery.
Panagiotis Kourtesis, Rayaan Amir,
Josie Linnell, et al. “Cybersick-
ness, Cognition, Motor Skills:
The Effects of Music, Gender, and
Gaming Experience.”
IEEE Transactions on Visualization
and Computer Graphics, 2023; 29
(5): 2326. https://ieeexplore.ieee.
org/document/10049731.
Panpan Yu, Yifan Liu, Ziqiang
Wang, et al. “Ultrahigh-density 3D
Holographic Projection by Scatter-
ing-assisted Dynamic Holography.
Optica, 2023; 10 (4): 481.
https://opg.optica.org/optica/
fulltext.cfm?uri=optica-10-4-
481id=528843.
Pierce, David. “Genies Digital
Avatars: How Characters Work in
the Metaverse..” Protocol, January
26, 2022. www.protocol.com/ge-
nies-digital-avatars-metaverse.
Reardon, Gregory, et al. “Shear
Shock Waves Mediate Haptic
Holography via Focused Ultra-
sound.” ScienceAdvances, vol. 9,
no. 9, March 1, 2023. https://doi.
org/10.1126/sciadv.adf2037.
“Samsung Unfear.” DAD, 2023.
www.dandad.org/awards/pro-
fessional/2023/237490/sam-
sung-unfear/.
Sandra Lilja, Xinxiu Li, Martin
Smelik, et al. “Multi-organ Sin-
gle-cell Analysis Reveals an On/
off Switch System with Potential
for Personalized Treatment of
Immunological Diseases.” Cell
Reports Medicine, February 28,
2023. https://doi.org/10.1016/j.
xcrm.2023.100956.
Shinshu University. “Source-shift-
ing Metastructures Composed
of Only One Resin for Location
Camouflaging.” ScienceDaily, May
releases/2023/05/230530125407.
htm.
Tsamitros, N et al. “The Application
of Virtual Reality in the Treat-
ment of Mental Disorders]. Der
Nervenarzt vol. 94,1 (2023): 27-33.
https://pubmed.ncbi.nlm.nih.
gov/36053303/.
Tedeneke, Alem. “World Economic
Forum Establishes Purpose-Driven
Metaverse to Tackle World’s Most
Pressing Challenges.”
World Economic Forum, Janu-
ary 17, 2023. www.weforum.org/
press/2023/01/world-economic-
forum-establishes-purpose-driv-
en-metaverse-to-tackle-world-s-
most-pressing-challenges/.
“The Tech, Mindset and Headsets
Making the Metaverse ‘Real.’”
Euronews, January 7, 2023. www.
euronews.com/next/2023/01/07/
ces-2023-from-a-scented-head-
set-to-neural-earbuds-this-tech-
is-making-the-metaverse-more-r.
University of Tsukuba. “Simulating
Cold Sensation Without Actual
Cooling.” ScienceDaily, October
releases/2023/10/231023124409.
htm.
University of Waterloo. “Re-
searchers Explore Why Some
People Get Motion Sick Playing
VR Games While Others Don’t.”
ScienceDaily, May 1, 2023.
es/2023/05/230501085850.htm.
William Chung, Michael Bar-
nett-Cowan. “Sensory Reweight-
ing: A Common Mechanism for
Subjective Visual Vertical and Cy-
bersickness Susceptibility. Virtual
Reality, 2023. https://link.springer.
com/article/10.1007/s10055-023-
00786-z.
World Economic Forum. “Interop-
erability in the Metaverse.” Brief-
ing paper. January 2023. www3.
weforum.org/docs/WEF_Interop-
erability_in_the_Metaverse.pdf
Xintao Yan, Zhengxia Zou, Shuo
Feng, et al. “Learning Naturalistic
Driving Environment with Statis-
tical Realism.” Nature Communi-
cations 14, 2023. www.nature.com/
articles/s41467-023-37677-5.
Yiming Liu, Chun Ki Yiu, Zhao
Zhao, et al. “Soft, Miniaturized,
Wireless Olfactory Interface for
Virtual Reality.” Nature Communi-
cations 14, 2023/ www.nature.com/
articles/s41467-023-37678-4.
SELECTED SOURCES

BIOENGINEERING

TECH
TABLE OF CONTENTS
BIOENGINEERING
259 Top Headlines
260 State of Play
261 Key Events
Developments
263 Why Bioengineering Trends
Matter to Your Organization
264 When Will Bioengineering
Trends Impact Your
Organization?
To Consider
268 Central Themes
269 Ones To Watch
270 Important Terms
272 AI and Biology
273 Pharmaceutical Companies
Invest in AI
273 Generative AI Gains Wide
Acceptance
282 Using the Human Genome Map
282 Unlocking Bioinformatics Data
283 Sequencing Ancient Genomes
283 Programmable Gene Editing
Proteins
284 Bioprinting, Organoids, and
Novel Organisms
285 Bioprinting Electronics
285 Bioprinting and Tissue
Engineering
285 Fabricating Organoids
286 Growing Organoids to Study
Long COVID
286 Organ-on-a-Chip
287 3D Printed Drugs
287 Bacterial Nanosyringes
287 Using Viruses to Deliver Big
DNA Payloads
288 Minimum Viable Lifeforms
289 Biocomputing and
Cyberbiosecurity
290 Organoid Intelligence
290 Training Biocomputers to
Learn New Skills
290 Biological Circuits
290 Programmable DNA Machines
291 Using DNA to Store Data
292 Biological Robots
292 Living Sensors
293 Cyberbiosecurity
294 New Materials
295 AI-Created New Materials
295 Modifying Fungi for Building
and Packaging
295 Mycelium Leathers
296 Biomolecule-Based Packaging
296 Intelligent and Active
Packaging
296 Biodegradable and Edible
Packaging
296 Durable Biofilms
297 Cultured Food Drinks and
Next-Gen Agriculture
298 Lab-Grown Meat Is Going
Mainstream
298 Synthetic Milk and Cheese
298 Precision Fermentation
299 Brewing Great Nonalcoholic
Beers
299 Upgrading Photosynthesis
300 Faster Flowering
300 A Return to Natural Farming—
With Technology
300 Regenerative Agriculture
300 A New Wave of Genetically
Modified Foods
301 CRISPR Animals
301 Genetically Engineered Space
Farming
302 Aging, Health, and Beauty
303 Cultivated Collagen
273 Text-to-Synthetic Compound:
LLMs Automate Chemistry
273 AI-Generated Proteins
274 Generative Biology
274 Simulating Molecular
Interactions
274 Spatial Biology Improves
with AI
276 Gene Editing CRISPR
277 CRISPR Treatment for Diseases
277 Next-Generation Gene Editors
278 New Editing Tools
279 In Vivo Gene Editing
279 Cell Therapy 2.0
280 Reading and Sequencing
Genomes
281 Next-Generation Genome
Sequencing
281 Metagenomics
281 Faster Gene Synthesis
282 Quantum Biology

TECH
TABLE OF CONTENTS
BIOENGINEERING
303 Growing Blood
303 Growing Sex Cells
303 Human Trials of Synthetic
Wombs
304 Universal Cancer Vaccines
304 Upgrading Embryos
Before Birth
305 Genetic Screening
for Pregnancy
305 Biobank Releases
306 Microbiome Metrics at Home
306 Age Reversal
306 Removing Zombie Cells
306 Skin Care and Beauty
307 Improving Gut Biomes
307 Optimizing Recreational Drugs
308 Climate and Sustainability
309 eDNA Detection
309 Using CRISPR for Sustainable
Wood Production
309 Better Plastics Recycling
317 Posthumous Sperm Retrieval
317 Gene Editing Ethics
318 Engineering Super Soldiers
319 Authors
310 Engineering Plants for Carbon
Capture
310 Greening Fashion
311 De-Extincting Lost Species
311 Rewilding Barren Terrains
311 CRISPR Mosquitoes
312 Reviving Ancient Viruses
313 Regulation and Policy
314 Regulatory Changes Toward
Genome-Edited Crops
314 Regulating DNA Recognition
Systems
314 Safeguarding Genetic Privacy
315 Defining Parenthood
315 National DNA Drives
315 International Collaborations to
Advance Bioengineering
316 Ethics, Trust and Acceptance
317 Resolving Bias in Genome
Research
317 Ethics in Indigenous Genomics

AI breakthroughs are
set to make this year a
turning point for biotech,
CRISPR, and synthetic
biology.
TOP HEADLINES
BIOENGINEERING
01
02
03
04
05
Lab-Grown Meat Is Going Mainstream
Cultivated meat and synthetic dairy are emerging as sustainable and ethical alternatives
to traditional animal products, and in 2024 production will start to scale.
CRISPR Will Treat Disease
US and UK regulators have approved the first CRISPR therapy, marking a significant
step in gene editing’s potential to cure genetic diseases.
Millions of New Materials Discovered
A new AI system from Google DeepMind predicted 2.2 million novel materials, signaling
big changes ahead for batteries, computer chips, and more.
AI Transforms Biology Today; Biology Powers Computers Tomorrow
Scientists created organoid intelligence, a new biocomputing field using brain cells for
AI, promising major efficiency gains over traditional computer systems.
We Aren’t Prepared for a Cyber-Bio Attack
Cyberbiosecurity has emerged as a crucial unmet need at the intersection of AI and
life sciences, with no formal oversight yet established.
TECH

STATE
OF PLAY
In the past year, AI breakthroughs have accelerated the convergence of
biology, information systems, and advanced platforms, with the potential
to fundamentally transform businesses and societies. While today, our
focus is on generative AI, very soon that focus will shift to generative bi-
ology: Here, AI models will decipher the complexities of biology, leading to
the creation of novel molecules, drugs, materials, consumables, and living
organisms.
Our orientation to biology and the living world will change in the near
future as CRISPR products, in the form of novel drug therapies and new
foods, enter the market. Soon, we will no longer be constrained by existing
materials to manufacture batteries, clothing, or buildings.
Lab-grown beef, chicken, and fish will scale, making it possible to con-
sume meat without slaughter. Through cellular reprogramming, we could
start to reverse the aging process. And we’re on the brink of new fertility
treatments that will challenge our ideas about parenthood.
These shifts will fundamentally alter our relationship with biology and the
natural world, but we lack comprehensive policies to navigate them. While
biotech promises to reshape our world, it also presents unprecedented
risks, underscoring the need for preparedness and governance.
Brace yourself for uncharted
impacts. This could be
a breakthrough year for
biotechnology.
TECH BIOENGINEERING
260

JUNE 13, 2023
Pushing Limits of Embryo Models
Israeli, UK, and Chinese research teams
advance embryo models beyond legal
limits, prompting ethical debates.
NOVEMBER 29, 2023
AI Unlocks New Material Secrets
Google DeepMind’s AI predicted structures of
over 2 million potential new materials, potentially
enhancing technologies like batteries, solar
panels, and chips. Scientists could create around
400,000 of these materials in labs soon.
JANUARY 18, 2024
Universal Cancer Vaccine Trials
Cancer patients in the UK receive
the first dose of a new mRNA
therapy designed to help the body
recognize and fight cancer cells.
DECEMBER 8, 2023
FDA OKs Gene Therapy for Sickle Cell
Casgevy becomes the first approved
commercial gene-editing treatment
using CRISPR technology.
FEBRUARY 6, 2024
GMO Seeds Sold Direct to Consumers
The FDA gives purple tomatoes, genetically
enhanced to be more nutrient-dense,
approval for home gardening.
KEY EVENTS
TECH BIOENGINEERING

RAPID ACCELERATION
The rapid integration of artificial intelli-
gence into bioengineering is catalyzing
unprecedented innovation. The near
future will see significant changes to
traditional industries such as meat, dairy,
textiles, and pharmaceuticals, while
advancements should pave the way for
sustainable solutions in carbon capture,
plastics recycling, and biodiversity en-
hancement. Near-term breakthroughs in
healthcare will lead to cataclysmic, long-
term disruption. Leaders should deepen
their understanding of bioengineering’s
vast potential and associated risks, en-
abling them to spearhead innovation in
products and services, streamline pro-
cesses, improve materials, and achieve
cost efficiencies. This convergence also
opens up new avenues for collaboration,
allowing companies to explore untapped
markets and forge strategic partnerships
that can drive forward their competitive
edge.
CRISPR’s Emerging Economic Impact
The size of the global market for CRISPR
technologies and their associated products is
projected to exceed $4 billion by mid-decade.
In the US alone, CRISPR could contribute $19
billion to GDP by 2032.
GMO Misinformation
A new GMO backlash looms as public mis-
information persists, with many unaware
that the latest GMOs aim to boost produce
and grains with enhanced nutrients, not just
modify them for convenience or yield.
Life Extension Backlash
Life extension breakthroughs promise longer
lives for some, yet they will strain social ser-
vices, pensions, and other support systems
for the elderly, challenging our preparedness
for an aging population.
Supply Chain Disruption
In the next year, new materials poised to im-
prove shipping’s environmental footprint may
disrupt traditional supply and cold chain op-
erations dependent on outdated tech, posing
significant threats to established companies.
Regulatory Pressures
Federal Trade Commission actions spark
fears of stifled innovation due to limits on
scaling therapies via acquisitions, while the
US Inflation Reduction Act ushers in a harsh-
er pricing climate, impacting the biopharma
industry’s future reimbursement strategies
for innovations.
Stark Health Divide
Biotech will revolutionize health care with
novel drug therapies, yet their inaccessibility
to developing economies risks creating a
stark health divide, exacerbating global
health inequalities.
Media
Telecom
Distribution
TECH BIOENGINEERING

As a modern “general
purpose technology,”
bioengineering has the
potential to influence
an entire economy and
alter society through
political, economic, and
social structures. By
2030, most people in
developed economies will
have used or consumed
a product created
with a bioengineering
technology.
Biotech advancements
will transform the value
network for businesses.
Without active
monitoring, disruptors
may emerge, threatening
established companies’
capacity to create value
and maintain their
competitive edge for
market share, investment,
and talent.
Advancements in
synthetic biology,
CRISPR, AI, and
engineering are opening
up new opportunities
across health care,
pharmaceuticals,
agriculture, food and
beverage, beauty,
chemicals, sustainability,
energy, and materials
production.
The convergence of
AI and biology will
impact every business.
Leaders will need a
better understanding of
bioengineering’s potential
and risks, so they can
innovate products
and services, develop
processes, enhance
materials, reduce costs,
and seek out new partners
and customers.
Near-term disruptions
in the traditional
meat, dairy, textile,
and pharmaceutical
industries are imminent,
with advancements
offering new options
for carbon capture,
plastics recycling, and
enhancing biodiversity,
signaling a pivotal
shift in environmental
management.
In response to
bioengineering advances,
businesses must develop
policies that embody
their values, ethics, and
culture, such as the use
of genetically enhanced
ingredients, insurance
coverage for novel genetic
therapies, whether and
how to promote genetic
privacy, and more.
Biotech Is a General
Purpose Technology
Biotech Will Change
Your Value Network
Advancements Should
Result in Optionality
AI-Bio Convergence
Requires New Strategy
Disruption Is on Your
Near-Term Horizon
Businesses Need
New Policies
WHY BIOTECHNOLOGY TRENDS MATTER TO YOUR ORGANIZATION
TECH BIOENGINEERING

TECH
264
WHEN WILL BIOTECHNOLOGY DISRUPT YOUR ORGANIZATION?
BIOENGINEERING
Medical Devices
Vaccines
OTC Medications
Poultry Farming
Defense
Beauty
Drug Development
Textiles
Luxury Fashion
CO2 Capture Farming
Ranching
Health Diagnostics
Recycling
Chemicals
Fast Fashion
Consumer Packaged Goods
Construction Building Materials
Paints Coatings
Data Storage
Grain
Automotive
Oil/Petroleum Sector
Electricity
Aerospace
Cold Chains
Living Materials
Computing
Life Extension
Off-Planet Exploration

TECH
265
WHEN WILL BIOTECHNOLOGY DISRUPT YOUR ORGANIZATION?
BIOENGINEERING
Advancements in bioengineering
will disrupt every industry, to some
degree, within the next two decades.
The Future Today Institute categorizes
bioengineering as a “general purpose
technology.” Like the steam engine
and internet before it, bioengineering
has the potential to influence an
entire economy and to alter society
through political, economic, and so-
cial structures. For most industries,
it’s not a matter of if these trends will
disrupt their futures but when.
Several factors are driving the mo-
mentum of bioengineering trends and
the probable timing an industry will
be disrupted:
Regulations
The pace of technology advancement typi-
cally far exceeds any changes to regulation.
Bioengineering is unique in that regulation
exists, but products and processes are treat-
ed differently in every country. Regulatory
and policy uncertainty could accelerate or
stifle growth.
Media mentions
Increased awareness and enthusiasm can
influence the momentum of a technology,
even when there’s been no real breakthrough.
Media bursts related to bioengineering will
drive momentum, especially if those stories
are favorable and—importantly—easily under-
stood by the general public.
Public perception
How the public understands, and responds to,
bioengineering advancements will create or
quell demand. This is especially true for food
and beverage, consumer packaged goods,
beauty and fashion, over-the-counter medi-
cines and vaccines, and new therapeutics.
RD developments
The pace of new research breakthroughs can’t
be scheduled to coincide with a board meet-
ing or earnings report. There are factors that
can improve the likelihood and speed of new
discoveries (funding, quality and size of staff,
access to resources). We closely monitor RD
developments but treat them as wild cards.
Scaling
While the pace of innovation is fast across
the spectrum of technologies, it takes time
for a promising new biotech development to
scale beyond the lab. Scaling requires disci-
pline, patience, effort, and time.
Costs
Bioengineering research is still costly,
though the price of components, equipment,
and materials drops every year. Once a dis-
ruptor can make a product cheaper with bio-
engineering rather than traditional produc-
tion, it will push faster into the mainstream.
Advancements in technology will eventually
bring down costs of production as we’ve
seen in other fields, such as computing.
Constraints on adoption
Even if a technology is maturing, constraints
on its adoption can hinder its influence in
an industry. For example, a business may re-
fuse to adopt an alternative bioengineering
technology because it challenges a proven,
successful strategy.

Threats
There is no alignment on a global framework governing bioengineering. As a
result, you can anticipate geopolitical conflict stemming from the development
and use of emerging bio-based technologies in the years to come.
Unresolved IP and patent issues in biotechnology raise serious concerns about
how patenting practices could hinder biotech development across various
countries.
While new agricultural methods may benefit the environment, they pose threats
to local communities in economies still reliant on traditional farming practices.
CRISPR therapies may not be universally accessible, potentially deepening
global health divides and exacerbating current disparities, leading to worsened
conditions and future conflicts.
Engineering novel organisms and using techniques like germline editing risks
uncontrollable cascading effects in nature, potentially creating invasive species
or pathogens, turning a lab accident into an ecological disaster.
Opportunities
In the coming decade, biotech advancements will cause leaders to confront their core
beliefs about their business models, products, and services. In the meantime, it’s
imperative that businesses seek out new partnerships, develop new pipelines for talent,
and align stakeholders on the moral and ethical uses of engineered biology.
Because bioengineering has been evolving over decades, it may seem premature for
action, but its advancements will compound. Like AI, which grew quietly before becoming
essential, bioengineering will soon be integral to operations. Leaders paying attention to
its progress and harnessing strategic foresight can gain first-mover advantages.
Biotech companies always face capital needs, particularly amid sector-wide valuation
challenges. Streamlining RD, general, and administrative expenses—as well as
exploring new financing options, and considering mergers with other biotechs—can help
them navigate the current market landscape.
Businesses should start exploring white spaces now. Where are the opportunities for
innovation and growth? What might threaten the organization’s ability to thrive? Where
are there downstream risks to partners or customers? Businesses that seek out and
anticipate near-term developments will gain a competitive advantage.
Generative biology (genBio) will unlock new ways to develop medicines, food, agricultural
systems, beauty and skincare products, textiles, packaging and building materials, and
more. Leaders should develop scenarios for using and scaling these genBio systems,
processes, and products.
TECH BIOENGINEERING

As companies consider AI’s
impact on their workforce,
they are neglecting to fo-
cus on future demands for
individuals skilled in both
AI and biology. Compa-
nies across various sectors
should proactively develop
insights into their future
workforce requirements
and start establishing the
necessary talent pipeline
now.
As VC investment floods
into groundbreaking bio-
tech platforms this year,
companies must distinguish
themselves by clearly de-
fining their uniqueness. Also
crucial is broadening their
focus from rare to major un-
met medical needs, ensur-
ing a solid value proposition
to secure crucial VC funding
for drug approvals.
As the biotech ecosystem
evolves, life sciences and
other companies will need
to undertake a new dig-
ital transformation that
includes AI. This should
include the creation of
a long-term strategy, an
expanded value network
map, and a comprehensive
execution plan to stay com-
petitive and innovative.
The AI-bio convergence will
spark myriad innovations
and demand unprecedent-
ed agility from companies.
Leaders must empower
their organizations to ex-
periment with new prod-
ucts and processes, and
ensure that they shape their
own futures rather than
being compelled to adapt
to external innovations or
react to regulatory shifts.
The uncertain regulatory
landscape offers a unique
opportunity for business
and government to collabo-
ratively envision the future.
Regulation is reactive.
Stakeholders can proac-
tively evolve frameworks to
address safety, update IP
and copyright processes,
align on commercialization
strategies, and thwart mis-
information.
To stay competitive, com-
panies must use strategic
foresight to understand
how the evolving biotech
ecosystem could impact
their existing products
and processes. Leaders
should prioritize moni-
toring, acting, and agile
decision-making in order
to adapt to the AI-bio
convergence.
1 4
2 5
3 6
TECH BIOENGINEERING

CENTRAL THEMES
268
TECH BIOENGINEERING
Gene Editing Realizes Its Promise
CRISPR’s journey to commercial success has been a
marathon, not a sprint. For more than a decade, the
gene-editing technology faced many hurdles, from
technical challenges to patent disputes to regulatory
approval, slowing its path to practical applications.
Finally, in 2023, CRISPR’s promise crystallized with the
approval of Casgevy, a first-of-its-kind therapy for sick-
le cell disease in the UK and US. This landmark therapy,
which deactivates a specific gene, highlights CRISPR’s
ability to disable genes with precision. More impor-
tantly, with government clearance and a real-world
use case, Casgevy opened the door to a vast array of
possibilities for gene editing and other biotechnology
applications. However, as CRISPR therapies move to-
ward broader application, challenges remain: namely,
the complexity and cost of treatments, accessibility
issues, and public trust. Despite these obstacles, CRIS-
PR’s trajectory toward commercialization marks a sig-
nificant milestone—though the marathon continues.
AI and Biology Are Converging
Increasingly, AI is being used in biological systems.
Scientists are no longer limited by a traditional human
team’s speed: New AI models now accurately predict
biological structures, a capability that will accelerate
scientific research that used to take decades. The AI-
bio convergence extends into computing itself, and
researchers are exploring the creation of biology-pow-
ered machines. These innovative systems promise to
be faster, more efficient, and consume significantly
less energy than traditional computers. Organoid in-
telligence aims to use human brain cells in a new type
of computer. Programmable DNA computers execute
complex operations through molecular manipulation.
Some researchers believe this fusion of computers and
biological processes is the real future of artificial in-
telligence; both are important because they offer novel
approaches to problem-solving and unlock new forms
of creativity. Biological computers potentially open
up unprecedented opportunities to improve compute
power, data storage, and sustainability.
Businesses and Governments Aren’t Prepared
Businesses and governments need to catch up as AI
and biology converge. Biotech fields are making dis-
coveries that not only deepen our insight and create
new options—they also introduce novel methods to al-
ter biological systems, with outcomes that remain un-
predictable. Such progress presents vast opportunities
for investment and for businesses to meet their ESG
goals. However, it poses challenges for incumbents
in supply chain management, agriculture, consumer
packaged goods, health, and biosciences that may
not see their value networks changing early enough to
take action. Cyberbiosecurity is a growing and unmet
need, as increasingly companies will need to protect
the biotech ecosystem from unauthorized access,
damage, attack, and other threats. As new biotech-
nologies emerge, a lack of alignment on purpose and
policy could result in the situation we see today with
AI. Without strategic foresight to prepare for the future,
the potential risks associated with biotechnological
advancements could surpass those of AI, underscoring
the urgent need for readiness.

Dr. Amy Trejo, director of RD and responsible
materials innovation at Procter Gamble, for
leveraging bioengineering for sustainability in
the consumer packaged goods space.
Dr. Arthur Levin, distinguished scientist at
Avidity Biosciences, for engineering a new RNA
platform that delivers therapy to previously
inaccessible tissue and cell types.
Dr. Cheryl Cui, CEO of Bota Biosciences, for
launching a new type of engine for biological
programming and discovery.
Chris Abbott, CEO of Pivot Bio, for scaling
sustainable biotech solutions to meet global
agricultural challenges.
Dr. Demis Hassabis and Dr. Shane Legg,
co-founders of DeepMind, for their ground-
breaking AI inventions that predict biological
structures and combinations.
Didier Toubia, CEO and co-founder of Aleph
Farms, for achieving the world’s first govern-
ment clearance to produce and sell cultured
beef.
Dror Bin, CEO for the Israel Innovation Author-
ity, for scaling innovation resources in biotech
and other critical technologies.
Eben Bayer, co-founder and CEO of Ecovative,
for innovating business models to integrate
mycelium technology into the production of
food and materials.
Dr. Emily Leproust, CEO of Twist Bioscience,
for breaking new ground in high-throughput
synthesis and sequencing of DNA.
Dr. Gaurab Chakrabarti, CEO of Solugen, for
decarbonizing the chemicals industry.
Dr. Hal Barron, Dr. Rick Klausner, and Hans
Bishop, founders of Altos Labs, for leading a
new effort on cellular rejuvenation program-
ming to reverse the human aging process.
Dr. J. Craig Venter, CEO of JCVI and serial en-
trepreneur, for advancing the fields of synthetic
biology and genomic research.
Dr. Jason Kelly, co-founder and CEO of Ginkgo
Bioworks, for scaling genetic engineering to
produce bacteria with a wide variety of applica-
tions.
Dr. Jianmin Fang, co-founder and executive
director of RemeGen, for overseeing one of the
largest partnership deals between a Chinese
biotech company and a Western company (Se-
agen) in history.
Josh Tetrick, co-founder and CEO of Eat Just,
for commercializing cultured meat in the US
and Singapore.
Dr. Kimberly Smith, RD chief at ViiV Health-
care, for her work to end the HIV epidemic
through her visionary practices and innovative
approach to clinical development.
Dr. Lisa Dyson, founder and CEO of Air Pro-
tein, for her work developing food from carbon
dioxide.
Dr. Mary Maxon, executive director of BioFu-
tures at Schmidt Futures, for developing and
leading a new program to maximize the poten-
tial of biotech for a circular bioeconomy.
Matthew McKnight, general manager of
biosecurity at Ginkgo Bioworks, for his role in
advancing biotechnology for national security,
public health, and pandemic preparedness.
Niyati Gupta, CEO of Fork Good, for building
and scaling new business models in food and
agribusiness.
Dr. Noubar Afeyan, CEO of Flagship Pioneering,
for inventing and building platform companies,
each with the potential to transform human
health and the planet.
Dr. Raymond Deshaies, senior vice president
for global research at Amgen, for his work on a
new frontier of small molecule design via RNA
degradation.
Dr. Sarah Reisinger, chief science and research
officer for DSM-Firmenich, for her continued
work bridging the gap between RD, technical
requirements, and commercialization.
Dr. Yin Ye, CEO and executive director of BGI
Group, for scaling the industrial application
of cutting-edge biotechnology and genomics
research.
ONES TO WATCH
TECH BIOENGINEERING

IMPORTANT TERMS
270
BIOENGINEERING DOMAINS
Innovations in biotechnology are currently defined
by five key areas: biomolecules, biosystems, bioma-
terials, biocomputing, and biomachine interfaces.
Major breakthroughs in one field either reinforces
or accelerates breakthroughs in the others.
Biocomputing
Biology is made up of code—and the goal is to
harness that code for computing. DNA and RNA can
be used as mediums for storing information and
data processing. While traditional supercomputers
use a lot of energy, heat up quickly, and require
costly cooling centers to function properly, biolog-
ical computing systems can perform computa-
tions without burning excess energy—and they are
infinitely scalable.
Biomachine iInterfaces
Innovative new interfaces are connecting living or-
ganisms to computers for many different purposes,
from restoring a stroke victim’s ability to walk to
someday controlling external computers simply
using thought.
Biomaterials
It is now possible to replicate or improve on raw
materials using bioengineering technology. One ex-
ADDITIONAL TERMS
Cas9 (CRISPR associated protein 9)
A special enzyme that can cut DNA sequences.
Cas9 is part of the “molecular scissors” method
of genome editing made possible by CRISPR.
Chimera
A living organism created by combining cells from
at least two genetically different organisms.
Chromosome
A thread-like structure made up of a single length
of DNA and found in the nucleus of each cell.
CRISPR (clustered regularly interspaced short
palindromic repeats)
A naturally occurring genetic engineering tool
found in bacteria that can be programmed to target
specific areas of genetic code and to edit DNA at
precise locations.
DNA (deoxyribonucleic acid)
A self-replicating two-stranded molecule, arranged
as a double helix, that contains the genetic instruc-
tions used in the development, functioning, and
reproduction of an organism.
Enzyme
A biological catalyst, usually a protein. Enzymes
speed up the rate of specific chemical reactions
in cells.
Ex vivo
Outside of cells or an organism.
Gain of function (GoF) research
Research intended to modify a biological pathway
in a cell line or organism to enhance or increase
certain biological functions.
Gene
The basic unit of heredity.
Genome
The complete set of DNA that makes up
an organism.
Genome editing
Intentionally altering cells or organisms by
inserting, deleting, editing, or otherwise modifying
a gene or gene sequence.
Heritable genetic change
Altering genes in a way that results in changes
that pass down through generations.
ample: a bioreplacement material that is produced
sustainably, at a lower cost than traditional raw
materials, and poses no harm to the environment.
Biomolecules (also known as omics)
A group of biological sciences collectively known as
“–omics,” including fluxomics (metabolic reactions
in cells), metabolomics (chemical species involved
in the reactions in cells), proteomics (the decoded
product, or proteins), transcriptomics (the RNA cre-
ated from each piece of genetic code), and genom-
ics (the DNA code that drives cellular processes) is
working to analyze the structure and functions of
biological molecules that translate into the func-
tion and dynamics of an organism. Learning about
and tinkering with the engineering of molecules
(think: DNA, RNA) will lead to new therapeutics and
innovative defenses against novel viruses, as well
as alternatives to the ways we currently grow food.
Biosystems
Biology is complex. Scientists are applying engi-
neering principles to understand and influence the
pathways, connections, and interactions within bi-
ological systems. Developing new processes could
lead to new opportunities to modify or even create
cells, tissues, organs, and potentially complex net-
works like respiratory systems.
TECH BIOENGINEERING

IMPORTANT TERMS
271
In vivo
Inside of cells or a living organism.
Induced pluripotent stem cells (iPSC)
Cells that have been reprogrammed back into an
embryonic-like state with the potential to develop
into other types of cells that can be used for thera-
peutic or reproductive purposes.
Mutation
A change in a DNA sequence.
Off-target effect
Typically an unintended direct or indirect conse-
quence of altering an organism.
Regenerative medicine
An emerging field seeking to repair or replace torn,
defective, or missing tissue using stem cells, engi-
neered cells, or biological processes.
RNA (ribonucleic acid)
A messenger chemical that carries instructions or
translates the genetic code of DNA into structural
proteins.
Stem cell
Nonspecialized cells that have the ability to
develop into other types of cells with specialized
functions.
Synthetic biology
A field of science rooted in both biology and
engineering that seeks to redesign organisms, or
design new organisms, to have new abilities.
TECH BIOENGINEERING

AI AND
BIOLOGY
TECH BIOENGINEERING

TECH
Pharmaceutical Companies Invest in AI
Historically, scientists struggled to mine big
biological data sets for insights using con-
ventional statistical tools. With the spotlight
now on AI, drug developers now see they’re
failing to capture the value of their important
asset: their data. AI’s potential to refine the
often unpredictable process of drug discovery
is significant, since even marginal enhance-
ments can substantially improve the speed
and efficiency of developing new drugs. Two
years ago, DeepMind’s AlphaFold made a
notable breakthrough in predicting protein
structures, an advancement that will soon
revolutionize the process of identifying mole-
cules with therapeutic potential. But the prac-
tical application of AI in drug development
is already underway. AstraZeneca now uses
reinforcement learning in 70% of its small
molecules in development. London-based
biotech startup E-therapeutics uses AI to de-
sign RNA molecules and algorithms to predict
their likely activity, in an effort to thwart dis-
ease-causing genes. Investment bank Morgan
Stanley projects that within the upcoming
decade, the pharmaceutical industry could
AI symptom checkers. As genAI technologies
continue to evolve and integrate within the
pharmaceutical and life sciences sectors,
strategic shifts in how health care and med-
ical research are conducted and managed
will occur.
Text-to-Synthetic Compound: LLMs
Automate Chemistry
While you were asking GPT-4 to write the lyr-
ics for a rap song about avocados, research-
ers have been enhancing the capabilities of
large language models to automate compli-
cated tasks in chemistry. One such research-
er is Philippe Schwaller, from the Swiss
Federal Institute of Technology in Lausanne,
whose team gave GPT-4 access to extensive
databases of molecules, chemical reactions,
and scientific research. They call the new
system ChemCrow, and they’re using it on a
wide range of chemistry challenges, includ-
ing drug synthesis and cost calculation.
ChemCrow successfully devised a practical
plan for synthesizing atorvastatin, a drug for
high blood cholesterol. On average, Chem-
Crow achieved over 9 out of 10 in human
evaluations for 12 chemistry tasks. Separately,
Gabriel Gomes at Carnegie Mellon Universi-
ty and his colleagues also upgraded GPT-4
with similar chemistry tools—with a twist.
This model is integrated in a remotely con-
trolled chemistry lab with liquid compounds
that could be mixed using robotic arms. They
asked the system to perform certain reactions
by writing in a prompt, which was then exe-
cuted by the robotic arms in the lab. But when
the team asked the system to whip up sarin
gas, the model—mercifully—refused. While
AI promises a new pathway to automate the
process of synthesizing compounds, it’s not
without potential danger. Public domain tools
could be used to create something potentially
dangerous.
AI-Generated Proteins
An AI system can now create new types of
proteins that don’t exist in nature. By focusing
on the protein’s amino acid building blocks,
researchers can design a protein with special
properties—like being really tough or flexible—
to make new materials that are like plastics
but better for the environment. Recently, ex-
spend roughly $50 billion a year on AI to
propel every stage of new drug development,
from ideation to compound identification to
clinical trial design and marketing.
Generative AI Gains Wide Acceptance
Just a few years ago, the pharmaceutical
and life sciences industries weren’t all that
interested in investing in or using AI to
automate and improve various processes.
Last year, when genAI took center stage, that
resistance started to fade, especially in the
Asia-Pacific region where genAI startups are
particularly active. Singapore’s Integrated
Health Information Systems entered a piv-
otal partnership with Microsoft and OpenAI,
which will create a health care–specific GPT
hosted on Azure. Once it’s finished, this
initiative aims to enhance the efficiency of
health care workers by providing valuable
insights and automating tasks. In Japan,
the introduction of an advanced genAI tool
is transforming the way doctors process
extensive patient interviews. Tokyo-based
Ubie is a startup that makes AI-based health
care products, hospital SaaS products, and
AI AND BIOLOGY
BIOENGINEERING

TECH
perts have made AI programs that can guess
the 3D shape of a protein from its amino acid
sequence. But figuring out the best amino
acid sequence to engineer a certain protein
shape is much harder. Researchers from MIT,
the MIT-IBM Watson AI Lab, and Tufts Univer-
sity used a generative AI model not unlike
DALL-E 2, which generates realistic images
from natural language prompts. Then, they
adapted the model so it generates amino acid
sequences of proteins that achieve specific
structural objectives. The researchers used
this architecture to build two machine learn-
ing models that can predict a variety of new
amino acid sequences that form proteins
meeting structural design targets. Meanwhile,
researchers at the University of Washington’s
Institute for Protein Design Baker Lab creat-
ed a new program, called RFdiffusion, which
allows scientists to upload a 3D model of a
cell and use software to identify the most
likely candidates for drug testing. RFdiffusion
increases the accuracy as much as 100-fold,
compared to previous protein design pro-
grams that didn’t use AI.
testing of drug compound candidates. This
new approach circumvents the traditional,
often prohibitively expensive, methods of
chemistry-based experimentation. But the
use of AI in this context is not just a matter
of cost-efficiency; it marks a transformative
step in bioengineering. By simulating molec-
ular interactions on computers, researchers
can rapidly iterate and refine drug com-
pounds, significantly accelerating the devel-
opment process. This method offers a more
sustainable, scalable, and potentially more
accurate alternative to physical testing.
What’s on the horizon: much faster, cheaper
RD. Our analysis suggests that organiza-
tions that adapt in silico molecular simula-
tions powered by AI will gain a competitive
edge—and be prepared for knock-on effects
that make up the broader movement toward
a more innovative, data-driven approach in
bioengineering and health care.
Spatial Biology Improves with AI
Spatial biology is a burgeoning field predi-
cated on gaining a deeper understanding of
Generative Biology
What if it was possible to generate novel
protein therapeutics using new computa-
tional tools, without having to discover them
through trial and error? That’s the promise
of Boston-based startup Generate Biomedi-
cines, which trained an AI to invent proteins
with structures that, as far as we know, don’t
exist anywhere in nature. Inspired by DALL-E
2, the powerful text-to-image AI system from
OpenAI, Generate’s platform asks the user
to describe the shape, size, and function
of a protein they’d like to see. It then uses
diffusion modeling to generate a structure
with the right amino acids folded correctly to
meet the description. Our understanding of
the genome, along with fundamental molec-
ular and network mechanisms, is now being
enhanced by innovative tools that allow us
to interact with, examine, and manipulate
biological systems in new ways.
Simulating Molecular Interactions
Increasingly, companies are using AI-pow-
ered simulations in place of the physical
AI AND BIOLOGY
BIOENGINEERING
A close-up view of fungal hyphae and spore struc-
tures. As AI and biology converge, it will be possible
for scientists to iterate on nature’s designs.

AI AND BIOLOGY
TECH
the human body using computer modeling
and generative AI. Spatial biology’s core lies
in examining cells and tissues within their
natural 2D or 3D habitats, akin to how GPS
technology pinpoints locations. It leverag-
es cellular and molecular data to map the
intricate architecture of cells, allowing for a
much more comprehensive understanding of
cellular interactions in their environment—a
perspective traditional methods like bulk
sequencing cannot offer. Why bother exam-
ining cells in super-high resolution? Because
a closer look will reveal complex cellular
interactions and functions, with the potential
to reach molecular or atomic level insights
as the technology improves. And increasingly,
spatial biology will produce complex data,
and companies will need advanced algo-
rithms to help mine it for insights. The field
is at the cusp of transforming our under-
standing of cellular mechanics and disease
pathology. Just as the James Webb Telescope,
with its super-high-resolution images, is
changing our understanding of the universe,
the technological advancements, improved
automation, and sophisticated data analy-
sis capabilities researchers will gain from
spatial biology will transform our under-
standing of life. The full potential of spatial
biology in diagnostics and treatment devel-
opment will usher in a new era of precision
medicine. This represents not just a leap in
medical science but also a paradigm shift
in our approach to health care and disease
management.
Emerging biotechnology techniques will allow scientists to create hyphae-like structures for any number
of new purposes.
Image credit: Future Today Institute and Dall-E.
BIOENGINEERING

GENE EDITING
CRISPR
TECH BIOENGINEERING

TECH
CRISPR Treatment for Diseases
Both the US Food and Drug Administration
and the UK medicines regulator have finalized
their groundbreaking approvals of the first
CRISPR gene therapy for Casgevy, a thera-
py built on the CRISPR-Cas9 gene editing
tool. Casgevy targets sickle cell disease and
β-thalassaemia, conditions characterized by
severe pain and the need for regular blood
transfusions. These approvals will pave the
way for future CRISPR therapies to potentially
cure a wide range of genetic diseases. CRISPR
technology, which earned its creators Dr. Em-
manuelle Charpentier and Dr. Jennifer Doudna
the Nobel Prize in chemistry in 2020, is used
in Casgevy to correct gene mutations respon-
sible for sickle cell disease and β-thalassae-
mia. These mutations affect hemoglobin
in red blood cells, leading to serious health
complications. To administer the treatment,
doctors first draw blood-producing stem cells
from patients and then use CRISPR-Cas9 to
precisely edit the hemoglobin genes. Casgevy
targets the BCL11A gene, which normally
Next-Generation Gene Editors
While CRISPR-Cas9 has been groundbreaking
in enabling precise DNA cutting, its applica-
tion is somewhat limited to diseases that
can be addressed by gene disruption. The
future of gene therapy lies in more versatile
gene editing tools that offer capabilities
such as activating genes, altering individ-
ual DNA bases, or introducing new genetic
sequences.
What’s on the horizon: an expansion of the
CRISPR toolkit, including base editing, prime
editing, and epi-editing. Last year’s regulato-
ry approval of CRISPR/Cas9 therapies opened
a new pathway to help next-gen gene edit-
ing technologies enter clinical trials faster,
because the groundwork for navigating the
regulatory and technological complexities of
novel gene therapies has now been estab-
lished. We are finally moving from ambition
and experimentation to practical application.
inhibits the production of fetal hemoglobin.
By disrupting BCL11A, Casgevy triggers the
production of a stable form of hemoglobin,
alleviating the symptoms of these diseases.
The treatment process involves preparing
the patient’s bone marrow to receive the
modified cells and a subsequent period of
hospitalization to ensure the cells’ effective
integration. While the treatment is effec-
tive, its future widespread use is still in
doubt. CRISPR therapies are expensive and
difficult to scale, due to the complexity and
technological requirements of treatments,
including the extraction, modification, and
reinsertion of blood stem cells. In the near
term, this will likely hinder their use in low-
and middle-income countries. While Casgevy
could serve as a catalyst for further RD
in gene editing and potentially transform
the future of medical treatments, there is a
pressing need for continued innovation and
investment to make such groundbreaking
therapies more universally accessible.
GENE EDITING CRISPR
BIOENGINEERING
In the future, scientists will coax enzymes to
interact with DNA bases to produce more precise
molecular activity.

TECH
GENE EDITING CRISPR
BIOENGINEERING
New Editing Tools
Base editing
First developed in 2016 by Harvard biochemist David Liu, base editing allows for the precise
alteration of individual DNA letters. This technique has shown promise in preclinical studies
for diseases like muscular atrophy and muscular dystrophy. Beam Therapeutics, co-founded
by Liu, initiated the first US clinical trials in base editing for leukemia treatment, with initial
data expected in 2024. Meanwhile, Verve Therapeutics’ in vivo base editing trial for high
cholesterol has shown encouraging preliminary results, potentially revolutionizing coronary
disease treatment.
Prime editing
Also conceived in Liu’s lab, prime editing offers even greater precision, capable of deleting
or adding small DNA segments. As impressive as CRISPR is, it can sometimes change the
wrong genes or accidentally break apart strands of a DNA’s double helix. The refinement to
CRISPR affords more precision and versatility. Prime editing has demonstrated success in
correcting various inherited diseases in animal models.
Epi-editing
Epi-editing is yet another novel CRISPR application that modifies the epigenome to regulate
gene expression without altering the underlying DNA. This approach has already gained at-
tention for its potential in treating diseases by editing gene expression regulatory networks.
Durham, North Carolina–based Tune Therapeutics has shown promising results in gene
silencing using epi-editing, and several companies, including San Francisco–based Epic Bio,
plan to commence clinical trials in the near future.
Next-generation gene editors will help expand the CRISPR toolkit, enabling scientists to perform more
precise edits.

TECH
In Vivo Gene Editing
It is now possible to inject CRISPR compo-
nents directly into the body and make chang-
es to genetic material in vivo, or “within the
living.” Unlike “ex vivo” editing, where cells
are modified outside the body and then rein-
troduced, in vivo editing involves introducing
the gene-editing tools (like CRISPR-Cas9) in
the body using viral vectors or lipid nanopar-
ticles. Practically speaking, this means that
treating cancer would no longer require
traditional chemotherapy—instead, cancer-
ous cells would be targeted and edited with
CRISPR. The FDA approved a phase 3 trial by
New York–based Regeneron Pharmaceuticals
and Cambridge, Massachusetts–based Intel-
lia Therapeutics, which was co-founded by Dr.
Jennifer Doudna. Their therapy, NTLA-2001, is
a groundbreaking one-time intravenous treat-
ment designed to target liver cells and specif-
ically aims to disable a gene responsible for
a progressive and fatal disease. Early clinical
trials of NTLA-2001 have shown remarkable
results, reducing disease-causing protein
levels by up to 93%, with these reductions
remaining stable for at least nine months.
to specific proteins on the surface of cancer
cells. Researchers at the Nanfang College
of Sun Yat-sen University and Huazhong
University of Science and Technology (both
in China) loaded nanocarriers with a new
set of genetic instructions and successfully
regressed leukemia in a mouse. The second
technique involves engineering synthetic
gene circuits in order to protect healthy cells
when delivering cellular therapy. CAR T cells,
or white blood cells that have been geneti-
cally modified in a lab to help fight cancer
more effectively, can be lethal to cells they
come into contact with, whether they’re can-
cerous or not. A new method of controlling
cell therapy, using engineered networks,
would offer doctors better precision.
These promising outcomes have also been
replicated in a variant of the disease af-
fecting heart tissues. Though the trial isn’t
expected to deliver concrete results until
2027, its impact is already being felt. This is
the first published instance of CRISPR being
used directly in the bodies of a large group
of individuals, marking a pivotal moment in
the field of gene editing. While this clearly
means a big leap forward, it also raises con-
cerns about potential off-target effects and
the implications for germline cells, which are
critical for reproductive functions.
Cell Therapy 2.0
For more than a decade, researchers have
transplanted healthy, viable cells to replace
or repair damaged ones. Most notably, cellu-
lar therapy has shown promise in helping a
person’s immune system fight cancer. But
cellular therapy carries associated risks,
which range from flu-like symptoms to
death. The field is evolving, and two emerg-
ing techniques are pushing cell therapy into
its next era. One is in vivo cell therapy, which
helps patients produce cells that can bind
GENE EDITING CRISPR
BIOENGINEERING
Cells have complex structures that can be studied
and manipulated using bioengineering techniques.

READING
SEQUENCING
GENOMES
TECH BIOENGINEERING

TECH
Next-Generation Genome Sequencing
When the first human genome was se-
quenced in 2003, it cost roughly $2.7 billion
and took 13 years to complete. In 2012, it cost
about $10,000 for researchers to sequence
a full genome, and today, you can sequence
your genome from the comfort of your home
for less than the price of a Black Friday TV
deal. The next generation of sequencers will
offer a monumental leap forward in speed
and efficiency, akin to the transition from
dial-up to high-speed internet. On the hori-
zon: ion semiconductor sequencing, which
converts chemical information into digital
data; nanopore sequencing, a technique that
reads molecular letters through tiny na-
no-size pores; and single-molecule real-time
sequencing, which observes the DNA replica-
tion process in real time. Ion Torrent, made by
Thermo Fisher Scientific, automatically turns
the basic building blocks of DNA (represent-
ed by the letters A, C, G, T) into a form that
computers can understand (0s and 1s)—and
it does it right on a tiny semiconductor chip.
This method combines straightforward
chemistry with advanced chip technology
in its approach, are pushing the entire field
of sequencing forward.
Metagenomics
Metagenomics represent a new approach
in a genomic analysis. Simply put: imagine
metagenomics is like dealing with one box
full of 10 different jigsaw puzzles. In this
analogy, each puzzle represents the DNA of
a different organism living in a particular
environment. The challenge of metage-
nomics is to sort out these pieces and put
together each individual puzzle correctly. As
researchers are considering new therapies or
trying to understand how a virus or pathogen
works, they need contextual data to under-
stand cause and effect. New metagenomics
tools help scientists solve several puzzles at
once to understand the diverse range of life
forms coexisting in a specific environment.
This is crucial for gaining insights into how
these microorganisms interact with each
other, with humans, and with the environ-
ment. It’s a complex task but offers valuable
information for various applications, from
health care to environmental science. For
example, metagenomics can detect viruses on
food items, like identifying viral contamina-
tion on lettuce. This helps trace the source of
microbial and viral contamination, improving
food safety. It’s effective in cleaning up pollut-
ants, by helping to identify microorganisms in
polluted environments that can degrade toxic
substances more efficiently than other meth-
ods. And it’s being used to identify how mi-
croorganisms compete and communicate in
different environments, from human digestive
tracts to deep-sea vents. Israel-based BiotaX
developed TaxonAI, a platform that can collect,
analyze, and predict multiple disease states
and calculate optimal interventions support-
ed by metagenomic AI analysis. Chile-based
KITAI’s lab-on-a-chip combines AI, microfluid-
ic, and metagenomics technologies to identify
biological pests, monitor water sources, and
analyze environmental pathogens.
Faster Gene Synthesis
Synthesis transforms digital genetic code into
molecular DNA, allowing scientists to design
and mass-produce genetic material. Twist
Bioscience is a pioneer in the field; it’s formed
offering a new way of reading DNA that’s not
only easier and faster but also more afford-
able and adaptable than what we’ve had.
Just as the microprocessor revolutionized
computing, shifting us from large, cen-
tralized mainframes to personal desktop
computers, semiconductor technology is
set to make DNA sequencing much more
accessible, allowing even small labs and
clinics to have this powerful tool at their
disposal. This could transform many indus-
tries, just like semiconductor chips did for
electronics. Oxford Nanopore makes devices
for nanopore sequencing that relays infor-
mation in real time. Unlike traditional DNA
sequencing methods, where scientists have
to wait until the end of the process to get any
data, nanopore sequencing lets them see
the results as they happen. This is great for
urgent situations, like identifying harmful
bacteria or viruses quickly. And once scien-
tists have the information they need, they
can stop the sequencing. This means labs
can clean and reuse their equipment (called
a flow cell), which is both time-efficient and
cost-effective. These methods, each unique
READING SEQUENCING GENOMES
BIOENGINEERING

TECH
as many as 300 base pairs of DNA, and these
snippets, or oligos, can be joined together
to form genes. Both the price for oligos and
the time to produce them is decreasing—
while base pairs are getting even longer and
more complex. It now costs an average of
just 7 cents per base pair—a 22% decrease
year-over-year. Twist’s DNA snippets can be
ordered online and shipped to a lab within
days; the synthetic DNA is then inserted into
cells to create target molecules, which are the
basis for new food products, fertilizers, indus-
trial products, and medicine.
Quantum Biology
Quantum biology is an emerging field that
combines quantum physics—the science
of the very small—with biology, the study of
living things. Researchers apply the princi-
ples that govern subatomic particles and to
understand how living organisms work at a
fundamental level. For business leaders, this
matters because quantum biology has the
potential to revolutionize various industries.
It can lead to breakthroughs in medicine,
by improving drug design or understanding
they had repeating segments or were sim-
ply too challenging to be recognized and
cataloged. As technology improves, so will
our ability to map a more detailed version of
human life on a granular scale. The Telo-
mere-to-Telomere Consortium, aptly named
after what’s called the end caps of chromo-
somes, published a new set of papers in
2022 that identified all but five of the hidden
areas of the map. Using various sequencing
technologies, including a novel nanopore
device capable of reading 100,000 bases at
a time alongside a sequencer with improved
accuracy, researchers discovered new areas
for gene evolution. In 2024 and beyond,
scientists will gain new insights into regions
of the human genome that haven’t been
fully explored, and that should in turn reveal
discoveries about human evolution, longev-
ity, and resiliency. Meanwhile, the National
Institutes of Health has initiated a ground-
breaking program with an initial investment
of $6.4 million to establish Diversity Centers
for Genome Research at three institutions:
the University of Texas Rio Grande Valley,
Meharry Medical College, and the University
of Hawai‘i at Mānoa. With this move, NIH is
targeting colleges and universities with a his-
tory of serving underrepresented communities
that haven’t recently received significant NIH
funding; the goal is that these centers will
enhance the universities’ research capabili-
ties and inspire students from diverse back-
grounds to pursue genomics. This initiative,
part of the National Human Genome Research
Institute’s $32.7 million commitment over the
next five to seven years, reflects a strategic
move to diversify the genomics field, recogniz-
ing that diverse perspectives spur creativity
and innovation.
Unlocking Bioinformatics Data
Rapid advancements in technology and a
steep decline in sequencing costs are advanc-
ing the use of bioinformatics data. Scientists
use this data—biological information stored
digitally, primarily focusing on genetic and
molecular data—to investigate all sorts of
questions: How do certain diseases affect our
bodies at the molecular level? Can we design
new medicines to treat these diseases? How
do different species evolve and adapt to their
diseases at a molecular level. In technology,
it could inspire new, more efficient ways of
data processing and energy storage. It’s an
exciting frontier that blends the most basic
elements of our universe with the complexi-
ty of life, opening up a world of possibilities
for innovation and advancement in multiple
fields. One experiment has already yielded
results: At the Johns Hopkins University
Applied Physics Laboratory in Maryland,
researchers found striking similarities be-
tween an enzyme central to human metab-
olism and a magnetically sensitive protein
found in birds. This deepens our understand-
ing of magnetosensitivity—but in practical
terms, it also potentially transforms our
approach to studying biological navigation
mechanisms.
Using the Human Genome Map
When the first human genome was deci-
phered two decades ago, it was mostly—but
not entirely—complete. That’s because
roughly 200 million base pairs of DNA, or
about 8% of the human genome, weren’t yet
readable by sequencing machines because
BIOENGINEERING

TECH
environments? But there are challenges in
understanding it. Sequencing an individual’s
entire genome now generates a staggering
100 gigabytes of raw data, a figure that more
than doubles post-analysis with the applica-
tion of deep learning and natural language
processing tools. This will result in a deluge of
data that experts estimate will need 40 exa-
bytes for storage by 2025—that’s eight times
the storage required for every spoken word in
human history. Genome analysis pipelines
are struggling to keep pace with this explo-
sion of data. The complexity and computa-
tional intensity of sequencing analysis, which
involves myriad steps to identify genetic
variations, are monumental tasks requiring
sophisticated technological solutions.
Recent advances in deep learning and AI gen-
erally are significantly improving the process
of DNA sequencing, making it faster, more
accurate, and less expensive. Nvidia, which
makes powerful GPUs, is applying AI to both
traditional (short-read) and newer (long-read)
DNA sequencing methods. This is making it
possible to sequence human genomes with
cently analyzed the DNA of a 10,000-year-old
skeleton found in Somerset, England. The
sequence showed that he likely had dark
skin and blue eyes, a genetic combination
that might have been common millennia
ago but today is rare. This expansion of the
genetic diversity map goes beyond modern
populations, offering insights into how spe-
cies have evolved and adapted over millen-
nia. But the impact of aDNA extends beyond
biology and archaeology; it fosters cultural
and political connections, enhancing our
collective understanding of the human
journey. Sequencing ancient genomes will
help historians develop a more accurate
understanding of what society might have
been like thousands of years ago and how
we compare today. As technology evolves,
aDNA research not only allows us to revisit
the past, it holds the key to unlocking future
discoveries, making it an invaluable asset
in the quest for knowledge about the human
experience.
Programmable Gene Editing Proteins
Researchers at the Massachusetts Institute of
Technology found that certain eukaryotic or-
ganisms (like plants, animals, and fungi) have
special enzymes that can cut DNA, similar to
how CRISPR technology works. These enzymes,
which seem to be related to some CRISPR pro-
teins, could potentially be used to edit human
DNA, which means they could play a big role in
future medical treatments and research. One
team at MIT, led by Feng Zhang, focused on
systems in eukaryotes called OMEGAs (Obli-
gate Mobile Element Guided Activity), which
could move small bits of DNA throughout
bacterial genomes. They discovered proteins
called Fanzors in various organisms, which
are capable of editing DNA. These Fanzors are
smaller than typical CRISPR proteins, making
them potentially easier to use in therapies.
Although they’re not as efficient as current
CRISPR methods yet, the team has already
improved their performance significantly. This
research could lead to new ways of editing
genomes more efficiently and perhaps with
fewer side effects.
high accuracy at a much lower cost. Reading,
sequencing, and analyzing bioinformatics
data using technological breakthroughs
have practical, real-world applications, such
as quickly identifying genetic disorders in
newborns or discovering new targets for
drug development.
Sequencing Ancient Genomes
Archaeology and genetics are merging,
hoping to surface new insights about the
history of life on Earth. The field of ancient
DNA (aDNA) research is starting to uncover
a wealth of insights, from identifying new
branches of the human family tree to reveal-
ing the genomes of long-extinct species. For
example, recent breakthroughs have traced
the origins of the Black Death to present-day
Kyrgyzstan and revealed lost Indigenous
populations. The sequencing of ancient ani-
mals and humans, from woolly mammoths
to Neanderthals, has provided a genetic
window into bygone eras. The 1000 Ancient
Genomes project, led by Pontus Skoglund at
the Francis Crick Institute in London, re-
BIOENGINEERING

BIOPRINTING,
ORGANOIDS,
AND NOVEL
ORGANISMS
TECH BIOENGINEERING

TECH
Bioprinting Electronics
In a groundbreaking development that blurs
the lines between biology and technology,
researchers at UK-based Lancaster Univer-
sity successfully 3D printed glowing shapes
inside nematode worms, demonstrating the
potential to embed electronics directly within
living organisms. The team leveraged a pho-
tonic 3D printer and a special ink that shapes
and activates the material within the organ-
ism. By feeding this ink to nematode worms,
the team was able to create intricate conduc-
tive circuits in the form of stars and squares
inside the living worms. This technique
suggests potential for improving traditional
electronic implants, such as pacemakers and
bionic ears, which have transformed medi-
cal treatments but come with their own set
of challenges, including infection risks and
maintenance difficulties. The Lancaster Uni-
versity team’s work is part of a growing trend
in bioprinting electronic implants and com-
puter-brain interfaces, which could replace
the medical devices we use today.
printed human hearts will be transplanted
into live pigs to see if they can keep the ani-
mals alive. Meanwhile, scientists at Harvard
University’s Wyss Institute have developed a
new 3D bioprinting technique for tissue. This
method creates thick, vascularized tissues
using living human cells consisting of a spe-
cial silicone mold to shape and support tis-
sue on a chip. In the mold, they first print a
network of large blood vessel channels with
endothelial cells in silicone ink. Then, they
add another layer with mesenchymal stem
cells in a different print. Finally, they fill the
remaining spaces with a liquid containing
fibroblasts and extracellular matrix, creating
a connective tissue that strengthens the
whole structure. Tissues are about 10 times
thicker than those made before and can last
up to six weeks.
Fabricating Organoids
It’s difficult and dangerous for scientists
to study how living human tissue responds
to viruses, medications, or other stimu-
li, because brain or heart tissue can’t be
removed from a living person. As an alterna-
tive, scientists are creating organoids—tiny
three-dimensional, multicellular clusters
grown from human stem cells that resemble
complex tissues like the heart and kidney. In
December 2023, scientists at Weill Cornell
Medicine used an organoid model to identify
a new pancreatic cancer treatment. A month
later, scientists at the Princess Máxima Cen-
ter for Pediatric Oncology in the Netherlands
successfully grew tiny brain organoids in a
dish from human fetal brain tissue. They also
revealed that the tiny blobs of tissue could
be reprogrammed to have certain diseases,
in order to study developmental disorders or
brain cancers. Scientists are already exper-
imenting with transplantation: In separate
experiments, researchers at Stanford and the
University of Pennsylvania successfully trans-
planted human brain organoids into damaged
rat brains. The organoid made connections to
the rest of the brain and responded to flashing
light stimuli. This raises both complex ethical
concerns and, perhaps, fears of a day when
super-rats emerge that can process informa-
tion as well as humans. This area of research
is controversial in some countries, including
Bioprinting and Tissue Engineering
There is a critical shortage of organ dona-
tions, and until now, the only path to organ
transplantation involved matching with a
donor, making sure the recipient’s immune
response doesn’t reject the organ, and mit-
igating the risk of infection. An emerging
solution is organ bioprinting, which leverag-
es stem cell technology to fabricate organs
tailored to the recipient’s cellular profile, and
in turn, should reduce the risk of rejection.
Researchers at Stanford University received
a landmark federal contract from the Ad-
vanced Research Projects Agency for Health
to grow human organs inside of bioreactors,
which are machines that provide a biologi-
cally active environment where cells, tissues,
or microorganisms can be grown or main-
tained under controlled conditions. The team
will grow all the cell types needed to produce
a human heart inside the bioreactor and
eventually feed the cells into a bioprinter to
fabricate a fully functional human heart. It’s
estimated that bioreactors could produce
needed cells by the billions, and eventual-
ly print a heart every two weeks. This year,
BIOPRINTING, ORGANOIDS, AND NOVEL ORGANISMS
BIOENGINEERING

TECH
the US, where bills introduced into both the
Senate and in many state legislatures call for
a ban on any research involving fetal tissue,
even if it was cultivated in a lab.
Growing Organoids to Study Long COVID
Organoids are being used to research the
lasting effects of SARS-CoV-2, the COVID-19
virus, in addition to other respiratory diseas-
es. Miniature brains, lungs, guts, and livers
are being grown in high-security labs and
infected with the virus, as are combinations
of different organs to test therapies and the
lasting impacts of long-haul Covid. Scientists
at the Karolinska Institute in Sweden infected
brain organoids with SARS-CoV-2 and discov-
ered that brain fog could be caused by the
destruction of connections between neurons.
Neurobiologists at the UK’s MRC Laboratory of
Molecular Biology in Cambridge used organ-
oids to learn that SARS-CoV-2 damages the
protective barrier of the brain. Meanwhile, sci-
entists at the Global Health Institute at Swiss
Federal Institute of Technology in Lausanne
are studying a harmful bacteria called Pseu-
domonas aeruginosa, which can cause tough-
processes, and physiological responses.
It turns out that these chips are better at
predicting real-world responses in humans
than the animals typically used in the lab.
Researchers in South Korea developed an
artificial nervous system that can simulate
a conscious response to external stimuli. It
includes an artificial neuron circuit, which
acts like a brain; a photodiode that converts
light into electrical signals; and a transistor
that acts as a synapse. All these compo-
nents are connected to a robotic hand. Think
of this as “wetware” rather than computer
hardware. This type of a system could help
people with certain neurological conditions
regain control of their limbs. It could even-
tually be worn or even embedded. Emulate,
a company that makes OoCs, tested 870
human liver-chips across a blinded set of
27 drugs with known toxicity issues—and
the chips did a better job of predicting drug
safety than the usual methods of studying
drug interactions. A team of bioengineers
at Harvard made a vagina-on-a-chip using
donated vaginal cells. The chip successfully
mimicked the vaginal microbiome and is
to-treat pneumonia by forming biofilms, or
thick layers, in our lungs. Understanding how
these biofilms develop has been difficult. To
better study this, the researchers grew mini
lung-like structures from stem cells, called
AirGels, that mimic the actual environment
of our airways, including the presence of
mucus and the air-liquid interface found in
our lungs. The team discovered that Pseudo-
monas aeruginosa quickly forms biofilms in
connection with lung mucus by pulling the
mucus together using tiny, retractable fila-
ments, called type IV pili. This study shows
that while mucus normally protects our lung
cells, it can also provide a place for harmful
biofilms to grow.
Organ-on-a-Chip
Picture something like a computer chip
but with a transparent circuit board that’s
connected to a biological system pumping
a blood substitute through tiny blobs of
tissue. Organ-on-a-chip systems (OoCs) are
synthetic organs made of multichannel,
three-dimensional microfluidic cell culture
technology that promotes organ functions,
BIOENGINEERING
Originally intended for manufacturing, 3D printing
techniques are being applied for human tissue
production and printable drugs.

TECH
actually more realistic than other existing
models currently used in labs. OoC academic
research and startups are attractive to both
venture funding and foundations, which view
the technology as foundational to new drug
discovery.
3D Printed Drugs
Drug manufacturing today requires enormous
facilities and doesn’t allow for customiza-
tion. Emerging techniques using 3D print-
ing technology could improve the variety of
medicines offered to patients. For example,
Chinese bioprinting company Triastek devel-
ops 3D printed medicines and operates the
production facilities to manufacture them.
Laxxon Medical, based in New York, developed
3D screen printing technology that results in
medicines in a variety of shapes intended for
oral, transdermal, and implantable formats. Or
what about printing medicine at home? Back
in 2012, Leroy Cronin, a chemist at the Univer-
sity of Glasgow in the UK, published a paper in
the journal Nature Chemistry describing “re-
actionware,” which are 3D-printed chemical
vessels containing the components needed
Bacterial Nanosyringes
In an emerging advancement bridging
microbiology and medicine, researchers are
transforming bacteria into nanosyringes
capable of targeting human cells for precise
protein delivery. This innovative approach,
redefining the boundaries of targeted medi-
cal treatments, could dramatically improve
the effectiveness and safety of therapies for
many different health conditions, including
cancer. Some of our most powerful drugs are
made up of small molecules that indiscrimi-
nately enter cells and cause unintended side
effects. Large molecules like proteins could
offer targeted and potent therapeutic ben-
efits, but have one big challenge: they can’t
get through cell membranes. This is where
the bacterial nanosyringes come into play,
offering a solution already found in nature.
Bacteria like Photorhabdus have evolved
cylindrical structures that function like mi-
croscopic syringes, injecting their contents
directly into targeted cells. Researchers at
the Zheng Lab at MIT, led by Joe Kreitz and
his team, managed to harness this natural
mechanism, using Google DeepMind’s Al-
phaFold AI program to adapt nanosyringes to
bind to specific human proteins. This break-
through technique has already demonstrat-
ed its potential in lab settings, successfully
delivering various proteins to targeted human
cells and even to neurons in mice.
Using Viruses to Deliver Big DNA Payloads
Bacteriophages, also known as phages,
are viruses that infect and replicate only in
bacterial cells. They are ubiquitous in the
environment and are recognized as the most
abundant biological agent on earth. Last year,
researchers modified a phage to deliver 20
times more DNA to human cells than has
ever been possible before in gene therapies.
This breakthrough, led by Dr. Venigalla Rao
at The Catholic University of America, could
unlock new frontiers in cell and gene thera-
pies, enabling complex, multifaceted modifi-
cations to human cells in a single treatment
step. The virus, equipped to carry DNA strands
up to 171,000 base pairs in length, offers an
unprecedented capacity to transport not only
large DNA sequences but also over a thousand
additional molecular components like RNAs
to carry out specific reactions. Adding the
right starting compounds would set off a
reaction resulting in the desired end product.
A decade later, Cronin and his team have 3D
printed a series of connected containers that
perform a variety of chemical reactions. This
setup involves 12 different steps, including
filtering and evaporating solutions. By care-
fully adding specific chemicals and liquids
at the right times and in the correct order,
they transformed basic, easily accessible
ingredients into a muscle relaxant known
as baclofen. Additionally, by changing the
design of these containers and using dif-
ferent chemicals, they were able to produce
other medications, such as an anticonvul-
sant and a drug used to treat ulcers and acid
reflux. But it’s not clear yet if the authorities
that check drug safety will agree to this new
method of making medicines. Regulators
like the FDA would have to change their safe-
ty rules: Rather than just checking the place
where drugs are made and the drugs them-
selves, they would need to make sure that
the new equipment used for making drugs
actually makes the right medicine.
BIOENGINEERING

TECH
and proteins. This capability could transition
gene therapy from mere treatment to po-
tential cures, especially for complex genetic
conditions like muscular dystrophy, which
have previously been hindered by the DNA
size limitations of existing viral vectors.
Minimum Viable Lifeforms
To understand how new organisms might be
created, scientists have long pursued devel-
oping an MVP—minimum viable product—for
biology. In 2010, scientist J. Craig Venter
and his team announced an astonishing
discovery: They could destroy the DNA of an
organism called Mycoplasma capricolum
and replace it with DNA they had written on a
computer that was based on a similar bac-
terium, Mycoplasma mycoides. They named
their 907-gene creature JCVI-syn1.0, or Syn-
thia, for short. It was the first self-replicating
species on the planet whose parents were,
technically, computers, and the project was
designed to help the team understand the
basic principles of life, from the minimal cell
up. In 2016, Venter’s team created JCVI-syn3.0,
a single-celled organism with even fewer
genes—just 473—which made it the sim-
plest life-form ever known. But the organism
acted in ways scientists hadn’t predicted,
producing oddly shaped cells as it self-repli-
cated. Scientists came to believe that they’d
taken away too many genes, including those
responsible for normal cell division. They
remixed the code once again, and in 2021
announced a new variant, JCVI-syn3A. It still
has fewer than 500 genes, but it behaves
more like a normal cell. Now, researchers are
working to strip down the cell even further.
They developed a new synthetic organ-
ism called, M. mycoides JCVI-syn3B, which
evolved for 300 days, proving that it could
still mutate. Researchers at Osaka Metropoli-
tan University created a synthetic bacterium
capable of swimming by introducing seven
proteins into it. With minimal genetic infor-
mation, the spherical synthetic bacteria are
thought to be the smallest mobile lifeform
to date. These minimal viable organisms
will help researchers design the future of life
from first principles.
BIOENGINEERING
Organs-on-a-chip are small devices that have tiny pieces of human tissue inside them, and they are spe-
cially made to keep the tissues working like they would in the human body.
Image credit: Penn Medicine News.

BIOCOMPUTING
CYBERBIOSECURITY
TECH BIOENGINEERING

TECH
Organoid Intelligence
In February 2023, scientists formed a new
field, called organoid intelligence (OI), which
is now considered the next frontier of biocom-
puting. To meet AI’s growing computational
needs, there’s a shift away from traditional
Von Neumann architecture toward more
innovative approaches. One is neuromorphic
computing, inspired by the brain’s structure,
which efficiently handles simultaneous infor-
mation storage and processing. That’s what
led researchers at Johns Hopkins to create OI,
which uses biological materials—most often
human brain cells—for information process-
ing, leveraging their inherent capabilities be-
yond silicon-based systems. This represents
a significant step in harnessing the brain’s
natural efficiency for AI applications. Late in
2023, a biocomputing system made of living
brain cells learned to recognize the voice of
one person from a set of 240 audio clips of
eight people pronouncing Japanese vowel
sounds. The clips were sent to the organoids
as sequences of signals arranged in spatial
patterns. Why bother inventing technology
that sounds like it was inspired by a dystopi-
squares—and after eight sessions, the bac-
teria played at an expert level. While the bac-
teria haven’t yet beat humans at the game,
there’s an interesting analogy worth remem-
bering: The benchmarks in computing and
specifically in AI have always been gameplay.
This isn’t the only biological computer. A
biocomputer called DishBrain learned how to
play the 1980s video game Pong. DishBrain
is made of ~1 million live human and mouse
brain cells grown on a microelectric array
that can receive electrical signals. The sig-
nals tell the neurons where the Pong ball is,
and the cells respond. The more the system
played, the more it improved. Cortical Labs
is now developing a new kind of software,
a Biological Intelligence Operating System
(biOS for short), which would allow anyone
with basic coding skills to program their
own DishBrains. Further demonstrations of
new, simple forms of neural networks made
from biology will likely occur this year.
Biological Circuits
Scientists are in the process of building
biological circuits, made of synthetic DNA,
and the software that operates them. A pro-
gram called DNAr, developed at the Federal
University of Mato Grosso do Sul in Brazil,
simulates chemical reactions, while another
called DNAr-Logic enables scientists to design
circuits. A high-level description of a logical
circuit is then converted into a chemical-re-
action network, which can be synthesized into
DNA strands. Dramatically speeding up the
design process for biological circuits could
drastically reduce the time it takes to discover
health treatments and new drugs.
Programmable DNA Machines
In a breakthrough that could redefine comput-
ing, scientists at Shanghai Jiao Tong Universi-
ty in China unveiled what might be the world’s
first programmable DNA computer, capable
of executing billions of unique circuits. This
could usher in a new era where computers
could solve complex mathematical problems
and potentially aid in diagnosing diseases.
Rather than relying on a traditional silicon
microchip-based framework, DNA computers
operate using the very molecules that have
been nature’s medium for storing life’s blue-
an sci-fi novel? As the world demands more
AI applications like ChatGPT, we’ll need more
energy-intensive computers and networks to
crunch all that data. OI might be able to per-
form all of those tasks using a fraction of the
resources required of a traditional computer.
Training Biocomputers to Learn New Skills
Scientists at the Spanish National Research
Council genetically modified a strain of E.
coli called Marionette so that it could sense
different chemicals and respond to them.
But that wasn’t all. They modified the strain
so its plasmids each encoded for a different
fluorescent protein (red and green). While
the researchers could alter the ratio of the
red and green with future chemical inputs,
without inputs, the ratio would simply stay
constant and, in a way, was a form of mem-
ory. Here’s where things got interesting: The
team grew the Marionette strain in eight
wells that correspond with the outer squares
of a grid and taught it how to play tic-tac-toe.
Initially, the bacteria played randomly, but
the Spanish National Research Council team
trained the strain by adding chemicals to the
BIOCOMPUTING AND CYBERBIOSECURITY
BIOENGINEERING

TECH
prints for eons. After all, biology has a code—
ACTG—not unlike binary code (1s and 0s) in
conventional computing. But while biology
can be expressed in code, DNA is organic—and
molecules have a habit of moving around ran-
domly. For that reason, the researchers took
inspiration from origami and designed DNA
sequences to fold into specific shapes, allow-
ing them to function like parts in a computer,
guiding the flow of data. In experiments, they
built a DNA computer with 30 logic gates
and 500 DNA strands, capable of calculating
square roots and identifying genetic markers
of kidney cancer.
To be fair, this new DNA computer takes
hours to perform simple computations and
won’t replace regular computers anytime
soon. But the research does hold promise for
certain biomedical applications. For example,
a DNA machine could detect specific genes
and respond with a DNA strand that triggers
biological reactions, useful in environmental
monitoring or disease treatment. What’s next:
cajoling DNA to perform complex algorithms
and disease diagnosis.
storage method involves converting digital
data into DNA sequences made of the ACTGs
(adenine, cytosine, guanine, and thymine)
you learned about in high school. The pro-
cess, which takes about eight hours for 1
kilobyte, involves chemically synthesizing
a unique DNA strand to match the desired
sequence, then drying and sealing it on a
chip to protect against oxygen. To access
the stored data, customers must send one
of their DNA cards to Biomemory’s partner,
US-based Eurofins Genomics. The retrieved
data, in the form of DNA sequences, is then
emailed back and can be decoded using
Biomemory’s DNA translation feature. Cards
aren’t rewritable—meaning, they can only be
decoded once.
The Intelligence Advanced Research Proj-
ects Activity, a group in the Office of the
Director of National Intelligence, intends to
store an exabyte of data—roughly a million
terabyte-size hard drives—in a blob of DNA. A
weird branch of biological science, yes, but
human computing has practical purposes:
DNA could solve our future data storage
problems. It’s durable, too: Evolutionary sci-
entists routinely study DNA that is thousands
of years old to learn more about our human
ancestors. In China, scientists at Tianjin
University stored 445 kilobytes of data in an E.
coli cell. In the US, Twist Bioscience is making
hyperdense, stable, affordable DNA storage by
using robots to create a million short strands
of DNA at a time from microscopic drops of
nucleotides on silicon chips. The end result
will be a tiny, pill-size container that could
someday hold hundreds of terabytes of ca-
pacity. Now, a consortium called the DNA Data
Storage Alliance is developing an interoper-
able storage ecosystem using DNA as a data
storage medium. Founders include Microsoft,
as well as Western Digital, Twist Bioscience,
and Illumina. Members of the Alliance, includ-
ing Los Alamos National Laboratory, Seagate,
FujiFilm, Dell Technologies, Lenovo, IBM, and
the University of Arizona’s Center for Applied
Nanobioscience and Medicine are hoping to
write megabytes of data per second on syn-
thetic DNA that will be readable for thousands
of years.
Using DNA to Store Data
In 2018, scientists from Microsoft Research
and the University of Washington achieved a
new milestone: They discovered how to cre-
ate random access memory on DNA at scale.
They encoded 200 megabytes of data on hu-
man DNA—including 35 video, image, audio,
and text files ranging from 29 kilobytes to 44
megabytes. In 2021, the team built a molec-
ular controller and DNA writer on a chip, with
a PCIe interface. Microsoft used the system
to store a version of the company’s mission
statement in DNA: “Empowering each person
to store more!” Flash forward to 2024, and
Paris-based startup Biomemory has intro-
duced DNA cards, a new form of data storage,
priced at $500. Each card offers one kilobyte
of storage, equivalent to a short email. This
could be useful for anyone wanting to save
ultrasensitive data. DNA storage is recog-
nized for its remarkable longevity, with a
potential lifespan of hundreds of thousands
of years in optimal conditions, far exceeding
traditional storage devices like hard drives.
(Biomemory’s DNA cards guarantee a mini-
mum lifespan of 150 years.) This innovative
BIOENGINEERING

TECH
Biological Robots
In 2020, a cluster of stem cells from an
African clawed frog served as the base for a
fortuitous experiment involving a supercom-
puter, a virtual environment, and evolutionary
algorithms. Researchers created 100 genera-
tions of prototypes before they had a tiny blob
of programmable tissue called a xenobot.
These living robots can undulate, swim, and
walk. They work collaboratively and can even
self-heal. And they’re tiny enough to be inject-
ed into human bodies, travel around, and—
maybe someday—deliver targeted medicines.
While technically they’re made up of living
cells, researchers are quick to point out that
xenobots lack the characteristics of a tradi-
tional biological life-form. In 2021, xenobots
got a design upgrade and new capabilities.
While before they needed the contraction of
heart muscle cells to move forward, upgraded
xenobots can self-propel using tiny hairs on
their surfaces. The current crop of xenobots
live longer, and they can sense what’s in their
environment. They can also operate in robot
swarms to complete a collaborative task. Xe-
nobots are being used to help researchers un-
that don’t already exist in natural organisms.
These characteristics will someday enable
their use as a biorobotics platform for vari-
ous medical and other purposes. For exam-
ple, with some modifications, it’s possible to
build cell clusters that could be dispatched
to repair damaged tissues within the body.
Living Sensors
Research is already underway to develop bi-
osensors that can detect deadly bioweapons
on the battlefield and harmful chemicals
in factories by identifying the presence of
specific DNA sequences. Researchers at the
University of California San Diego created a
bacterium called Acinetobacter baylyi capa-
ble of detecting a single DNA letter mutation
in a gene that’s present in many cancers.
With a focus on the microbiome, they en-
gineered the bacterium to detect mutated
DNA sequences while living inside of the gut.
While still very early in development, living
sensors could someday be used to detect
viral outbreaks in a community’s sewage
system, cholera in drinking water, and other
pathogens.
derstand how defects in the hairlike struc-
tures in our lungs, called cilia, can result in
diseases. Also in progress: xenobots that can
travel to a damaged spinal cord and repair it
with regenerative compounds.
Meanwhile, another type of living robot,
anthrobots, were developed in 2022 from
donated human tracheal cells. Covered in
cilia, these anthrobots harnessed the struc-
tures like flexible oars to propel themselves
around. When grown in a petri dish, scien-
tists discovered the bots could be assem-
bled into super-anthrobots to perform tasks.
A team at Tufts University grew a sheet of
human neural cells and scratched a few off,
to create a defect roughly a millimeter wide.
With super-anthrobots on the other side,
bots catalyzed healing. While some skep-
tics claim that the significance of biologi-
cal robots are overhyped—they’re not really
programmable robots after all—it’s useful to
think about this tissue in a broader context.
Instead of viewing the cell clusters merely as
tiny tissue samples for studying human biol-
ogy, they have distinct shapes and behaviors
BIOENGINEERING
General DNA integrated circuits can solve math
problems and identify molecular biomarkers, show-
ing promise that DNA computers might one day
prove useful for clinical and diagnostic applications.

TECH
Cyberbiosecurity
Cyberbiosecurity is a relatively new inter-
section between biology, biosecurity, and
cybersecurity that will soon become a criti-
cal frontier in the safeguarding of the global
bioeconomy against burgeoning threats.
As AI continues to mesh with life scienc-
es, advancements underscore a dual-edge
potential: On the one hand, we’ll see faster
development of new vaccines, therapeu-
tics, and materials, which stand to promote
economic growth while improving health
and creating more options to battle climate
change. On the other hand, these new tools
could be misused. Technologies with AI-bio
capabilities bear the risk of triggering biologi-
cal catastrophes. As of now, there is no single
body charged with overseeing cyberbiosecu-
rity. The Nuclear Threat Initiative, a nonpar-
tisan global security organization focused
on reducing nuclear and biological threats
imperiling humanity, convened a panel of
leading experts on synthetic biology, machine
learning, bioinformatics, and international
security policy in January 2024, however there
is no formal governing organization yet. (If
this sounds somewhat familiar, it’s because
it happened before with AI. More than two
decades ago, there were calls to establish a
similar governing organization for AI. Today,
the regulatory landscape is a mishmash of
different policies often in conflict and hard
to enforce.)
BIOENGINEERING
This magnified image shows a brain organoid produced in the lab of Dr. Thomas Hartung, a professor of en-
vironmental health and engineering at the Johns Hopkins Bloomberg School of Public Health and Whiting
School of Engineering in Baltimore. The culture was dyed to show neurons in magenta, cell nuclei in blue
and other supporting cells in red and green.
Image credit: Courtesy Jesse Plotkin and Johns Hopkins University.

NEW
MATERIALS
TECH BIOENGINEERING

TECH
AI-Created New Materials
Last year, the Google DeepMind team revealed
its Graphical Networks for Material Explo-
ration (GNoME)—a sort of AlphaFold, but for
new materials. Remarkably, it predicted the
structures of 2.2 million new materials, and
some 700 have already been created in labs
for testing. GNoME has significantly expanded
the repository of stable materials to 421,000,
a nearly tenfold increase, showcasing its
efficiency in identifying viable new materials.
Here’s why this is significant: the tradition-
al methods used to discover new materials
involve experimenting with element combina-
tions from the periodic table, a slow, tedious,
and inefficient process. DeepMind uses
two deep-learning models instead: the first
generates structures by tweaking existing
materials, while the second predicts material
stability based purely on chemical formulas.
This dual-model strategy broadens the search
for new materials. GNoME’s role is to evaluate
these candidates, focusing on their decompo-
sition energy to determine material stability,
a crucial factor for engineering applications.
Newly discovered materials can be used to
able, novel solution for fire retardation within
building materials. Researchers at RMIT
University in Melbourne, Australia, have used
this technology to develop a new biological
material for fire retardancy. Unlike asbestos,
which has terrific fire retardant properties
but releases harmful particles upon com-
bustion, engineered mycelium are safe and
effective. Fire retardant sheets are grown and
then layered into protective mats only a few
millimeters thick. This innovative fabrication
process results in a material that not only
delays the ignition of building components,
but also exhibits a unique fire response
mechanism. In case of a fire, the mycelium
sheets momentarily ignite, discharging water
vapor and carbon dioxide, thereby reducing
the oxygen available for the fire to sustain
itself. This reaction culminates in the forma-
tion of a carbonaceous layer, which effectively
stops the propagation of flames. New York–
based Ecovative is also harnessing fungi
in its Mushroom Packaging, using organic
waste and mycelium grown in custom molds
without light, water, or chemicals. This pro-
cess takes five to seven days and results in a
durable, lightweight material that is heat-treat-
ed to halt growth, yielding a completely natural
and compostable packaging solution. It’s an
eco-friendly alternative to conventional styro-
foam or bubble packaging.
Mycelium Leathers
Besides fire retardancy and packaging, luxury
companies are eying mycelium as an alter-
native to leather. Hermès created a myceli-
um version of its classic luxury Victoria bag,
while Adidas launched a pair of Stan Smith
shoes made out of the leather alternative. Bolt
Threads and Ginkgo Bioworks announced a
multi-project collaboration to optimize the
production efficiency of mycelium leathers
and to develop novel proteins for biomaterials.
Research and Markets, a provider of data and
analysis, forecasts that the global market for
synthetic leather materials could approach
$67 billion by 2030. Meanwhile, the market for
bio-based leather, which comprises materials
of natural origin, was valued at approximately
$650 million in 2021, according to Polaris Mar-
ket Research. Until now, there have been many
leather alternatives made from plastics such
make better solar cells, batteries, computer
chips, and more. Meanwhile the Lawrence
Berkeley National Laboratory created a new
autonomous lab using GNoME called A-Lab.
It uses a materials database, incorporates
findings from GNoME, and employs ma-
chine learning along with robotic arms to
create new materials autonomously, with-
out human intervention. A-Lab was able to
synthesize 41 out of 58 proposed compounds
over 17 days, showcasing a much faster pace
of material creation compared to traditional
labs. This efficiency is critical in a field where
experimentation can otherwise be dauntingly
slow. AI will significantly enhance the pro-
cess of discovering and creating new materi-
als, which will ultimately lead to better solar
cells, batteries, computer chips, and more.
Modifying Fungi for Building and Packaging
In an era marked by escalating fire risks to
residential structures, synthetic biology may
help enhance the safety of buildings. Ultra-
thin sheets can be made from the expansive,
root-like networks beneath mushrooms
known as mycelium—and they may be a scal-
NEW MATERIALS
BIOENGINEERING

TECH
as polyurethane or polyvinyl chloride, common-
ly referred to as PVC, leading to the somewhat
disparaging nickname “pleather.” Mycelium
offers a viable, sustainable alternative.
Biomolecule-Based Packaging
Packaging made from natural materials like
plants or proteins is biodegradable and safe
for the environment. However, single-material
packages can have drawbacks, so research-
ers are mixing different natural substances
to make stronger, better packaging. Depend-
ing on the food and storage method, these
eco-friendly packages can protect food by
keeping out germs, preventing spoilage, and
reducing water loss, among other things.
There’s also a new type of packaging that
includes natural preservatives to keep food
fresh longer by stopping bacteria growth or
preventing the food from going bad. Another
innovative idea in development is “smart”
packaging that changes color or shows signs
to indicate how fresh the food is, helping
everyone from producers to consumers keep
track of food quality.
color when the product is no longer safe to
consume, giving consumers a clear, visual
signal about the state of their food.
Biodegradable and Edible Packaging
Smart packaging will drive agricultural
advances and investment. Biopolymers such
as polysaccharides, proteins, and lipids can
be used to fabricate edible films or coatings
as packaging. Rather than throwing away
your strawberries’ packaging, you can eat the
wrapping. University of Minnesota research-
ers are developing polymers that self-de-
struct or “unzip” when exposed to light, heat,
or acid. Saltwater Brewery designed biode-
gradable and edible plastic rings for six-
packs of beer—so sea turtles can eat them
rather than get tangled in them. Infarm cre-
ated a renewable plastic that folds around
objects. It uses seaweed-based agar-agar gel
to grow microgreens and herbs that don’t
need water. At the end of 2022, Prince Wil-
liam awarded a $1.2 million Earthshot Prize
to Notpla, a startup that uses seaweed to
produce naturally biodegradable packaging.
Durable Biofilms
A biofilm is essentially a community of bacte-
ria living together in a structured formation.
While biofilms can be beneficial, such as
in treating wastewater, they can also cause
significant problems, including damaging
infrastructure through corrosion and being
involved in up to 60% of infectious diseases.
Bacteria in biofilms become tougher, harder to
remove from surfaces, and more resistant to
antibiotics compared to their solitary counter-
parts. Researchers at the University of Roches-
ter have developed a method to create bio-
films through 3D printing. They’ve genetically
modified bacteria to produce biofilm compo-
nents, allowing these bacteria to be printed in
a hydrogel. This process forms biofilms with a
dense network that mimics natural biofilms’
structure. The technique offers precise control
over the bacteria’s distribution and density,
enabling detailed studies on biofilm behav-
ior and the creation of biofilms with specific
patterns for various applications, opening up
a wide array of practical applications in man-
ufacturing, supply chain, transportation, food
and beverage, and beyond.
Intelligent and Active Packaging
Active packaging works by incorporating
substances like antimicrobials and antiox-
idants directly into the packaging to extend
the food’s shelf life and enhance its safety
and taste. For example, a company might
use packaging with built-in antimicrobials
to keep bread fresher for longer by prevent-
ing mold growth. Intelligent packaging, on
the other hand, includes smart indicators
that show changes in the food’s condition,
such as freshness, quality, or safety. These
indicators can react to environmental
changes like temperature, humidity, or the
presence of certain gasses. Researchers at
the NOVA School of Science and Technology
in Portugal are developing bio-based sensors
made from natural extracts and biopolymers
that can act as smart food packaging, with
indicators showing various factors such as
freshness. Eventually, this could mean the
end of expiration dates, which aren’t actu-
ally connected to whether food is spoiled
or fresh. Expect to see meat packaging that
changes color if the meat starts to spoil,
or a milk carton with a label that shifts
NEW MATERIALS
BIOENGINEERING

CULTURED
FOOD DRINKS
AND NEXT-GEN
AGRICULTURE
TECH BIOENGINEERING

TECH
Lab-Grown Meat Is Going Mainstream
There is growing interest in cultivated meat,
which is produced from animal cells in a lab
or biomanufacturing plant, which offers an
environmentally friendly (and, let’s face it,
animal-friendly) alternative to traditional
meat production. Concerns about food secu-
rity underline this push into bloodless meat
cultivation. Late in 2020, Singapore approved
a competitor to the slaughterhouse by allow-
ing a bioreactor—a high-tech vat for growing
organisms—run by a US company to produce
cultured chicken nuggets for its residents.
The company, Eat Just, manufactured chicken
in bioreactors using cells taken from healthy,
live chickens. In 2023, the company opened
a 30,000-square-foot facility in Singapore,
and its bioreactors now have the capacity
to produce tens of thousands of pounds of
slaughter-free meat. By 2030, Eat Just plans
for cultured meat products to cost at or lower
the current price points for chicken, beef, and
pork. Its successful entry into Singapore, a
highly regulated country that’s also one of the
world’s most important innovation hotspots,
is accelerating interest in cultured meat’s
developed by Beyond Meat and Impossible
Foods, cell-based meat cultivation results in
muscle tissue that is molecularly identical
to animals grown for our consumption—and
in some cases, improved. Lab-grown meat
also doesn’t require the hormones and anti-
biotics used at conventional facilities.
Synthetic Milk and Cheese
Synthetic milk is coming. It’s a promising
substitute for cow’s milk, offering a similar
taste, look, and texture that plant-based
options like oat, nut, and soy milks don’t
match. Hailed as the milk of the future,
synthetic milk is considered an eco-friendly
choice that could disrupt the dairy sector
and potentially disadvantage small-scale
dairy farmers. Synthetic cow’s milk is culti-
vated by artificially reproducing the proteins
in casein and whey. Casein genes are added
to yeast and other microflora to produce
proteins, which are purified and transformed
using plant-based fats and sugars. Perfect
Day makes lab-grown dairy products—yogurt,
cheese, and ice cream—that are now sold in
thousands of US grocery stores. Remilk, an
Israeli company, has established a large-scale
production facility in Denmark dedicated to
manufacturing cheese, yogurt, and ice cream.
Nestle and Danone, two of the world’s largest
food and beverage corporations, have been on
an acquisition spree, buying lab-grown dairy
startups around the world. In the next few
years, the focus will be on scaling cultured
dairy operations and lowering costs of produc-
tion. New Culture, which makes animal-free
cheeses for pizza that stretch, melt, and of
course, taste like what you’d find at your
favorite local restaurant, upgraded its fermen-
tation process last year. It can produce 25,000
pizzas’ worth of cheese in a single run.
Precision Fermentation
Precision fermentation is an advanced version
of a very old technology: brewing. For hundreds
of years, it’s been used to multiply microbes
to create specific products, from beers to
medicines. Today, precision fermentation can
be used for a host of purposes. Food technol-
ogists can use genome sequencing and gene
editing as part of a precision fermentation
process, which results in microbes engineered
startup ecosystem. Israel is a global leader
in the cultured meat sector, with groups
like Aleph Farms, which got regulatory
approval to sell cultured steaks, and Steak-
holder Foods, which teamed up with Singa-
pore-based Umami Meats to produce fish
filets without contributing to the overfishing
of declining fish stocks. On the infrastruc-
ture side, Turkey-based Biftek is working on
new technologies and serums to reduce the
cost of lab-grown meat, and in Mexico, Micro
Meat creates technologies to scale up pro-
duction. In Israel, MeaTech uses 3D printing
to produce whole cuts of cell-based meat,
while Israel-based SuperMeat has developed
what it calls a “crispy cultured chicken.” Sev-
eral startups are bringing cultured meat to
market. Finless Foods, based in California, is
developing cultured bluefin tuna meat, from
the sought-after species now threatened by
long-standing overfishing. Other companies,
including Mosa Meat (in the Netherlands)
and Upside Foods (in California, formerly
known as Memphis Meats) are cultivating
meats in factory-scale labs. Unlike the exist-
ing plant-based protein meat alternatives
CULTURED FOOD DRINKS AND NEXT-GEN AGRICULTURE
BIOENGINEERING

TECH
for specific purposes. For example, feeding
engineered microbes into a precisely tuned
fermenter could create synthetic coconut oil or
palm oil. Genetically altered microbes, which
are already in use to produce plant-based meat
substitutes, could soon form the basis for non-
dairy cheeses. With more consumers seeking
out vegan options and climate change impact-
ing dairy production, Nestle, Danone, Mars,
General Mills, and Unilever are developing pre-
cision fermentation platforms to meet future
supply chain constraints and market needs.
Precision fermentation can also produce new
forms of stabilizers and preservatives.
Brewing Great Nonalcoholic Beers
People are becoming more interested in
healthier, more responsible ways of drink-
ing, leading to a big increase in nonalcoholic
beers. But anyone who’s tried a nonalcoholic
beer will be quick to complain that it doesn’t
taste or smell as good as beer fermented the
traditional way: The beers typically lack the
pungent, hoppy smell and can leave an odd
aftertaste. The reason has to do with how they
are made. Brewers either stop the brewing
Upgrading Photosynthesis
Genetically modifying crops with upgrades
could dramatically increase crop yields with-
out needing to increase the other resources
required for cultivation. Researchers are
working on a number of projects that would
increase photosynthesis—the biological
process green plants and some organisms
use to harness sunlight to produce energy
out of CO2 and water. Simply over-exposing
plants to sunlight doesn’t have the same ef-
fect—more light can damage cells unless they
turn on a biological system called quenching
that’s capable of flushing out the excess en-
ergy. On cloudy days, plans turn off quenching
to retain the excess energy, but the process
of turning quenching on and off is time-con-
suming, unpredictable, and inefficient. Scien-
tists hope that with genetic engineering, they
can speed up the quenching process, which
would lead to more efficient photosynthesis.
In 2022, modified soybean plants were shown
to yield 20% more thanks to a jacked-up
photosynthesis system. Researchers are also
working on cowpeas and rice.
process early to avoid creating alcohol, or
they let the beer ferment and then remove
the alcohol. Both methods tend to remove
the hop aromas, which are important for
making beer smell good. But biotech startup
EvodiaBio seems to have found a solution.
Its scientists developed a method to create
monoterpenoids, the compounds respon-
sible for the hoppy flavor, and add them to
beer after brewing, restoring the flavor that
usually gets lost. By using baker’s yeast cells
as tiny biofactories, the team can generate
these hop aromas in fermenters, avoiding
the waste of expensive hops that typically
lose their flavor during the brewing process.
This approach not only enhances the taste
of nonalcoholic beer but is also much more
eco-friendly compared to using tradition-
al hops. For example, in the US, aroma hop
farming happens mainly on the West Coast—
which means that anyone outside the area
must rely on an extensive transportation
and refrigeration cold chain, not to mention
the considerable amount of water needed
for cultivation (about 2.7 tons of water to
produce just 1 kilogram of hops).
BIOENGINEERING
Bioluminescent Firefly petunias give off an eerie
glow in the dark.
Image credit: Light Bio

TECH
Faster Flowering
Using the CRISPR gene-editing tool, research-
ers at the University of Georgia Warnell School
of Forestry and Natural Resources and at
Franklin College of Arts and Sciences figured
out how to make trees mature faster. They
used CRISPR to edit a flower repressor gene
and drastically shortened the time it takes a
poplar tree to flower—from 10 years down to
just three months. It would typically take the
plant a year to develop the systems to even
produce flowers, and the team engineered the
plant to mature in just a few days. The prom-
ise of this research is an accelerated time
frame for tree breeding, as well as to enhance
the natural defenses of trees from extreme
heat, cold, and drought.
A Return to Natural Farming—With Technology
Farming faces a big challenge: how to provide
plants with nitrogen to feed more people with-
out harming the environment or reducing crop
sizes. Nitrogen is essential but expensive, and
the usual synthetic fertilizers cause pollution
and contribute to climate change. They’re
also not very reliable because they can easily
resistance to flooding and drought. In 2017,
the Rodale Institute launched the Regen-
erative Organic Certified program to start
creating an official standard. It builds on the
USDA certified organic seal by adding soil
health, animal welfare, and human rights
requirements. General Mills announced that
it would accelerate regenerative agriculture
by dedicating a million acres of farmland
to it by 2030. Meanwhile, several brands,
including Patagonia, Timberland, Allbirds,
Gucci, and Balenciaga, have launched efforts
to promote regenerative agriculture.
A New Wave of Genetically Modified Foods
A recent Pew Research study showed that
most Americans see food using genetical-
ly modified organisms as worse for their
health than a food that has no genetic
modification at all, while just 7% see them
as healthier than other foods. GMOs have a
public perception problem because some of
the earliest modified crops (corn and soy-
beans) were genetically changed to tolerate
herbicides like glyphosate, which sells under
the brand name Roundup, and last year in
the US, 91% of domestic corn production used
these herbicide tolerant seeds. So, it’s under-
standable that people are wary of a new crop
of GMOs. But the promise of supercharged
foods enhanced to produce additional nutri-
ents—rather than modified to respond to a
particular fertilizer or pesticide—may change
people’s minds. Biofortified foods are genet-
ically enhanced to provide a denser dose of
nutrients. California-based Fresh Del Monte
created a pink pineapple that’s been modified
to have a higher level of lycopene, an antioxi-
dant that gives peaches, tomatoes, and water-
melon their rosy hues. Early in 2024, the Purple
Tomato, developed by Norfolk Plant Sciences,
was approved to be marketed directly to home
gardeners. It was the first time that genetical-
ly modified foods were available to noncom-
mercial producers in the US. The tomatoes
are bright purple thanks to color genes from
a snapdragon flower that were added to the
plant. They’re not only unusual looking, they
have high levels of anthocyanin, which has
antidiabetic, anticancer, anti-inflammatory,
antimicrobial, and anti-obesity effects and is
used to prevent cardiovascular diseases.
wash away or evaporate. Healthy soil, full of
microbes like bacteria and fungi, naturally
supports plant growth by recycling nutri-
ents, but chemicals can harm this balance.
Synthetic biology offers a solution. Pivot Bio,
a biotech company, has developed a way to
enhance a soil microbe’s ability to supply
nitrogen directly to plants, offering a steady
and environmentally friendly source of this
crucial nutrient throughout the growing sea-
son without genetic modification from other
organisms.
Regenerative Agriculture
Regenerative agriculture describes farm-
ing and grazing practices that rebuild soil
organic matter and restore degraded soil
biodiversity. There’s a clear need for this
technology-led practice: Decades of using
chemicals, salt-based fertilizers, carbon
mining, and harsh insecticides deplete soil.
Planting multiple types of crops together,
rotating crops, cutting back on tilling, and
reducing reliance on harsh chemicals can
revitalize depleted soil, leading to improved
yields, nutrient-rich crops, and improved
BIOENGINEERING

TECH
CRISPR Animals
CRISPR is making farmed animals bigger,
stronger, and (hopefully) healthier. Research-
ers at Auburn University in Alabama intro-
duced a gene from alligators into catfish, not
aiming to give the catfish alligator charac-
teristics, but rather to boost their ability to
resist diseases. This is because alligators
are exceptionally good at warding off infec-
tions, and a slight increase in resilience could
significantly impact fish farming. Currently,
about 40% of fish raised in farms globally
don’t survive until harvest, so reducing even a
fraction of this loss could be transformative.
Scientists in Japan used CRISPR to modify the
myostatin gene in red sea bream, resulting in
fish that are larger and heavier by about 17%
compared to nonmodified fish, even though
both groups were fed the same quantity of
food. Researchers have long experimented
with CRISPR on animals, and so far have used
it to create super-muscly pigs, cattle, sheep,
rabbits, and goats. But most animals did
not live past infancy, and, somewhat weirdly,
many developed unusually large tongues.
Genetically Engineered Space Farming
Growing plants in space, on the moon, and
Mars is important for keeping astronauts
healthy and happy. NASA has been looking
into this for years, focusing on building the
right equipment to grow plants, choosing the
best types of plants for space nutrition, and
studying how plants react to being in space,
including how they interact with microbes. Re-
cent advances in gene-editing technology, like
CRISPR and other tools, have made it easier
to tweak plant genes for space needs. These
tools, along with new ways to deliver gene-ed-
iting materials to plants and the use of big
data and machine learning to analyze plant
genes, are opening up possibilities for creating
plants that can thrive in space. Looking ahead,
experts believe it’s crucial to focus on using
these technologies to develop plants that meet
the specific needs of space missions, mak-
ing space agriculture more sustainable and
effective. Space agriculture is quickly becom-
ing a multibillion-dollar industry. NASA and
Germany’s space agency are now investing in
a variety of space agricultural projects that
could someday support off-planet habitats.
BIOENGINEERING
Researchers in China created double-muscled pigs by introducing a mutation into the pigs’ genetics that
keeps the muscles developing beyond the point they would naturally. Note: this image is AI-generated, and
does not show a live animal.

AGING, HEALTH,
AND BEAUTY
TECH BIOENGINEERING

TECH
Cultivated Collagen
Israel-based Aleph Farms has already brought
cultured steak to market. Their next goal:
cell-cultured collagen, which is part of a
broader strategy to replace the entire cow.
The company is developing a way to produce
collagen directly from cow cells, avoiding the
need to raise and harvest livestock. Tradition-
ally, collagen is derived by boiling cow hides
and bones, a process used across various
industries. Aleph Farms will use techniques
from its steak production, such as bovine cell
sources and a growth medium free of ani-
mal components, to create different types of
collagen that are identical to those found in
nature. This cultivated collagen will include a
full range of proteins found in the extracellu-
lar matrix, which makes up skin, bones, and
joints. There are other startups now working
on collagen cultivation, too. Jellatech, a North
Carolina–based startup, created a full length,
triple helical, bio-identical and functional
human collagen made from its proprietary
cell line.
go longer between treatments. After that,
researchers will set their sights on manufac-
turing lab-grown blood for rare blood types
that don’t typically have large donor pools.
Growing Sex Cells
Last year, Dr. Katsuhiko Hayashi from Osaka
University successfully created eggs from
cells harvested from male mice, with the
eventual goal of developing new fertility
treatments. The process begins with taking a
skin cell from a male mouse and converting
it into a stem cell, which has the potential to
develop into various cell types. Since these
cells are male, they carry XY chromosomes.
The team then removes the Y chromosome,
replicates the X chromosome, and combines
the two X chromosomes—a modification that
enables the stem cell to develop into an egg.
Hayashi’s work builds on groundbreaking re-
search from fellow Japanese scientist Shinya
Yamanaka, who in 2006 showed that it was
possible to make gametes derived from
human-induced pluripotent stem cells. Ya-
manaka’s process includes harvesting cells
from a skin biopsy or blood sample (both
quick and relatively painless). Those cells are
turned into stem cells, grown in a medium
that resembles what would exist in a human
womb, and developed into precursor sex cells,
which mature into sperm or stem cells. Then,
once IVF kicks in, those cells are used to create
an embryo. One or more of the healthiest em-
bryos are then implanted into the uterus and,
if all goes well, develop into a healthy, viable
fetus. The idea is that someday soon, couples
suffering from infertility or individuals who
desire to have a baby without a partner would
have access to a reliable fertility treatment.
Human Trials of Synthetic Wombs
Researchers at the Children’s Hospital of
Philadelphia (CHOP) created an artificial womb
called a biobag and used it to successfully
keep premature lambs alive and developing
normally for 28 days. Now, CHOP researchers
are seeking approval to begin the first human
clinical trials for a device they’ve developed,
called the EXTra-uterine Environment for Neo-
natal Development, or EXTEND. The team has
clarified that this technology is not designed
or capable of supporting the full spectrum of
Growing Blood
For people who live with rare blood types
(AB negative, AB positive, B negative) or who
have blood disorders, acquiring blood for
surgery or a transfusion can mean the dif-
ference between life and death. For decades,
scientists have attempted to grow blood
cells in a lab at scale, but until recently, the
process has failed to produce enough blood
cells to make an impact. But late in 2022,
scientists at the National Health Service
Blood and Transplant in the UK announced
that they had grown red blood cells in a lab
and successfully transfused them into a
living person, a world first. It took 500,000
stem cells to generate 50 billion red blood
cells, which then needed to develop. (In a
healthy adult, 50 billion red blood cells rep-
resents about 1% of their total blood volume.)
Last year, the same researchers transfused
red blood cells that were grown in a lab into
another person requiring that blood. This
technique is a pioneer in transferring lab-
grown cells to another person as a part of a
blood transfusion. Going forward, patients
who need regular blood transfusions could
AGING, HEALTH, AND BEAUTY
BIOENGINEERING

TECH
development from conception to birth, but
is intended to increase survival and improve
outcomes for extremely premature babies by
replicating a natural womb environment. The
FDA is still working with independent advisers
to determine regulatory and ethical consider-
ations for synthetic wombs and ethical con-
siderations for what human trials could look
like. Other teams around the world are devel-
oping similar devices, while bioethicists are
working out the broader implications. What
if synthetic wombs aren’t available to low-
er-income people? How might they factor into
debates over reproductive rights? And what if,
further in the future, they eliminate the need
for a person to carry a pregnancy at all?
Universal Cancer Vaccines
Early in 2024, the first patient in the UK re-
ceived a dose of a cancer vaccine as part of
a larger clinical trial. Designed to treat sol-
id-state tumor cancers, such as melanoma,
this application of immunotherapy harness-
es the immune system to fight cancer cells.
(“Vaccine” is a bit confusing here, since most
vaccines are designed for prevention, while
analyses to develop custom mRNA vaccines,
which encode protein-containing mutations
unique to the tumor. The immune system
uses those instructions to search and
destroy similar cells throughout the body,
which is similar to how the Covid vaccines
work. BioNTech is running clinical trials for
personalized vaccines for many cancers,
including ovarian cancer, breast cancer, and
melanoma. Moderna is developing similar
cancer vaccines and announced that its per-
sonalized cancer vaccine, when combined
with Merck Co.’s immunotherapy treat-
ment Keytruda, cut recurrence and risk of
death of the most deadly skin cancer com-
pared with immunotherapy treatment alone.
In the trial, the mRNA vaccine revved up the
immune response.
Upgrading Embryos Before Birth
Researchers are developing a new tech-
nique that might someday enable people
to optimize their children’s genes before
birth. Using algorithms to understand the
tiny variations in DNA—single nucleotide
polymorphisms, or SNPs—these researchers
hope to make accurate gene-based predic-
tions about an individual’s future. SNPs are
important markers of genomic variants at a
single base position in the DNA—and these
single letter changes to our genetic code are
contributors to conditions like diabetes. If
SNPs were read in vitro, before embryos were
implanted, they could reveal whether that ge-
netic combination had a higher probability of
developing diabetes or even heart disease. If
an embryo were edited using CRISPR, embryos
could also be optimized with the best possible
traits, given the raw genetic material. Theoret-
ically, parents could influence myriad traits
for their offspring, including hair texture,
resistance to a virus such as HIV, or protection
against Alzheimer’s disease. This intervention,
like the gene drive edit to make mosquitos
unable to transmit malaria, would have a
permanent, heritable effect. It could eradi-
cate certain diseases passed from parents to
children, and in the process improve the entire
gene pool.
this treatment is for people who have already
developed a tumor.) Called mRNA-4359, the
treatment contains a molecule that can relay
instructions to cells. It works by directing
cells to produce proteins typically found on
the surface of solid cancer tumors. Once
these proteins are made, they are introduced
to the immune system, training it to recog-
nize and attack cancer cells.
This vaccine is classified as a “universal”
cancer vaccine, meaning it is premade and
can be administered to patients with certain
types of cancer straight from the shelf. In
contrast, other mRNA cancer vaccines being
developed are customized based on the indi-
vidual patient’s cancer, such as a pancreatic
cancer vaccine that uses genetic material
from the patient’s own tumors for a more
personalized approach.
Long before they were making Covid vac-
cines, both Moderna and BioNTech were
researching immunotherapies for cancer.
After analyzing a tissue sample from a can-
cerous tumor, the companies ran genetic
BIOENGINEERING

TECH
Genetic Screening for Pregnancy
Genetic screening tests promise a kind of
prenatal fortune-telling: Many companies
say they can predict the health outcomes of
growing embryos. California-based MyOme
and New Jersey–based LifeView use the ge-
netic sequences of parents, along with cells
retrieved during a biopsy, to generate an
embryo’s entire genome. Next, they use algo-
rithms to calculate the probabilities of cer-
tain ailments. Couples going through fertility
treatments have a limited number of embryos
to choose from and would be able to select
the embryos they like based on those results.
Other startups offer to calculate scores and
optimize for other genetic traits such as
height and intelligence. LifeView provides
genetic report cards to would-be parents: they
deliver a report showing whether the embryo
has the right number of chromosomes, risk
assessment grades for heart attacks, certain
cancers, diabetes and more.
Genetic screening is a booming market in the
US—and last year, the FDA moved to regulate
prenatal testing. The proposed regulations
cal research projects last year. This extensive
collection of sequences offers an invaluable
tool for exploring the genetic foundations
of human health and various diseases. Labs
worldwide have the opportunity to access
these data sets, with the UK Biobank having
granted approval to over 30,000 researchers
from around 90 countries to date. Scientists
have published more than 9,000 peer-re-
viewed studies using Biobank data, investi-
gating genetic influences on a range of con-
ditions and traits, including Alzheimer’s risk,
heart disease, personality traits, and even
sexual orientation—though that last one has
sparked debate. Whole genome analysis, un-
like exome studies, lets scientists examine
associations between traits and rare genetic
variations in both the protein-coding and
noncoding segments of the genome. While
it’s understood that noncoding regions
play roles in gene regulation among other
functions, much about their contribution
to human biology remains to be discovered.
The expansion of this data set should drive
significant breakthroughs in the near future.
aim to clarify that the FDA has the authority
to regulate the noninvasive prenatal tests
developed and used by individual laborato-
ries. These tests have rapidly evolved from
niche laboratory trials to a major indus-
try; over a third of pregnant women in the
US receive a simple blood test in the first
trimester to check for fetal genetic abnor-
malities. While these tests are highly reliable
for detecting common genetic conditions
such as Down syndrome, the accuracy of
newer tests for rare abnormalities is often
questionable, frequently producing incorrect
positive results. Such inaccuracies can lead
to significant anxiety for expectant parents,
sometimes prompting unnecessary and
costly follow-up procedures. Under the new
FDA oversight, the marketing and availability
of such tests would require government ap-
proval, especially for tests considered “high
risk” because they could influence critical
medical decisions.
Biobank Releases
The UK Biobank made 500,000 genome se-
quences available to scientists for biomedi-
BIOENGINEERING
Soon, it may be possible to upgrade embryos before
birth using new biotechnology techniques.

TECH
Microbiome Metrics at Home
The human microbiome is a mini-universe of
the genetic materials living on and inside our
bodies, inherited from our gestational moth-
ers. It’s vast: there are 200 times more genes
in the bacteria, fungi, protozoa, and viruses
that make up our microbiomes than in the hu-
man genome. That microbiome weighs nearly
5 pounds, and it lives mostly in your gut and
on your skin. Microbiomes differ greatly from
person to person, even if you’re comparing sib-
lings who live in the same city. How well you
digest lactose, how vulnerable you are to skin
cancer, how well you sleep, your probability of
developing anxiety or becoming obese—all of
these traits are linked to the microbiome and
influenced by what you eat and drink, whether
you smoke, what chemicals your body comes
into contact with, and what medications you
take. Data about your microbiome used to be
collected over several visits to an allergist, but
today, at-home tests can determine its genetic
makeup. Some companies will mix together
special probiotic compounds to mitigate con-
ditions or optimize the symbiotic relationship
your body has with all those microorganisms.
Labs is developing biological reprogramming
technology. In 2022, Altos, which raised a
staggering $3 billion in funding over just
one round, announced a partnership with the
Center for iPS Cell Research and Application
at Japan’s Kyoto University to study cellular
rejuvenation programming.
Removing Zombie Cells
Senescent cells are damaged cells that stop
functioning but don’t die, accumulating in
the body like cellular zombies—and they’re
linked to aging. But scientists are research-
ing the use of senolytic drugs, which remove
these worn-out immune cells, as a way to
treat diseases like multiple sclerosis. In
MS, the immune system attacks the myelin
sheath around nerves, and while it’s charac-
terized by phases of relapse and recovery, it
can eventually progress into a phase where
symptoms continuously worsen without pe-
riods of remission. In older animals, myelin
damage leads to lots of senescent cells. But
when researchers at Georgetown University
injected older mice with a toxin to damage
myelin and then treated some with seno-
lytic drugs, the treated mice showed a 65%
greater increase in a myelin-rebuilding protein
compared to untreated mice. This finding
indicates that removing senescent cells could
improve myelin repair, and could mean that
senolytic drugs offer a new treatment strategy
for MS, particularly in its progressive stage—if
it works in humans as well as it does mice,
which for now is a big if. But if human trials
show promising results, it is plausible that
senolytic drugs could be developed to treat
a host of diseases and ailments, along with
conditions associated with aging.
Skin Care and Beauty
Synthetic biology–derived compounds are
producing improved ingredients in skin care
products. Amyris, one of the first commercial
synthetic biology companies, created a suite
of products developed with biosynthesis to
create squalene, a key antioxidant found
in moisturizers. Bay Area startup Geltor is
engineering animal-free collagen for use in
serums and creams, designed to plump skin
and reduce the appearance of fine lines and
wrinkles. Conventional collagen is usual-
Age Reversal
As we age, while our DNA sequence might
stay constant, chemical changes do occur.
Observing those changes could lead to new
techniques to halt or even reverse age-relat-
ed disease. Columbia University research-
ers discovered that it might be possible to
record and store information about cells as
they age. The technique, a sort of biological
DVR, uses the CRISPR-Cas system over a pe-
riod of days. In the future, if we can quantify
aging at a cellular level, we might be able to
reverse it. Synthetic biologist George Church
and a team at Harvard’s Wyss Institute com-
bined three different gene therapies related
to cellular decay into a single compound. The
intent: reverse obesity and diabetes while
also improving kidney and heart function.
Remarkably, the technique seemed to work
(in mice, at least). Maybe that’s why last year
there were so many funding and partnership
announcements in the field. The Saudi royal
family launched the Hevolution Foundation,
a not-for-profit with an annual budget of $1
billion to support basic research on the bi-
ology of aging. Meanwhile, the startup Altos
BIOENGINEERING

TECH
ly drawn from bovine sources, but modern
bioengineering means it can be grown in a
lab. Skin care brand Algenist uses bio-fer-
mented microalgae in its antiaging products,
and it developed alguronic acid that makes
skin look more youthful. One Ocean Beauty,
founded by former Burberry Beauty President
Marcella Cacci, produces a bio-fermented ex-
opolysaccharide isolated from brown kelp, an
elasticity-promoting glycoprotein, and a blue
light-repellant microorganism—all from cells
found naturally in the ocean.
Improving Gut Biomes
A mass extinction event is happening right
now in our guts and in the environment. The
widespread use of antibiotics, along with
diets rich in processed foods, have led to a
staggering decline of microorganisms inside
the people and animals living in wealthy na-
tions. During the past 12,000 years of human
evolution, we’ve shifted nature’s balance—our
diets are now relatively narrow, compared to
our far-distant ancestors. Recently, scientists
studied modern hunter-gatherer tribes in Tan-
zania, Peru, and Venezuela, and found their
Melinda Gates Foundation as one of its
investors. The American Gastroenterological
Association and OpenBiome will track 4,000
patients over 10 years to learn about fecal
microbiomes.
Optimizing Recreational Drugs
When it comes to recreational drugs like
marijuana, genetic factors can determine
whether someone feels pleasantly relaxed or
anxious and listless. New diagnostic tests
promise to optimize recreational drugs for
someone’s unique genetic profile. Atai Life
Sciences NV, based in Berlin, is researching
genetics, depression, and small molecules
within cells in an effort to repurpose psyche-
delics as therapies for depression and PTSD.
MindMed in New York is developing a plat-
form to help patients determine which drugs
to take—therapeutics based on MDMA and
DMT—depending on genetics and other data.
microbiota had 50% more bacterial species
than those in the West today. Unlike those
tribes, we no longer hunt and eat wild flora
and fauna. Those from wealthier countries
now eat very little dietary fiber, a limited
variety of fruits and vegetables, and only four
species of livestock: sheep, poultry, cattle,
and pigs. Worse, widespread use of antibiot-
ics in farm animals—used not necessarily to
prevent disease but to increase weight gain
and therefore the volume of meat available—
means that we’re ingesting compounds that
are helping to destroy our own microbiomes.
Humans are complex, composite organ-
isms, made up of layers and layers of cells.
Researchers now think that our gut micro-
biome is directly linked to our metabolism,
our immune systems, our central nervous
systems, and even the cognitive functions
inside our brains. It’s an inherited problem:
Most of our microbiomes come to us from
our mothers as we pass through the birth
canal. A number of researchers are now look-
ing at the future of our microbiomes. Vedan-
ta Biosciences is making gut bacteria that
can be turned into drugs and counts the Bill
BIOENGINEERING
Scientists are learning more every day about the
relationship between our gut microbes and our
bodies.

CLIMATE AND
SUSTAINABILITY
TECH BIOENGINEERING

TECH
eDNA Detection
Environmental DNA, or eDNA, is genetic ma-
terial found in the environment. Feces and fur
from animals, as well as hair and saliva from
humans are just some of the organic matter
found in soil, seawater, snow, and air. As a
fish moves through water, it’s continuously
shedding bits of itself. Likewise, when a cyclist
rides on a trail, her sweat, mucus, and dead
skin cells wind up mixed into the gravel and
dirt. These fragments of nuclear or mitochon-
drial DNA can reveal invaluable insights about
an environment. Scientists from the United
States Geological Survey and the Monterey Bay
Aquarium Research Institute are developing
a new mobile eDNA sampler that can float
through rivers and streams, collecting materi-
al and detecting pathogens or invasive species
autonomously. As detection systems advance,
eDNA detection will serve as early warning
systems for potential outbreaks. But there’s
another interesting use for eDNA: reconstruct-
ing ancient ecosystems. Scientists excavated
eDNA from frozen soil in the Arctic desert, and
were able to piece together a lost world nearly
2 million years old. The eDNA revealed a coast-
ments could make forestry more sustainable
and efficient, offering new economic and
environmental benefits.
Better Plastics Recycling
Despite global efforts to recycle plastic prod-
ucts, there are numerous barriers: Consum-
er-facing plastics come in different varieties,
they’re often coated with labels or print, and
they have colors and other added features.
The mess of waste—used iPhone cases, emp-
ty shampoo containers, soda bottles—can’t
be easily managed at scale, so a lot of it piles
up. A potential solution is microorganisms
like some bacteria and fungi that use spe-
cial enzymes to break down various types of
plastics. But turning plastic into something
these microbes can eat isn’t as simple as
just mixing them together. The plastics need
to be pre-damaged by sunlight or chemicals,
and the microbes need just the right condi-
tions to do their work. Even so, each type of
microbe can only eat certain plastics, and
it can take them weeks or months to break
down just a small amount.
Now, an emerging synthetic biology process
offers a new solution. France-based Carbios
developed a process using an enzyme that’s
especially good at breaking down PET plastic
into its basic building blocks, making it possi-
ble to recycle PET into high-quality new plastic.
After improving the enzyme and testing it in
an industrial setting, Carbios is now building
its first site dedicated to this bio-recycling
process. Meanwhile, researchers at SLAC Na-
tional Accelerator Laboratory and the National
Renewable Energy Laboratory used a microp-
orous material called a zeolite that contains
cobalt nanoparticles as a catalyst to break
down different polymer molecules, turning
the majority into propane. At the University of
Texas at Austin, researchers used a machine
learning model to generate novel mutations to
natural enzymes that allow bacteria to break
down the plastics found in soda bottles and
most consumer packaging. The enzyme, called
FAST-PETase (functional, active, stable, and
tolerant PETase), could operate efficiently and
work at an industrial scale. The first real-world
application: setting the enzyme loose to clean
up landfills.
al forest with conifers, black geese, horse-
shoe crabs, lemmings, and mastodons—a
natural wonderland unlike any in existence
today.
Using CRISPR for Sustainable Wood
Production
Trees are a valuable natural resource, but
improving their wood through traditional
breeding is slow and complicated due to their
complex genetics. CRISPR technology offers
a way to quickly change trees to improve
their wood—optimizing it for paper-making or
reduced carbon emissions, for example. Sci-
entists have used CRISPR to change multiple
genes in poplar trees, resulting in wood that’s
easier to process and better for the environ-
ment. This breakthrough means we can now
grow trees that are more suited for our needs
while also helping the planet. The key chal-
lenge has been dealing with lignin, a natural
part of wood that’s tough to break down. But
by precisely editing genes related to lignin,
researchers at North Carolina State University
have created poplar trees with wood that’s
much easier to turn into fiber. These advance-
CLIMATE AND SUSTAINABILITY
BIOENGINEERING

TECH
Engineering Plants for Carbon Capture
Carbon dioxide is the undisputed culprit
when it comes to climate change. But what if
we could just suck it out of the air? Trees do
that naturally, but with deforestation, there’s
not enough to make a sizable impact. The
Salk Institute’s Harnessing Plants Initiative is
working on an innovative approach that relies
on our existing carbon storage mechanisms
to help solve climate change. It’s developing
engineered crops that can store more carbon
in the ground for long periods of time. The
crops have a larger root mass, are deeper,
and contain more suberin, a plant tissue that
already relies on CO2 and can store signifi-
cant amounts without causing harm to the
plant. Salk researchers are hoping to develop
strains of rice, wheat, corn, and other plants
that both produce edible crops and store car-
bon, for improved soil health. Meanwhile, an
artificial leaf developed at Harvard harnesses
solar energy. When connected to a strain of
bacteria, it converts atmospheric CO2 and
nitrogen into organic forms that can benefit
living organisms. Those hungry, solar-fed bac-
teria essentially overeat, to the point where
veloped silks that are stronger, tougher, and
lighter than previous versions, with an eight-
fold increase in yield. This innovation could
provide an eco-friendly alternative to tradi-
tional textiles, drastically reducing waste
in the fashion industry. The engineered
silk, combining mussel foot proteins with
spider silk properties, has already reached
production levels sufficient for real-world
product testing, marking a significant step
toward its commercial use. Already, several
other successful trials of synthesized tex-
tiles have occurred: Bolt Threads developed
a synthetic fabric called Microsilk that’s
engineered from spider DNA, and Japanese
startup Spiber synthesized enough fibers to
manufacture a limited-edition parka. While
manufacturing new textiles is on the hori-
zon, getting bio-sourced materials such as
PHAs, spider silk, and chitosan into the sup-
ply chain process remains a challenge. Mills
and manufacturers don’t have incentives to
risk using new materials that may not work
seamlessly with their existing production
equipment.
30% of their body weight is excess energy—
stored CO2 and nitrogen. These microbes
then get mixed into soil, and release all that
nitrogen near the roots of plants, acting as
an organic fertilizer. At that point, they also
release the CO2, yet it remains trapped un-
derground. The result: enormous crop yields
without the environmentally poisonous side
effects typically associated with chemical
fertilizers.
Greening Fashion
The textile and clothing industry is a noto-
rious polluter but is making steps toward
more sustainable practices. Fuzhong Zhang,
a professor at Washington University in
St. Louis, has advanced the production of
synthetic spider silk, which could lead to
sustainable clothing manufacturing. Using
synthetic biology, his team created a process
to yield more silk from microbes, aiming
to meet the fashion industry’s demand for
renewable materials and reduce the envi-
ronmental impact associated with clothing
production. By incorporating a protein from
a common shellfish—mussels—the team de-
BIOENGINEERING
The Salk Institute’s Harnessing Plants Initiative is
working on an innovative approach that relies on
our existing carbon storage mechanisms to help
solve climate change.
Image credit: Salk Institute.

TECH
De-Extincting Lost Species
Woolly mammoths were once a “keystone
species,” one that other species in the eco-
system depended on in many ways for stabil-
ity. They stomped around in herds, knocking
down trees and packing down snow layers
as they searched for dead grasses to eat, and
that helped keep the permafrost layer stable.
Once the mammoths and other large grazing
animals stopped compacting the snow and
eating dead grasses, the ecosystem began to
change: The snow melted more easily, which
allowed the sun to reach the permafrost. The
permafrost layer is now melting at an alarm-
ing rate and releasing greenhouse gasses into
the atmosphere, which creates a vicious cycle:
Hotter temperatures lead to more melting,
which releases more gasses, which causes
hotter temperatures, and on and on it goes. Re-
searchers are helping to de-extinct the woolly
mammoth and other species using synthetic
biology techniques: Starting with a fully intact
healthy cell from a closely related species and
working backward, with genetic fragments
from preserved specimens, they could develop
a version of the animals that once existed.
bison were released in a woodland near Can-
terbury—the hope is that over time, the her-
bivores will revitalize a stretch of southeast
England and allow vegetation to grow again,
which should in turn boost biodiversity.
CRISPR Mosquitoes
Gene-drive technology, which has the po-
tential to spread antimalarial genes among
mosquito populations, is showing prom-
ise. Malaria kills hundreds of thousands of
people every year; 96% of the deaths are in
African countries, and particularly impact
children under age 5. The idea of gene drive,
which uses special DNA pieces that copy
themselves and spread quickly through an
organism’s genes, was thought of years ago.
However, it had problems because these
DNA pieces could end up in important parts
of the genes and cause harm. CRISPR-Cas9
technology, which allows for precise genetic
edits, enables the safe transfer of antimalar-
ial genes from one generation of mosquitos
to the next. This advancement offers hope
for controlling malaria by genetically modi-
fying mosquitoes to resist or eliminate the
disease. There have now been several pilots
around the world to edit mosquitoes so they
no longer carry malaria. In 2021, biotech com-
pany Oxitec launched a controversial field test
of specially engineered mosquitoes in Flori-
da in a move toward reducing the spread of
deadly diseases such as dengue, yellow fever,
and the Zika virus. Its tiny capsules contain
an engineered form of the Aedes aegypti
mosquito, called OX5034. Because only female
Aedes aegypti bite and spread disease, Oxitec
engineered males to pass on a gene that kills
female offspring before they mature. Male
offspring then continue mating and passing
on the altered gene, which should change the
population of disease-carrying mosquitoes.
The US Environmental Protection Agency
said this pilot poses no human threat, while
local authorities, who have been dealing with
steadily growing cases of dengue fever and
West Nile virus, hope that a smaller mosqui-
to population will curb the diseases without
insecticides or poisonous chemicals. The EPA
later approved an expanded plan to release
2.4 million genetically engineered mosquitoes
in more US sites, including California. Similar
Rewilding Barren Terrains
Rewilding is a direct human intervention
into nature using technology and science,
a holistic approach to conservation that
focuses on restoring the natural phenom-
ena of wilderness ecosystems, providing
connective corridors between wild spaces,
and reintroducing keystone species to their
natural habitats. A term coined more than 30
years ago, “rewilding” has gained renewed
attention in the past few years as the cli-
mate crisis has grown more dire and new
technologies have promised to protect and
rehabilitate ecosystems. In 2017, research-
ers plunged into the waters off Lizard Island
on the northeastern coast of Australia with
some unexpected equipment in tow—a set
of underwater loudspeakers. Their destina-
tion was a coral reef that had been all but
abandoned by a once-thriving population
of sea life. The researchers hoped that by
broadcasting the telltale sounds of a healthy
reef, they might lure back some of its vital
inhabitants. Remarkably, it worked. This
experiment was a unique instance of rewild-
ing, but there have been others. In 2022, four
BIOENGINEERING

TECH
experiments are underway in Malaysia and
Panama. While they can transmit malaria, re-
searchers are also thinking about how to use
engineered mosquitoes to deliver a defense
against deadly viruses. What if in the future
mosquitoes are flying syringes capable of de-
livering vaccines? University of Washington
scientists are working on a weakened form
of malaria-causing Plasmodium parasites
that won’t get people sick but will cause the
body to create antibodies. On the other hand,
could meddling in the genetic code of in-
sects, reptiles, and animals have catastrophic
consequences that no one anticipates? If that
sounds familiar, it’s because you’ve seen that
movie before: “Jurassic Park.”
Reviving Ancient Viruses
As the Arctic warms, the thawing perma-
frost is raising concerns about the release of
ancient viruses that could pose health risks
to animals and humans. Scientists highlight
the potential danger of these long-dormant
viruses, along with the release of chemical
and radioactive waste from the Cold War
era, emphasizing the importance of keeping
permafrost frozen. The permafrost acts as
a natural time capsule, preserving not only
viruses but also extinct animals, which
scientists have been able to study. The rapid
warming of the Arctic threatens to disrupt
this frozen archive, with temperatures there
rising up to four times faster than the global
average. Research into “zombie viruses”
found in Siberian permafrost has shown that
some of these viruses are still infectious,
which could mean a potential future risk of
outbreaks from ancient pathogens. Global
warming and increased activity in the region
could heighten the possibility of a spillover
event. Scientists advocate for proactive
surveillance and research to understand the
risks posed by thawing permafrost and to
mitigate the impact of climate change on
the release of pathogens.
BIOENGINEERING
As the permafrost layer melts, ancient viruses could thaw and release pathogens harmful to the
modern world.
Image credit: Image credit: Future Today Institute and Dall-E.
As the permafrost layer melts, ancient viruses could thaw and release pathogens harmful to the
modern world.
Image credit: Image credit: Future Today Institute and Dall-E.

REGULATION
AND POLICY
BIOENGINEERING
TECH

TECH
Regulatory Changes Toward Genome-Edited
Crops
Starting last year, countries have been reeval-
uating their regulatory stance on genome-ed-
ited crops, marking a significant change in
agricultural biotechnology policies. China’s
recent safety assessment guidelines for
these crops show a partial shift from tradi-
tional regulations on genetically modified
organisms, signaling a move toward a more
nuanced approach that seeks to balance
scientific innovation with public trust. Last
July, the European Commission proposed
legislative changes to relax the rules on gene
editing, suggesting a departure from stringent
GMO laws for certain next-generation tech-
niques. This proposed legislation argues that
edits achievable through traditional breeding
methods should not be subject to the same
stringent regulations as GMOs, streamlining
the approval process for these innovations.
Still, more complex modifications involving
foreign DNA would still fall under existing GMO
regulations, maintaining a layer of oversight
on more radical genetic alterations. Howev-
er, there is still no global framework, and as
produced a 3D image of a potential suspect
based on the DNA evidence—not a real photo-
graph but a predicted appearance including
skin tone, eye and hair color, and other facial
features. The company even incorporated a
haircut and mustache based on a witness’
description, not the DNA. In a move to gather
public tips, the police department released
this generated face, leading (unsurprisingly)
to controversy. In 2020, more controversy
ensued when a detective requested to run
the DNA-based facial reconstruction through
facial recognition software, a move consid-
ered problematic by civil liberties experts and
against Parabon NanoLabs’ policies. But while
this was one of the first known instances of
a police department using this tactic, the
question of how and when to use someone’s
DNA is increasingly common. The COVID-19
pandemic accelerated widespread use of
infectious disease surveillance techniques,
from saliva tests at airports and border cross-
ings, to nasal swabbing at testing centers. To
ease testing bottlenecks, which sometimes
resulted in hours-long lines, alternative
testing centers opened up: Private companies
dispatched workers, who often had no medical
training, to vans or small tents to administer
PCR or rapid response tests. It wasn’t immedi-
ately clear where the test results would be sent,
or who might also gain access to the data. With
the growing size and scale of third-party test
results and genetic databases, anyone with the
right skills could identify individuals—and we
don’t yet have safeguards against widespread
genetic surveillance. As of the start of 2024,
there are few restrictions on private companies
buying and selling genetic data in the US and
in many places around the world.
Safeguarding Genetic Privacy
Genetic privacy will be increasingly difficult
to safeguard—yet big genetic data sets are
required to perform the kind of research that
leads to new therapeutics. Sharing a person’s
complete genetic code online can help scien-
tists but also poses privacy risks, as people
with bad intentions might use it to learn about
the health of an individual or their family.
Recently, experts have started using AI to
generate artificial, but scientifically useful, ge-
netic data that keeps people’s identities safe.
we’ve seen already, biology has a tendency to
replicate and evolve in unpredictable ways.
The developments in China and the EU reflect
an urgent global need to update and reform
regulatory frameworks for genome-edited
crops, driven by a desire to foster innovation
while ensuring safety and maintaining public
confidence. As countries continue to adapt
their policies, further changes will shape the
future of agricultural biotechnology, poten-
tially making genome-edited crops more
accessible and widely used across the globe.
Regulating DNA Recognition Systems
Can you produce a photo of someone’s face
solely from their DNA? In the US, police de-
tectives used a DNA sample to generate a 3D
model of a suspect’s face using facial recog-
nition tools. Detectives from California’s East
Bay Regional Park District Police Department
decided to use genetic information from a
1990 cold case involving the death of Maria
Jane Weidhofer to create a lead. They sent the
DNA from the crime scene to Parabon Nano-
Labs, which specializes in generating faces
from DNA. Using machine learning, Parabon
REGULATION AND POLICY
BIOENGINEERING

TECH
Scientists at the University of Tartu, Estonia,
use neural networks to develop novel seg-
ments on human genomes. Because genetic
data is sensitive, the hope is that an artificial
human genome will allow researchers to study
DNA without infringing on anyone’s privacy.
Researchers at the University of Montpellier
in France developed a novel method that uses
both AI and known information about how
genes change in our bodies to work with big
data sets more easily. They cut genetic data
from thousands of people from different back-
grounds into pieces, based on where genes
often mix during reproduction, and used it to
train the AI system. It now creates artificial
populations of genetic data that are diverse
and realistic but don’t risk anyone’s privacy.
Defining Parenthood
The relationship between creating a child and
being the parent of that child is becoming
more complex due to advances in reproductive
technology and changes in the law. Emerg-
ing technologies like induced pluripotent
stem cells (see: Growing Sex Cells trend) will
someday allow children to be conceived with
National DNA Drives
Several countries are in the process of de-
veloping their own national DNA databases,
notably for the purpose of medical research
advancement. The United Arab Emirates is
currently working to sequence its entire pop-
ulation; the goal is to aid scientific research,
as well as to map and sequence the genes of
UAE nationals, which will assist in prevent-
ing and treating chronic illnesses. Because
existing databases are overwhelmingly made
up of Caucasian Americans and Europeans,
people of Arab descent have been exclud-
ed from the benefits of genetic research.
While the UAE’s program is voluntary, there
is a different strategy being used in China.
Over the last decade, China has launched a
comprehensive national effort to collect, se-
quence, and store the genetic information of
its citizens, integrating DNA databases into
a broader surveillance system fueled by the
government’s ambitions in artificial intelli-
gence. This initiative has particularly targeted
the Uyghur population, under the guise of
public health programs like “Physicals for
All,” leading to the collection of genetic data
without clear consent and raising concerns
over privacy and human rights violations. As
China builds a vast and unparalleled genetic
database, encompassing both minority groups
and the majority Han Chinese population, it
faces minimal domestic opposition to its ge-
netic research and surveillance practices, con-
trasting with the ongoing debates over genetic
privacy in the US, Canada, the EU, and the UK.
International Collaborations to Advance
Bioengineering
Researchers are building the first-ever com-
prehensive map of all 37.2 trillion human cells
in the body. The effort includes 130 software
engineers, mathematicians, computational
scientists, biologists, clinicians, and physicists
hailing from Israel, the Netherlands, Japan, the
UK, the US, and Sweden. Although a cell atlas
has long been theorized, new biological tools
and more powerful computers have turned this
one-time vision into a reality. These scientists
believe this mapping will give the medical
community a new way of understanding how
our bodies work and will help diagnose, moni-
tor, and treat disease.
synthetic eggs and sperm or grown in arti-
ficial wombs, challenging society’s current
ideas about procreation and parenthood. In
the UK, regulators are considering reforms
to make surrogacy simpler and to address
the issue of commercial surrogacy, which is
technically illegal but practiced in a gray area.
One controversial proposal is to allow surro-
gacy without a genetic link between the child
and the commissioning parents, which raises
questions about whether this constitutes as-
sisted reproduction or a form of adoption. The
reforms also consider granting commission-
ing parents full parental rights from birth,
moving away from the surrogate’s default
parental rights. These discussions highlight
the tension between the desire for genetic
parenthood and reproductive freedom, espe-
cially in cases where a biological connection
to the child does not exist. Such debates will
become increasingly relevant with the intro-
duction of technologies like ectogenesis, po-
tentially redefining what it means to “make”
a baby and become a parent.
REGULATION AND POLICY
BIOENGINEERING

ETHICS,
TRUST AND
ACCEPTANCE
TECH BIOENGINEERING

TECH
Resolving Bias in Genome Research
Overwhelmingly, the majority of people who
have had their genomes sequenced come from
affluent Caucasian Americans and Europeans;
fewer than 2% are from Africa. This excludes
an enormous number of people from the
benefits of genetic research, so there is now
increased attention and funding to diversify
this pool. H3Africa works with African investi-
gators to determine genomic and environmen-
tal determinants of common diseases. The
Non-Communicable Diseases Genetic Heritage
Study consortium, based in Nigeria, is creating
a comprehensive catalog of human genetic
variation among Nigerians. A decade-long
Three Million African Genomes project is also
underway to locate missing genetic variants
from ancestral genomes in Africa. It would
build an African biobank of clinical informa-
tion and could lead to a more equitable future
of genetic research.
Ethics in Indigenous Genomics
Biological materials from Indigenous peoples
are still missing from genetic databases,
basic research, and clinical studies. One major
to another problem: The pool of genetic data
in the US doesn’t include Indigenous peoples.
New initiatives could make genetic datasets
and research more inclusive. The Summer In-
ternship for Indigenous Peoples in Genomics
trains budding scientists, while the Center for
the Ethics of Indigenous Genomic Research
works to promote Indigenous-led research in
biobanking and precision medicine. Canada
and New Zealand are both working on gov-
erning frameworks and libraries to include
Indigenous peoples, relying on direction from
local communities.
Posthumous Sperm Retrieval
In the wake of the Hamas terrorist attack on
October 7, 2023, when hundreds of young
men were among the Israelis who lost their
lives, there was an unprecedented surge in
requests for posthumous sperm retrieval
(PSR) by embryologists and IVF special-
ists. This process involves retrieving viable
sperm from the deceased’s testicular tis-
sue shortly after death and preserving it in
liquid nitrogen. Families sought to preserve
their loved one’s genetic legacy by extract-
ing and freezing their sperm, hoping for the
possibility of conceiving a child in the future.
As a previously rare procedure, PSR required a
family court order in Israel for unmarried men,
while the spouse of a married individual could
directly request the procedure. But the crisis
led to Israel’s Ministry of Health easing regula-
tions, allowing hospitals to process requests
from parents without court involvement. It also
brought PSR into mainstream view, raising
questions about whether it is ethical to retrieve
and use sperm after death.
Gene Editing Ethics
In 2018, Chinese scientist He Jiankui caused
a global uproar by announcing he had created
the world’s first gene-edited children using
CRISPR technology, targeting embryos to make
them resistant to HIV. This led to the birth of
twins, marking a controversial milestone in
genetic editing. Jiankui’s actions, deemed
“illegal medical practices” in China, resulted
in a three-year prison sentence for him and
his two associates, partly because the genetic
alterations could be passed down to future
generations. Following the scandal, China
reason why can be traced back to how mem-
bers of the Havasupai tribe were treated by
Arizona State University (ASU) in the late 20th
century. In 1990, Havasupai were grappling
with an increase in diabetes. They allowed
ASU researchers to collect blood samples,
hoping the research would help them eradi-
cate the disease. But then, unbeknownst to
the Havasupai, the researchers changed the
scope of the project to encompass genetic
markers for alcoholism and various mental
disorders. They went on to publish many
papers in academic journals highlighting
their results, which led to news stories about
inbreeding and schizophrenia among tribe
members. The Havasupai were, understand-
ably, horrified and humiliated, and they filed
a lawsuit against ASU in 2004. ASU eventually
settled the suit in 2010, returned the blood
samples to the tribe, and promised not to
publish any more research. Subsequently,
the Navajo Nation, the second-largest group
of Indigenous peoples in the US, banned all
genetic sequencing, analysis, and related re-
search on its members. Although their objec-
tions were absolutely warranted, they’ve led
ETHICS, TRUST AND ACCEPTANCE
BIOENGINEERING

ETHICS, TRUST AND ACCEPTANCE
TECH
tightened regulations on human gene editing
and banned He from conducting any repro-
ductive technology services. Despite these
restrictions, late in 2023, He proposed a new
study focused on editing mouse and human
embryos to investigate potential protection
against Alzheimer’s disease, citing the ur-
gent need to address the challenges posed by
an aging population and the current lack of
effective treatments for Alzheimer’s. His latest
proposal has reignited ethical debates and
concerns within the scientific community. The
proposal’s reception remains mixed, reflect-
ing ongoing dilemmas over the boundaries of
genetic research.
A dozen countries have now banned germ line
engineering in humans, though their ranks
do not include China, which tightened regula-
tions without banning the practice outright.
Federal law in the US regulates the use of
federal funds for research on human germline
gene therapy—laws are notoriously politicized
and have changed a few times in the past
decade. The EU’s Convention on Human Rights
and Biomedicine said tampering with the gene
immune responses. The team plans to further
their research by transforming these mod-
ified cells into blood-making cells to help
humans survive acute radiation sickness,
suggesting additional benefits in protecting
against diseases such as cancer and dia-
betes. The experiment was deemed legal as
it was conducted on cultured cell lines in a
lab. But what happens when that research is
ready to leave the lab for the real world?
pool would be a crime against human dignity
and human rights. But all those declarations
were made before it was actually possible to
precisely engineer the germ line. Now, with
CRISPR, it is possible.
Engineering Super Soldiers
Last year, a team of military medical sci-
entists in China reported that they had
enhanced human embryonic stem cells’
resistance to radiation by inserting a gene
from the water bear, a microorganism known
for its extreme survivability. Using CRISPR
technology, they achieved a high survival rate
of these modified cells under lethal radiation
exposure. The research, led by professor Yue
Wen at the Academy of Military Sciences in
Beijing, has sparked interest (read: alarm)
since its publication, because of the impli-
cation: What if this is used to create a new
version of superhumans, capable of surviv-
ing extreme conditions like nuclear fallout?
Scientists around the world raised concerns
about the safety and ethical implications
of transferring genes across species, with
the risk of harmful mutations or unknown
BIOENGINEERING
Some researchers are concerned that biological ex-
perimentation could someday produce supersoldiers
who have been enhanced with special capabilities.
Image credit: Image credit: Future Today Institute
and Dall-E.

AUTHORS
CONTRIBUTORS
BIOENGINEERING
TECH

AMY WEBB
Recognized as the global leader in strategic fore-
sight, Amy Webb advises business leaders through
disruptive change, enabling them to navigate an
unpredictable future with confidence and take
actions that address global challenges, create
sustainable value, and ensure a company’s long-term growth. As founder and CEO of
the Future Today Institute, Amy pioneered a unique quantitative modeling approach
and data-driven foresight methodology that identifies signals of change and emerg-
ing patterns very early. Using that information, Amy and her colleagues identify white
spaces, opportunities, and threats early enough for action. They develop predictive
scenarios, along with executable strategy, for their global client base. In 2023, Amy
was recognized as the #4 most influential management thinker in the world by
Thinkers50, a biannual ranking of global business thinkers. She was also featured on
the 2021 Thinkers 50 list, was shortlisted for the 2021 Digital Thinking Award, and re-
ceived the 2017 Thinkers50 Radar Award. Forbes called Amy “one of the five women
changing the world,” and she was honored as one of the BBC’s 100 Women of 2020.
Amy also serves as a professor of strategic foresight at New York University’s Stern
School of Business, where she developed and teaches the MBA-level strategic fore-
sight course with live case studies. She is a Visiting Fellow at Oxford University’s Säid
School of Business. She was elected a life member of the Council on Foreign Relations
and is a member of the Bretton Woods Committee. She is a Steward and Steering
Committee Member for the World Economic Forum, a founding member of the Forum’s
Strategic Foresight Council, a member of the Forum’s Risk Advisory Council, and serves
on the Forum’s Global Futures Council. She was a Delegate on the former U.S.-Russia
Bilateral Presidential Commission, representing US interests in technology.
Regarded as one of the most important voices on the futures of technology (with spe-
cializations in both AI and synthetic biology), Amy is the author of four books, including
the international bestseller The Big Nine and her most recent, The Genesis Machine,
which was listed as one of the best nonfiction books of 2022 by The New Yorker. To
date, her books have been translated into 19 languages. A widely published and quoted
thought leader, Amy regularly appears in a wide range of publications and broadcasts.
Managing Director
MELANIE SUBIN
Creative Director
EMILY CAUFIELD
Editor
ERICA PETERSON
Copy Editor
SARAH JOHNSON
CHERYL COONEY
BIOENGINEERING
TECH

SELECTED
SOURCES
TECH BIOENGINEERING

Attal-Juncqua, Aurelia, Galen
Dods, Nicole Crain, et al. “Shap-
ing the Future US Bioeconomy
Through Safety, Security, Sus-
tainability, and Social Responsi-
bility.” Trends in Biotechnology,
org/10.1016/j.tibtech.2023.11.015.
Balaguer, Angels, Vinamra
Benara, Renato Luiz de Freitas
Cunha, et al. “RAG vs Fine-Tun-
ing: Pipelines, Tradeoffs, and a
Case Study on Agriculture.” ArX-
iv, January 17, 2024. https://doi.
org/10.48550/arXiv.2401.08406.
Baldock, Sara J., Punarja Kevin,
Garry R. Harper, et al. “Creating
3D Objects with Integrated
Electronics via Multiphoton Fab-
rication In Vitro and In Vivo.” Ad-
vanced Materials Technologies
8, no. 11 (2023): 2201274. https://
doi.org/10.1002/admt.202201274.
Ball, Philip. “Synthetic Morphol-
ogy Lets Scientists Create New
Life-Forms.” Scientific American,
May 1, 2023. www.scientificamer-
ican.com/article/synthetic-mor-
phology-lets-scientists-cre-
ate-new-life-forms/.
Ballard, Shawn, and Washington
University in St Louis. “Syn-
thetic Biology Meets Fashion
in Engineered Silk.” Phys.org,
April 20, 2023. https://phys.org/
news/2023-04-synthetic-biolo-
gy-fashion-silk.html.
“Biomemory—DNA Digital Data
Storage.” www.biomemory.com/.
“Bi-Terminal Fusion of Intrin-
sically-Disordered Mussel Foot
Protein Fragments Boosts Me-
chanical Strength for Protein Fi-
bers.” Nature Communications,
April 14, 2023. www.nature.com/
articles/s41467-023-37563-0.
Boiko, Daniil A., Robert
MacKnight, and Gabe Gomes.
“Emergent Autonomous Sci-
entific Research Capabilities
of Large Language Models.”
ArXiv, April 11, 2023. https://doi.
org/10.48550/arXiv.2304.05332.
Boyle, Alan. “Seattle Hub for
Synthetic Biology Plans to Trans-
form Cells into Tiny Recording
Devices.” GeekWire, December 7,
2023. www.geekwire.com/2023/
seattle-hub-for-synthetic-biolo-
gy-recording-cells/.
Bran, Andres M., Sam Cox, Oliver
Schilter, et al. “ChemCrow: Aug-
menting Large-Language Models
with Chemistry Tools.” arXiv,
October 2, 2023. https://doi.
org/10.48550/arXiv.2304.05376.
Brown University. “Eric Sorge:
Leveraging Synthetic Biology to
Clean Arsenic-Poisoned Water.”
January 18, 2024. www.brown.
edu/news/2023-08-24/sorge.
Burnard, Callum, Alban Man-
cheron, and William Ritchie.
“Generating Realistic Artificial
Human Genomes Using Adver-
sarial Autoencoders.” BioRxiv,
org/10.1101/2023.12.08.570767.
Chulikavit, Nattanan, Cheng
Wang, Tien Huynh, et al. “Fire-
proofing Flammable Composites
Using Mycelium: Investigating
the Effect of Deacetylation on
the Thermal Stability and Fire
Reaction Properties of Myce-
lium.” Polymer Degradation
and Stability 215 (September
1, 2023): 110419. https://doi.
org/10.1016/j.polymdegrad-
stab.2023.110419.
Commissioner, Office of the.
“FDA Approves First Gene Thera-
pies to Treat Patients with Sickle
Cell Disease.” FDA, December
8, 2023. www.fda.gov/news-
events/press-announcements/
fda-approves-first-gene-thera-
pies-treat-patients-sickle-cell-
disease.
Crispo, M., A. P. Mulet, L. Tesson,
et al. “Efficient Generation of
Myostatin Knock-Out Sheep
Using CRISPR/Cas9 Technol-
ogy and Microinjection into
Zygotes.” PLOS ONE 10, no. 8
(August 25, 2015): e0136690.
https://doi.org/10.1371/journal.
pone.0136690.
Dalal, Vasundhra, Nagarjuna Pa-
supuleti, Gyaneshwer Chaubey,
et al. “Advancements and Chal-
lenges in Ancient DNA Research:
Bridging the Global North–South
Divide.” Genes 14, no. 2 (February
2023). https://doi.org/10.3390/
genes14020479.
“Data Science Artificial Intel-
ligence: Unlocking New Science
Insights.” AstraZeneca. Accessed
January 31, 2024. www.astrazen-
eca.com/r-d/data-science-and-
ai.html.
DeAngelis, Allison. “VC Firm
Starts a New Synthetic Biology
Incubator, Headed by a Former
Broad Institute Scientist.” STAT
(blog), July 27, 2023. https://
www.statnews.com/2023/07/27/
ra-capital-synthetic-biology-in-
cubator/.
Dixon, Thomas A., Paul S.
Freemont, Richard A. Johnson,
and Isak S. Pretorius. “A Global
Forum on Synthetic Biology: The
Need for International Engage-
ment.” Nature Communications
13, no. 1 (June 18, 2022): 3516.
022-31265-9.
Dusséaux, Simon, Victor For-
man, and Sotirios C. Kampranis.
“Precision Fermentation to
Make Non-Alcoholic Beer Taste
Great.” Trends in Biotechnology,
org/10.1016/j.tibtech.2023.10.016.
Gim, Gyeong-Min, Dong-Hyeok
Kwon, Kyeong-Hyun Eom, et al.
“Production of MSTN-Mutated
Cattle without Exogenous Gene
Integration Using CRISPR-Cas9.”
Biotechnology Journal 17, no.
7 (2022): 2100198. https://doi.
org/10.1002/biot.202100198.
SELECTED SOURCES
TECH BIOENGINEERING

Guo, Rihong, Yongjie Wan,
Dan Xu, et al. “Generation and
Evaluation of Myostatin Knock-
out Rabbits and Goats Using
CRISPR/Cas9 System.” Scientific
Reports 6, no. 1 (July 15, 2016):
srep29855.
Hahn, Jaeseung, Suwan Ding,
Jongwon Im, et al. “Bacterial
Therapies at the Interface of
Synthetic Biology and Nanomed-
icine.” Nature Reviews Bioengi-
neering, October 10, 2023: 1–16.
023-00119-4.
Han, Ping, Wei Zhe Teo, and Wen
Shan Yew. “Biologically Engi-
neered Microbes for Bioreme-
diation of Electronic Waste:
Wayposts, Challenges and Fu-
ture Directions.” Engineering Bi-
ology 6, no. 1 (February 26, 2022):
23–34. https://doi.org/10.1049/
enb2.12020.
Honu, Mandana Manzari, Bog-
dan Knezevic, and David Yen.
“Make Better, Data-Backed Deci-
sions to Hit Biotech Milestones
Faster.” Trends in Biotechnology,
January 3, 2024. https://doi.
org/10.1016/j.tibtech.2023.11.011.
“How Artificial Intelligence Is
Revolutionizing Drug Discov-
ery.” Bill of Health, Petrie-Flom
Law Center at Harvard Law
School, March 20, 2023. https://
blog.petrieflom.law.harvard.
edu/2023/03/20/how-artifi-
cial-intelligence-is-revolutioniz-
ing-drug-discovery/.
“Human Fetal Brain Self-Or-
ganizes into Long-Term Ex-
panding Organoids”. Cell 187,
issue 3 (February 1, 2024). www.
cell.com/cell/fulltext/S0092-
8674(23)01344-2.
Imperial News. “First UK Patients
Receive Experimental mRNA
Therapy for Cancer.” Imperi-
al College London, February
4, 2024. www.imperial.ac.uk/
news/251213/first-uk-pa-
tients-receive-experimental-mr-
na/.
Intellia Therapeutics. “MAGNI-
TUDE: A Phase 3, Multinational,
Multicenter, Randomized, Dou-
ble-Blind, Placebo-Controlled
Study to Evaluate the Efficacy
and Safety of NTLA-2001 in
Participants With Transthyretin
Amyloidosis With Cardiomyop-
athy (ATTR-CM).” Clinical trial
registration. January 19, 2024.
https://clinicaltrials.gov/study/
NCT06128629.
Jones, Sam. “How Genetically
Modified Mosquitoes Could
Eradicate Malaria.” Nature 618,
no. 7967 (June 28, 2023): S29–31.
https://doi.org/10.1038/d41586-
023-02051-4.
Kale, Sampreeti Sri Sai, Rohith
Kode, Sai Kiran Kuchana, and L.
V. Simhachalam Kutikuppala.
“Breakthrough in the Scientific
World: Lab-Grown Red Blood
Cells Used in Transfusions.”
Asian Journal of Transfusion
Science 17, no. 1 (2023): 143–44.
https://doi.org/10.4103/ajts.
ajts_148_22.
Kasman, Laura M., and La Donna
Porter. “Bacteriophages.” In
StatPearls. Treasure Island (FL):
StatPearls Publishing, 2024.
www.ncbi.nlm.nih.gov/books/
NBK493185/.
Kitson, Philip J., Guillaume Marie,
Jean-Patrick Francoia, et al. “Dig-
itization of Multistep Organic
Synthesis in Reactionware for
On-Demand Pharmaceuticals.”
Science 359, no. 6373 (January
19, 2018): 314–19. https://doi.
org/10.1126/science.aao3466.
Kreitz, Joseph, Mirco J. Friedrich,
Akash Guru, et al. “Program-
mable Protein Delivery with a
Bacterial Contractile Injection
System.” Nature 616, no. 7956
(April 2023): 357–64. https://doi.
org/10.1038/s41586-023-05870-
7.
LeMieux, Julianna. “New Organ-
oid AirGels May Boost Infectious
Disease Research.” Genetic
Engineering and Biotechnolo-
gy News, August 1, 2023. www.
genengnews.com/topics/
infectious-diseases/new-or-
ganoid-airgels-may-boost-infec-
tious-disease-research/.
Lin, Belle, Isabelle Bousquette,
and Angus Loten. “Synthetic
Biology Moves From the Lab to
the Marketplace.” Wall Street
Journal, September 19, 2023.
www.wsj.com/articles/synthetic-
biology-moves-from-the-lab-to-
the-marketplace-3f409a87.
Lv, Hui, Nuli Xie, Mingqiang Li, et
al. “DNA-Based Programmable
Gate Arrays for General-Purpose
DNA Computing.” Nature 622, no.
7982 (October 2023): 292–300.
023-06484-9.
“Machine Learning for Molecu-
lar Simulation. Annual Review
of Physical Chemistry 71 (April
2020). www.annualreviews.
org/doi/10.1146/annurev-phy-
schem-042018-052331.
Melton, Lisa. “Synbio Salvag-
es Alcohol-Free Beer.” Nature
Biotechnology 40, no. 1 (Janu-
ary 1, 2022): 8–8. https://doi.
org/10.1038/s41587-021-01202-0.
Merchant, Amil, Simon Batzner,
Samuel S. Schoenholz, et al.
“Scaling Deep Learning for Mate-
rials Discovery.” Nature 624, no.
7990 (December 2023): 80–85.
023-06735-9.
“Metagenomics.” www.genome.
gov/genetics-glossary/Metage-
nomics.
MIT News. “AI System Can
Generate Novel Proteins That
Meet Structural Design Targets.”
April 20, 2023. https://news.mit.
edu/2023/ai-system-can-gen-
erate-novel-proteins-structur-
al-design-0420.
SELECTED SOURCES
TECH BIOENGINEERING

MIT News. “Harnessing Syn-
thetic Biology to Make Sustain-
able Alternatives to Petroleum
Products,” July 20, 2023. https://
news.mit.edu/2023/viso-
lis-startup-sustainable-alter-
naties-petroleum-0720.
New Atlas. “Synthetic Biology
Breakthrough Fixes CO2 from
the Air Better than Nature.” Jan-
uary 3, 2024. https://newatlas.
com/biology/synthetic-biolo-
gy-co2-fixing-better-nature/.
Ni, Bo, David L. Kaplan, and
Markus J. Buehler. “Generative
Design of de Novo Proteins
Based on Secondary-Struc-
ture Constraints Using an
Attention-Based Diffusion
Model.” Chem 9, no. 7 (July
13, 2023): 1828–49. https://
doi.org/10.1016/j.chem-
pr.2023.03.020.
Northwestern Engineering.
“When Bioelectronics and
Synthetic Biology Unite.” www.
mccormick.northwestern.
edu/news/articles/2023/08/
when-bioelectronics-and-syn-
thetic-biology-unite/.
Pathak, Kalyani, Riya Saikia,
Aparoop Das, et al. “3D Printing
in Biomedicine: Advancing Per-
sonalized Care through Additive
Manufacturing.” Exploration
of Medicine 4, no. 6 (December
29, 2023): 1135–67. https://doi.
org/10.37349/emed.2023.00200.
Peel, Michael, and Clive Cook-
son. “Scientists Reprogramme
Microbes in Quest to Create
New Materials.” Financial Times,
January 11, 2024. www.ft.com/
content/19989e64-6f52-49ff-
bcb6-124c7e41ca7f.
“Programmable DNA Machines
Offer General-Purpose Comput-
ing.” IEEE Spectrum September
26, 2023.” https://spectrum.ieee.
org/dna-computing.
“Researchers Use Quantum
Biology to Understand Human
Response to Earth’s Magnetic
Field.” Johns Hopkins Universi-
ty Applied Physics Laboratory,
March 13, 2023. www.jhuapl.edu/
news/news-releases/230313-us-
ing-quantum-mechanics-to-un-
derstand-biology.
Rice University. “Rice University
Launches Rice Synthetic Biology
Institute to Improve Lives.” Jan-
uary 10, 2024. https://news.rice.
edu/news/2024/rice-universi-
ty-launches-rice-synthetic-biolo-
gy-institute-improve-lives.
Schneider, Gisbert. “Automating
Drug Discovery.” Nature Reviews
Drug Discovery 17, no. 2 (Feb-
ruary 2018): 97–113. https://doi.
org/10.1038/nrd.2017.232.
“Semiconductor Sequencing
Technology - US.” www.ther-
mofisher.com/us/en/home/
life-science/sequencing/
next-generation-sequencing/
ion-torrent-next-generation-se-
quencing-technology.html.
“Shapes 3D-Printed into Living
Worms Could Be Future of Brain
Implants.” New Scientist, April
17, 2023. https://cacm.acm.org/
news/272112-shapes-3d-printed-
into-living-worms-could-be-fu-
ture-of-brain-implants/fulltext.
Snyder, Alison. “Scientists Build
Yeast with Artificial DNA in a Ma-
jor Synthetic Biology Advance.”
Axios, November 9, 2023. www.
axios.com/2023/11/09/synthet-
ic-biology-yeast-artificial-ge-
nome.
South China Morning Post.
“Team Behind Extreme Animal
Gene Experiment Eyes Human
Nuclear Resistance.” March 29,
2023. www.scmp.com/news/
china/science/article/3215286/
chinese-team-behind-extreme-
animal-gene-experiment-says-
it-may-lead-super-soldiers-who-
survive.
Stock, Michiel, and Thomas E.
Gorochowski. “Open-Endedness
in Synthetic Biology: A Route to
Continual Innovation for Biolog-
ical Design.” Science Advances
10, no. 3 (January 19, 2024):
eadi3621. https://doi.org/10.1126/
sciadv.adi3621.
Sulis, Daniel B., Xiao Jiang, Chen-
min Yang, et al. “Multiplex CRIS-
PR Editing of Wood for Sustain-
able Fiber Production.” Science
381, no. 6654 (July 14, 2023):
216–21. https://doi.org/10.1126/
science.add4514.
Symes, Mark D., Philip J. Kit-
son, Jun Yan, et al. “Integrated
3D-Printed Reactionware for
Chemical Synthesis and Anal-
ysis.” Nature Chemistry 4, no. 5
(May 2012): 349–54. https://doi.
org/10.1038/nchem.1313.
“Synthetic Biology Market
Trends.” Mordor Intelligence.
Accessed January 28, 2024.
www.mordorintelligence.com/
industry-reports/synthetic-biol-
ogy-market/market-trends.
“Telomere-to-Telomere.” National
Human Genome Research Insti-
tute. www.genome.gov/about-ge-
nomics/telomere-to-telomere.
“Transformative Life Sci-
ences Trends for 2024.” Bio-
Spectrum, January 2, 2024.
www.biospectrumasia.
com/analysis/27/23518/
transformative-life-scienc-
es-trends-for-2024.html.
“US Trends in Synthetic Biolo-
gy Research Funding.” Wilson
Center. www.wilsoncenter.org/
publication/us-trends-synthet-
ic-biology-research-funding.
Wang, Kankan, Hongsheng Ouy-
ang, Zicong Xie, et al. “Efficient
Generation of Myostatin Muta-
tions in Pigs Using the CRISPR/
Cas9 System.” Scientific Reports
5, no. 1 (November 13, 2015):
srep16623.
SELECTED SOURCES
TECH BIOENGINEERING

Williams, Rory L., and Chang C.
Liu. “Accelerated Evolution of
Chosen Genes.” Science 383, no.
6681 (January 26, 2024): 372–73.
https://doi.org/10.1126/science.
adn3434.
Wong, Carissa. “UK First to
Approve CRISPR Treatment for
Diseases: What You Need to
Know.” Nature 623, no. 7988
(November 16, 2023): 676–77.
https://doi.org/10.1038/d41586-
023-03590-6.
World Economic Forum. “How
Investment into Synthetic Biolo-
gy Could Help Achieve Net Zero,”
January 2, 2024. www.weforum.
org/agenda/2024/01/synthet-
ic-biology-potential-at-scale-in-
vesting/.
Yang, Fei, Kaili Zheng, and Yu
Yao. “China’s Regulatory Change
Toward Genome-Edited Crops.”
Trends in Biotechnology, De-
cember 30, 2023. https://doi.
org/10.1016/j.tibtech.2023.12.008.
Yin, Wenqi, Chao Qiu, Hangyan
Ji, et al. “Recent Advances in Bio-
molecule-Based Films and Coat-
ings for Active and Smart Food
Packaging Applications.” Food
Bioscience 52 (April 1, 2023):
fbio.2023.102378.
Zaman, Qamar U., Ali Raza, Jorge
Lozano-Juste, et al. “Engineering
Plants Using Diverse CRIS-
PR-Associated Proteins and
Deregulation of Genome-Edited
Crops.” Trends in Biotechnology,
org/10.1016/j.tibtech.2023.10.007.
Zarnack, Kathi, and Eduardo
Eyras. “Artificial Intelligence and
Machine Learning in RNA Biolo-
gy.” Briefings in Bioinformatics
24, no. 6 (November 15, 2023):
bbad415. https://doi.org/10.1093/
bib/bbad415.
Zhu, Jingen, Himanshu Batra,
Neeti Ananthaswamy, et al. “De-
sign of Bacteriophage T4-Based
Artificial Viral Vectors for Human
Genome Remodeling.” Nature
Communications 14, no. 1 (May
30, 2023): 2928. https://doi.
org/10.1038/s41467-023-38364-1.
SELECTED SOURCES
TECH BIOENGINEERING

ENERGY • CLIMATE

TECH
TABLE OF CONTENTS
ENERGY CLIMATE
330 Top Headlines
331 State of Play
332 Key Events
Developments
334 Why Energy Climate
Trends Matter to Your
Organization
335 When Will Energy
Climate Trends Impact
Your Organization?
To Consider
338 Central Themes
340 Ones To Watch
341 Important Terms
342 Energy Production
343 New Solar
343 Concentrated Solar
343 Multitasking Solar
343 Dispatchable Solar
347 Digitalization
348 Small Scale Hydro
348 Hydro as a Water Battery
348 New Turbine Design
349 Clean Fuels
349 Hydrogen
349 Reducing the Cost of Hydrogen
Production
349 New Base Materials
349 Biofuels
349 Biofuels From Hemp
350 Biofuels From Algae
350 Syngas From the Sun
351 Fossil Fuel Innovation
351 Rededication of Contaminated
Land
351 Reducing the Carbon Footprint
of Fossil Fuels
351 Methane Emissions
Reductions
352 New Nuclear
352 Fusion
352 Fusion Momentum
352 Privately led Fusion Projects
352 Molten Salt Reactors
352 Laser-Driven Fusion
352 Small Modular Reactors
354 Emerging Forms of
Energy Production
354 Energy From Thin Air
354 Wave Power
354 Tidal Turbines
354 Mimicking Photosynthesis
355 Alternatives to Minerals and
Rare Earths
355 Local Sourcing
355 Alternative Materials
355 Rare Earth Avoidance
356 Scenario: What If the World
Runs on Thermal Energy?
357 Energy Infrastructure
358 Energy Storage
358 TPV Batteries
358 Iron Batteries
358 Solid State Batteries
359 Gravitational Energy Storage
359 Flow Batteries
359 Capacitors
359 Compressed Air Storage (CAES)
360 Energy Transport
360 UHV Power Lines
360 Superconductors
361 The Grid
361 Dynamic Line Rating (DLR)
Systems
361 Balancing the Flow of Power
Within the Grid
362 Emissions Removal
363 Carbon Capture and Storage
363 Natural CSS
363 Direct Air Capture
363 CO2 Storage
343 Perovskite Cells
344 Dye-sensitized Solar Cells
344 Organic Solar Materials
344 Solar Thermophotovoltaics
345 New Wind
345 Offshore Floating Wind
Turbines
345 Vertical Wind Turbines
345 Going Bladeless
345 Airborne Wind Energy
346 AI Wind Farms
346 Dispatchable Wind Energy
Production
347 Other Renewables
347 Geothermal
347 Supercritical Geothermal
347 Using Geothermal for
Energy Storage
347 Ocean Thermal Energy
Conversion
347 Hydropower

TECH
TABLE OF CONTENTS
ENERGY CLIMATE
364 CCS-as-a-Service
365 Carbon Utilization
365 Carbon-Based Food
365 Diamonds
365 Hair Care, Soap, and Laundry
Detergent
365 Chemical Production
366 Carbon Tracking
366 Funding Influx
366 Tracking Platforms
366 Government Investment and
Regulation
366 Individual Consumer Tracking
367 Scenario: What If CO2 Is the
Foundation for Consumer
Products?
368 Emissions Reductions
369 Green Processes
369 Construction
369 Carbon Neutral or Negative
Building Materials
374 Mushroom Plastic
374 Self-Healing Materials
375 Lab Grown Leather
376 Green Transport
376 Micromobility
376 Cars
376 Heavy Trucks
376 Trains
377 Air
377 Ocean
378 Environmental Manipulation
379 Earth
379 Rewilding: Animals/Nature
379 Bioengineering
380 Sky
380 Geoengineering
380 Solar Geoengineering
380 Cloud Seeding
381 Ocean
381 Microplastics
381 Living Breakwater
381 Ocean Chemistry
381 Gene Editing
382 Scenario: What If a Sunshield
Leads to a New World Order?
383 Effects of Climate Change
384 Monitoring Climate Change
384 Emissions
384 Extreme Weather
384 Crops
385 Community-based Efforts
385 Digital Twins of Earth
386 Living in a New Reality
386 Floating Cities
386 Other Concepts for Cities
386 Underground Climate Change
386 Indonesia’s New Capital
386 Domed Cities
387 Climate Economy
388 Carbon Credits
388 Investment From Traditional
Banks
388 Verification Methodology
388 Focus on Asset Management
388 Blockchain Integration
388 Measuring Carbon
Sequestration
389 Blue Economy
389 Blue Carbon Offsets
389 Ocean land mapping
389 Foundational Data US
389 Data Platforms
390 Fishing
390 Desalination
391 Authors
369 3D Printed Houses
370 Heavy Industries
370 Steel and Iron
370 Chemicals
371 Agriculture
371 Plants Without Plants
371 Fertilizer Innovation
371 CRISPR Edited Crops
371 Waste
372 Digital Emissions
372 Reducing Carbon Emissions
372 Measuring Emissions
373 Recycling
373 AI Waste-Sorting Robots
373 Food
373 Plastic
374 Green Materials
374 Cross-Laminated Timber
374 Mycelium

ENERGY CLIMATE
TECH
Private and public
climate efforts widen
and accelerate in
the light of the most
extreme weather we
have ever experienced
01
02
03
04
05
Investment is shifting from clean energy to enabling technologies
As wind and solar provide more power, venture capital is targeting bottlenecks
in supporting infrastructures such as the grid and monitoring technology.
Extreme weather sets new records
2023 was the hottest year in the history of humankind; as the ocean warms,
the collapse of the Gulf Stream is projected to be much more likely.
Tracking carbon has become popular with regulators
Regulators have shifted their focus from establishing the carbon market to
adding transparency, standards, and verification frameworks.
Polluters being held accountable
A slew of groundbreaking litigation set precedents for holding governments and
companies accountable for the effects of climate change caused by their (in)action.
Governments are considering a broader set of tools
Regulators are exploring solar geoengineering, considering softer regulations for
CRISPR crops, and redesigning urban concepts to adapt to hostile environments.
TOP HEADLINES

STATE
OF PLAY
In 2023, attention shifted from scaling of renewable energies and electric
vehicles to ensuring that these scaled technologies could be viably inte-
grated into existing systems. Venture capital moved from transportation,
energy, and food and land use to the built environment and heavy industry
decarbonization, making significant innovation likely. Governments tight-
ened ESG regulations, putting pressure on companies to bring transparen-
cy into carbon emission reporting, even while standardization frameworks
for measuring and reporting these emissions along the entire supply
chain are still being developed. These increased reporting pressures run
the risk of leading to a focus on the wrong metrics, crippling effective envi-
ronmental initiatives.
Environmental impacts are now defined more broadly, which increases
companies’ responsibilities. The framework passed at the 2022 UN’s Bio-
diversity Conference amplified the protection of biological ecosystems, eq-
uitable access, and human rights. In 2023, COP28 had its first-ever health
day, and nations committed to include food systems into their updated
Nationally Determined Contributions. Further responsibilities might come
from a series of lawsuits which decide if actors can be held responsible for
the effects of climate change they caused. If courts rule they are, prepare
for drastically changed efforts in regulation and business practices.
From revolution to evolution:
The global focus shifts to the
creation of an enabling ecosystem
for climate action at scale.
TECH ENERGY CLIMATE
331

TECH
MARCH 8, 2023
Carbon injected in the sea
Denmark became the first nation
to import CO2 for the purpose of
burying it in the North Sea.
MARCH 14, 2023
Forever chemicals ban
In the US, toxic PFAS are
banned from water while
the EU backpedals.
JULY 5, 2023
Looser GMO rules in EU
The European Commission
proposes looser restrictions in
light of innovation in the field.
JUNE 29, 2023
Fraud task force for ESGs
The Commodity Futures Trading
Commission announces a task
force to combat ESG related fraud.
DECEMBER 3, 2023
First Ever COP Health Day
The acknowledgment of the intercon-
nectedness of health and environment
at the UN Climate Change Conference
gives hope for integrated actions.
KEY EVENTS
ENERGY CLIMATE

TECH
CHANGE IS THE NEW NORMAL
Two opposing forces will make the
corporate landscape highly volatile
and unpredictable. On the one hand,
active regulatory bodies, scaling of
renewable energy production, and
increased investment in innovation
that aims to solve the remaining
bottlenecks will require and empower
corporations to integrate sustain-
ability in ways not imaginable just
a few years ago. On the other hand,
economic headwinds in the form of
heavy inflation and a looming reces-
sion might lead consumers to priori-
tize affordability over sustainability,
and escalating geopolitical tensions
could strain the supply chain, in-
crease the price of raw materials, and
hinder collaboration in research—
slowing down innovation.
Price Beats Ethics
As inflation soars and the fear of a recession
festers, consumers might put affordability
above climate considerations, at least when it
comes to their wallets. This will put additional
strain on business owners, as they determine
how to adjust for climate demands.
Shifts in Mobility
More and more people, at least in Europe, are
committing to not use air travel. The resulting
smaller spheres of mobility could lead to an
increasing need for companies to have local
hubs, especially as the terms of remote work
are still being negotiated.
Rethinking Supply Chains
Governments are expanding their require-
ments for reliable and consistent reporting of
direct and indirect emissions (scope 1, 2, and
3), putting pressure on corporations to curate
their vendor networks and on the vendors to
ensure their operations are still profitable
under increased standards.
From Early Bird to Night Owl
Excessive heat impacts daily lives around the
globe, forcing people to stay indoors during
the day and only leave their houses in the eve-
ning. As these heat phases expand, industries
dependent on people’s physical presence will
have to rethink operations as habits and tim-
ing of activities shift.
Accountability Changes
The resolution of a number of upcoming
court cases will begin to determine govern-
ment and corporate responsibility for climate
change. If verdicts fall on the side of holding
institutions responsible, we could see funda-
mental changes in how climate is addressed.
Climate Upskilling
With climate regulation expanding and evolv-
ing, and climate technology innovation accel-
erating (thanks to AI) and scaling, companies
need to make sure they have the necessary
know-how in-house to understand and moni-
tor relevant developments.
Media
Telecom
Distribution
ENERGY CLIMATE

TECH
Energy and climate
technologies can help
businesses transform in
revolutionary ways that
increase profitability
significantly: Changed
processes, new materials,
updated supply chains, a
different set of partners
and collaborators, and
investment in research
can lead to new products
and services—and a
competitive edge.
Embracing climate
technology brings
long-term savings.
Reduced energy use and
operational expenses
enhance overall
efficiency, positively
impacting the bottom
line. This positions the
company for sustainable
growth, demonstrates
commitment to a green
future, and fosters a
culture of environmental
responsibility.
Increased frequency,
severity, and wider
occurrence of extreme
weather events force
all businesses to think
about how to mitigate
impacts, not just those
in disaster-prone regions.
Flexible measures are
needed to maintain
continuity, protect
assets, and promote
sustainable growth in the
face of an increasingly
unpredictable climate.
As climate-conscious
funding continues to
surge, businesses with
innovative climate
solutions gain greater
access to capital,
enabling them to
scale operations and
accelerate the transition
to a more sustainable
future. This provides a
unique opportunity to
stay ahead of regulatory
requirements and tap into
growing green markets.
Businesses that
invest in their own
renewable energy
production can achieve
a high degree of energy
independence, reducing
their vulnerability to
fluctuations in pricing
and increasing grid
failures. They’ll gain
greater control over energy
supply, with insulation
from market uncertainties
and enhanced long-term
stability.
Implementing climate
technologies and
adhering to environmental
regulations may raise
production costs, due to
upfront investment or
compliance expenses,
impacting profitability.
Strategic planning
and collaboration are
crucial to navigating this
transition successfully.
Opportunity for
Innovation
Cost Savings
Efficiencies
Climate Risk
Resiliency
Investment Financing
Opportunities
Energy
Independence
Attack on
Profitability
WHY ENERGY CLIMATE TRENDS MATTER TO YOUR ORGANIZATION
ENERGY CLIMATE

TECH
335
WHEN WILL ENERGY CLIMATE TRENDS DISRUPT YOUR ORGANIZATION?
Agriculture
Energy generation
Transportation
Construction
Insurance
Real estate
Hospitality
Retail
City management
Manufacturing
Pharma
Financial services
Consumer goods
Health care
Information technology
Water management
Infrastructure development
Waste management
Space exploration All trends will be relevant
ENERGY CLIMATE

Threats
Assessing the existing expertise and hiring or training employees with
the necessary skills to integrate climate technologies is a must yet can be
challenging in an already tight labor market and an overworked workforce.
Companies might be confronted with a much higher cost of doing business
after integrating sustainable processes and raw materials. This might pose
existential risks or fundamental shifts in operations and threaten profitability.
Geopolitical tensions and a higher frequency of extreme weather events
increase the risk for supply chain disruptions, as well as rising prices and
scarcity of necessary base materials and labor.
As tracking capabilities expand through scope 1, 2, and 3 emissions,
businesses should be prepared to find that their carbon emissions footprint
is much more significant than expected (and less under their control).
Proactive regulators are needed to ensure we reach our climate goals. However,
new regulations as well as changes to existing rules will redesign business
environments and create a strain for companies as they try to navigate their
operational transformation.
Opportunities
Look for new partners to collaborate with and selectively outsource activities for
implementing sustainability measures as part of a new supportive ecosystem.
This keeps costs at bay and expands the opportunity for cross-pollination of
ideas.
Price volatility and stricter regulation will trigger a rethink of business operations.
This can lead to new efficiencies in tangential contexts outside of environmental
considerations and positively affect margins.
First movers will be able to define the new normal. They’ll build competitive
advantage by transforming common business procedures in their industry and
setting standards as they improve their business operations.
Think beyond the changes necessary to comply with sustainability requirements
by also considering structural changes that could make the company more
flexible and responsive to innovation.
Energy and captured CO2 emissions are just two examples of products that the
integration of climate technologies can add to a company’s portfolio. Thinking
expansively can open up avenues into new industries, including but not limited
to energy.
TECH
ENERGY CLIMATE

TECH
With innovation making
huge strides in carbon
tracking, invest time into
finding the platform and
sensor ecosystem that’s the
best fit for your business
and supply chain. These
technologies will touch ev-
ery aspect of your business,
making the implementation
of new hardware and soft-
ware very expensive.
Consider all workforce as-
pects: Companies need to
weigh the skills and exper-
tise available, where to hire
or upskill, whether structure
and hierarchies need to
evolve, as well as where
employees should be
working. All these factors
are relevant for the carbon
footprint but also need to
be considered for maximiz-
ing profitability.
Sustainability is becoming
a broader effort and now
includes biodiversity, ethi-
cal practices, and protec-
tion of indigenous commu-
nities. Work toward gaining
insights about these areas
throughout your entire
supply chain to prepare for
regulatory pressures.
Explore new avenues to not
only stay on top of innova-
tion but to become part of
the ecosystem of investing,
research and develop-
ment. This ensures that the
bottlenecks and problems
specific to your business
are being solved, and might
even lead to new business
models licensing the devel-
oped technology.
Global demographics will
change significantly over
the next few decades,
with migration, aging, and
overpopulation all con-
tributing factors. Investi-
gate how this affects your
current customer base,
and where new markets,
either in regards to location
or customer profile, could
emerge globally.
Clearly define goals,
benchmarks, and deadlines
for the assessment and
implementation of climate
technologies in your com-
pany. Identify a sensible
organizational structure,
relevant stakeholders, re-
sponsible parties, and how
the company will efficiently
integrate the findings and
technologies.
1 4
2 5
3 6
ENERGY CLIMATE

CENTRAL THEMES
TECH
338
ENERGY CLIMATE
Proactive Regulators
On both sides of the Atlantic, regulators stepped up
to enable significant climate action. The EU’s Green
Deal Industrial Plan includes easier access to financ-
ing, simplification of permitting processes, regulatory
sandboxes for member states, and the announcement
of the European Hydrogen Bank to quickly scale the
fuel’s use in the region. In the US, regulators banned
certain forever chemicals (PFAS) in drinking water,
strengthened pollution standards for cars and trucks,
and widened eligibility for clean-energy tax credits. To
avoid greenwashing, the EU proposed the Green Claims
Directive to eliminate misleading messaging. In the
US, the Commodity Futures Trading Commission cre-
ated a task force to combat environmental fraud, and
the Securities and Exchange Commission proposed
stricter climate disclosure rules for publicly traded
companies.
Focus on Enabling Technologies
After passing the $1 trillion investment mark in 2022,
overall venture capital in climate tech dropped 40%
in 2023. However, most of that decrease took place in
later stages and growth funding; seed funding actual-
ly grew 23%, and the number of deals rose by 34%. In
addition, investors are showing enthusiasm for solving
new problems, after renewables have moved into the
scaling phase. Investment in the big three—transpor-
tation, energy, and food and land use—saw significant
drops, while investment in industry processes and the
built environment, both historically underfunded, saw
gains. But there is still much to be done: Transforming
the grid to accommodate non-dispatchable energy
sources; searching for alternative materials for batter-
ies, solar cells, and wind turbines; capturing and stor-
ing carbon; tracking emissions; and bringing transpar-
ency and validity to carbon markets are just some of
the bottlenecks that urgently need innovation.
Alternative Materials
The race to find alternative materials for renewable
energy production and tangential technologies (such
as batteries) is on. In a tense geopolitical environment
where the location of needed raw materials is limit-
ed to areas often plagued by political and economic
instability, there’s an increased desire for nations to be
resource independent. Recently, Sweden and Norway
made inroads on this independence when the larg-
est rare earth and mineral deposits were found in the
region. While China responded to the US’s 2022 export
controls on advanced computing and semiconductors
manufacturing equipment with its own restrictions on
exporting gallium and germanium (needed for solar
technologies), impact is less significant thanks to the
ability to expand in alternative markets in the mid-
term. Innovation intensified last year around finding
alternative materials for climate technologies, pre-
dominantly in electric vehicle motors (rare earth) and
batteries (metals).

CENTRAL THEMES
TECH
339
ENERGY CLIMATE
Looking Beyond Carbon
While carbon dioxide emissions are front and center
in the climate conversation, a more holistic approach
has gained significant traction and increased public
awareness. Both the US and the EU have taken steps to
control methane emissions, and global leaders came
together during the United Nations Biodiversity Con-
ference (COP 15) in December 2022 to agree on global
action on addressing biodiversity loss, restoring eco-
systems, and protecting indigenous rights. As a result,
30% of the planet and 30% of degraded ecosystems will
be under protection by 2030. EU regulators have agreed
on the Nature Restoration Law, which requires coun-
tries to work toward restoring at least 20% of land and
sea areas, and the US expressed its intention to join
the High Ambition Coalition on Biodiversity Beyond
National Jurisdiction, supported by the UN. Currently,
95% of the ocean falls beyond national jurisdiction,
and the coalition represents a group of countries that
have pledged to protect that ecosystem. Rewilding
efforts have also increased, often on state, local, and
even individual levels.
Tracking Climate Change
The tightened regulatory requirements for companies
to reduce their carbon footprint increases the need for
cohesive and reliable emissions tracking, as well as
common measurement standards. In the carbon credit
market, frameworks to verify actual results of carbon
projects and create a viable rating system are crucial
for credibility and effectiveness. As extreme weather be-
comes more frequent, severe, and widespread, govern-
ments have increased their efforts to build better pre-
diction systems to protect lives and ecosystems. Amid
a landscape where insurers are pulling out of states like
California and Florida because risks deemed uninsur-
able, these systems can provide the industry with the
data necessary to calculate risks. The development of
these monitoring technologies is being accelerated by
artificial intelligence and innovation around imaging,
such as hyperspectral imaging. Also helping to unify the
integrity frameworks are joint efforts by various stake-
holders in the voluntary carbon markets.
Blue Economy on the Rise
Originally used by small developing island states, the
term “blue economy” today encompasses socially
equitable, environmentally sustainable, and economi-
cally profitable ocean-based industries and spaces. In
2023, the US launched a global initiative of more than
$800 million to protect oceans and support devel-
oping nations. This follows the EU’s previous efforts
to create the Atlantic Smart Ports Blue Acceleration
Network to transform Atlantic ports, the G20 Supreme
Audit Institutions cooperating to meet the challenge of
auditing the blue economy, and African funds’ focus on
innovation to advance blue economy solutions for the
continent. These initiatives triggered a flurry of innova-
tion focusing on the ocean, from means and devices to
collect data underneath the surface, to platforms that
synthesize and analyze the data and make it available
to all relevant stakeholders, to advanced desalination
methods and increased support for sustainable fish-
ing. As the ocean is also moving more into the climate
conversation for carbon sequestration and power
generation, we can expect to see an acceleration of not
just technological advances but also novel business
activity in this space.

Dr. Stefaan De Wolf, professor of material
science and engineering of the KAUST So-
lar Center, for setting the world record for
tandem solar cell efficiency.
Alexander Bormann, founder of EnerKite,
for expanding the EV-charging infrastruc-
ture with airborne wind.
Dr. Motiar Rahaman, research associate
at University of Cambridge, for leading
research on a solar-powered reactor that
converts CO2 captured from industrial
waste or air into sustainable fuel.
Andrew Ponec, co-founder and CEO of
Antora Energy, for developing thermal
energy storage that turns renewable energy
into on-demand power usable by the heavy
industries.
Xiaomeng Liu, researcher at University of
Massachusetts Amherst, for his research
on creating energy from air.
Mateo Jaramillo, CEO of Form Energy, for
building a 1 gigawatt hour demo system of
an iron air battery, expected to come online
in 2025.
John Connell, senior scientist at NASA
Langley Research Center, for advancing
research on a solid state battery to be used
in aviation.
Lee Suk-bae, Ji-Hoon Kim, and Young-
Wan Kwon of the Quantum Energy
Research Centre, for their preprint on
developing the first room temperature
superconductor.
Hudson Gilmer, co-founder and CEO of
LineVision, for creating the only non-con-
tact overhead power line monitoring sys-
tem, used by UK’s National Grid.
Isabella Arzeno-Soltero, postdoctoral
scholar at Stanford University, for deter-
mining that seaweed farming cannot scale
enough to meet climate goals.
Dr. Graciela Chichilnisky, co-founder of
Global Thermostat, for creating one of the
largest direct air capture machines ever
operated.
Talal Hasan, CEO of 44.01, for leading the
first CO2 mineralization project in the Mid-
dle East that utilizes seawater.
Pasi Vainikka, CEO of Solar Foods, for
creating food out of renewable energy and
carbon dioxide.
Freddie Lintell, founder and CEO of Ree-
wild, for creating a carbon tracking app for
consumers.
Josh Dorfman, co-founder and CEO of
Plantd, for transforming perennial grass
into carbon-negative building materials.
Virginia San Fratello, architect, for 3D
printing houses out of indigenous materi-
als such as salt and clay.
Dr. William Dichtel, chemistry professor at
Northwestern, and Brittany Trang, Sharon
Begley Science Reporting Fellow at STAT,
for developing a low temperature, inexpen-
sive method to break down PFAS.
Shimrit Bar-El, co-founder and CRO at
Novella, for growing botanical ingredients
without the plant.
Hunter Swisher, founder of Phospholu-
tions, for reducing the amount of phospho-
rus in fertilizer by 50%.
Nathalie Berezina, founder and CEO of
Norbite, for transforming plastic waste
into sustainable products with the help of
moths.
Khaled Hassounah, co-founder and CEO
of Ample, for providing an alternative to EV
charging: battery swapping.
Tim Duehrkoop, co-founder and CEO of Xil-
va, for developing a methodology to assess
forest sequestration projects.
Dr. Tom Jackson and his team at Lough-
borough University, for the creation of the
“Data Carbon Ladder,” which enables the
tracking of digital emissions.
Bob Mumgaard, CEO of Commonwealth
Fusion Systems, for being a frontrunner in
the nuclear fusion race.
ONES TO WATCH
TECH
ENERGY CLIMATE

IMPORTANT TERMS
341
Ammonia
A compound of nitrogen and hydrogen. It can
be used directly as a fuel in direct combustion
processes, as well as in fuel cells or as a hydrogen
carrier. To be a low emissions fuel, ammonia must
be produced from low-carbon hydrogen and the ni-
trogen separated through the Haber process using
electricity generated from low-carbon sources.
Bioenergy
Energy content derived from biomass feedstocks
and biogas. It comes in solid, liquid, and gaseous
form. Its liquid form is often labeled biofuel.
Biogas
A mixture of methane, CO2, and small quantities
of other gases produced by anaerobic digestion of
organic matter in an oxygen-free environment.
Carbon capture, utilization, and storage (CCUS)
The process of capturing CO2 emissions from fuel
combustion, industrial processes or directly from
the atmosphere. Captured CO2 emissions can be
stored in onshore or offshore underground geologi-
cal formations, or used as an input or feedstock in
manufacturing.
biomass, and used as a source of energy or fuel.
Depending on the resource and energy type used
to produce it, hydrogen production can be emis-
sions-intensive or carbon neutral.
Liquid bioenergy (biofuel)
Liquid fuels derived from biomass or waste feed-
stock. They include ethanol, biodiesel, and biojet
fuels.
Conventional: Fuels produced from food crop feed-
stock such as sugar cane or vegetable oil, among
others.
Advanced: Fuels produced from non-food crop
feedstock that don’t directly compete with food and
feed crops for agricultural land.
Process emissions
Emissions from industrial processes that involve
chemical or physical transformations (separate
from fuel combustion).
Pyrolysis
Pyrolysis is the process of heating organic materi-
als, such as biomass, in the absence of oxygen.
Solid bioenergy
Charcoal, fuelwood, dung, agricultural residues,
wood waste and other solid wastes.
Traditional: Refers to the use of solid biomass with
basic technologies, such as a three-stone fire, often
with no or poorly operating chimneys.
Modern: Refers to the use of solid bioenergy in im-
proved cook stoves and modern technologies using
processed biomass such as pellets.
Solar photovoltaics (PV)
A process which converts sunlight into electricity
using a technology based on the photoelectric ef-
fect. With the photoelectric effect, materials absorb
photons (light) and release electrons, generating
electricity.
Zero carbon-ready buildings
A zero carbon-ready building is a highly energy
efficient building. It uses either renewable energy
sources directly or energy sources that can be de-
carbonized, for example electricity or district heat.
Zero emission vehicles (ZEVs)
Vehicles that operate without emitting CO2 emis-
sions (such as battery electric and fuel cell vehicles).
Direct air capture
A technology that captures CO2 from the atmo-
spheric air through a chemical reaction.
Dispatchable generation
Dispatchable generation is a source of electricity
that can be turned on or off, such as nuclear, mean-
ing it can be controlled. Non-dispatchable energy
sources, such as wind and solar photovoltaics,
cannot be controlled by operators.
Electrolysis
Electrolysis is a process where electric current
passes through a substance to effect a chemical
change. In hydrogen production, electricity is used
to split water into hydrogen and oxygen. If the
power used for the process comes from sustain-
able energy sources, the process does not produce
greenhouse gas emissions.
Energy intensity
Energy intensity is the amount of energy used to
produce a certain level of output.
Hydrogen
Hydrogen is the simplest and most abundant
element in the universe. It can be produced from a
variety of resources such as water, fossil fuels, or
TECH ENERGY CLIMATE

TECH
ENERGY CLIMATE
ENERGY
PRODUCTION

TECH
Concentrated Solar
Concentrated solar power (CSP) uses para-
bolic mirrors or “heliostats” to focus sunlight
and generate extremely high temperatures. A
common implementation of CSP is the “pow-
er tower,” in which concentric circles of helio-
stats all focus on the receiver of a single, cen-
tral tower hundreds of meters off the ground.
Although the concept was first developed in
the 1970s and ’80s, new methods and plant
designs have inspired a modest resurgence,
particularly in Australia and Africa. Bolstering
the case for CSP plants is the dispatchability
of the energy generated. Because CSP gen-
erates thermal energy, it can be transferred
using liquid sodium and stored long term in
molten salt reservoirs. Due to the abundance
of the elements required, large-scale thermal
storage would be relatively cheap when com-
pared with the rare earth elements needed for
electric batteries, such as lithium and cobalt.
For this reason, concentrated solar is being
viewed as an economical alternative to photo-
voltaic solar for nighttime use.
Dispatchable Solar
Dispatchable power enables a generation
source to scale up or down based on fluctu-
ations in demand. Solar power installations,
traditionally implemented in a way that
always maximizes energy output, are getting
smarter, able to adjust individual panel an-
gles to change their aggregate power gen-
eration and better accommodate the needs
of the grid. On large-scale solar farms, this
system of dispatching can go so far as to
independently adjust subsections of panels,
or even individual panels, to compensate
for others that are temporarily obscured
by cloud cover or experiencing operation-
al issues, thus ensuring a responsive and
consistent output. Counterintuitively, many
of the problems with static solar power gen-
eration stem from creating too much pow-
er: When the sun is at its zenith and solar
generation peaks, solar plants can potential-
ly produce so much energy they overwhelm
local electric grids. Dispatchable solar power
eliminates this problem, creating a more
dynamic and responsive source of power.
Perovskite Cells
Perovskite is a crystalline compound that
can be used as a semiconductor in solar cells
as an easier, cheaper, and more sustain-
able alternative to silicon. Furthermore, it is
transparent and flexible, making it easier than
silicon to integrate into the landscape. In April,
researchers at the King Abdullah University
of Science and Technology in Saudi Arabia
developed a new combination silicon/per-
ovskite solar cell, which achieved an efficiency
of 33.2%, a new world record for two-junction
solar cells. Advances in perovskite-based cells
are occurring quickly, and the technology is
regularly setting new efficiency and lifetime
records.
Multitasking Solar
Companies are building new solar installa-
tions with a mind toward additional benefits
beyond electrical power generation. When
installed in tandem with wind turbines, solar
panels are arranged to take advantage of
the Venturi effect, essentially creating wind
tunnels. These configurations increase air-
flow and wind turbine output by up to 60%.
Solar windows are becoming transparent
enough not to impact aesthetics in building
construction, thanks to organic semicon-
ductors that can be liquified and spread as a
coating on glass. In California, a pilot pro-
gram dubbed Project Nexus is placing solar
panels above irrigation canals in the San
Joaquin Valley. Researchers estimate that
placing similar solar panel canopies above
the state’s 4,000 miles of open canals would
generate 13 gigawatts of power while simul-
taneously saving 63 billion gallons of water
annually in a region that has seen severe
drought over the past decade. The solar can-
opies cool water temperatures, halt evapora-
tion, and prevent the growth of water-based
plants.
NEW SOLAR
ENERGY CLIMATE

NEW SOLAR
TECH
Dye-sensitized Solar Cells
Dye-sensitized solar cells (DSC) are made
using organic dyes that capture photons from
light. They are cheaper to manufacture than
silicon solar cells, more flexible in their de-
sign, and can even be semitransparent. Fur-
ther, they operate on a wider spectrum of light
than traditional silicon solar cells. Ambient
Photonics, an Amazon-backed startup, has
completed construction on a new, large-scale
DSC manufacturing facility. The company
claims its products can operate at low levels
of light, such as indoor environments, and is
aiming to eliminate the need for batteries in
small electronic devices.
Solar Thermophotovoltaics
While normal photovoltaic cells respond
only to less than half of the sun’s rays that
are visible, solar thermophotovoltaics create
electricity from heat waves. Antora Energy,
a US-based startup, has completed a large-
scale manufacturing facility that produces
thermophotovoltaic cells with 40% efficien-
cy. Meanwhile, researchers at the University
of Houston have developed a new design for
thermophotovoltaic cells with an improved
intermediary layer, which prevents thermal
energy from being wastefully radiated away.
This development could push efficiency lev-
els even higher.
Organic Solar Materials
Organic solar materials that can be printed
or stuck onto surfaces have continued to
shrink in size. These photovoltaics can be
50 microns thin, less than the width of a
human hair, and can be adhered to surfac-
es after they’re manufactured, ostensibly
making any surface a power-generating
opportunity. Because of the small footprint
of organic solar cells, they can also be ad-
hered to transparent surfaces. As a result,
solar windows are reaching the point where
they are transparent enough not to impact
aesthetics and could be used more widely in
buildings. Recent experimentation aims to
increase these organic solar cells’ viability
for power production on larger areas.
ENERGY CLIMATE
Flexible, transparent solar cells might soon be
adhered to building windows.

TECH
Offshore Floating Wind Turbines
Offshore floating wind technology is growing
with changes in base designs, innovative tur-
bine configurations, and strengthened regu-
latory backing in the US and Europe. In recent
years, “super-sized” structures have become
popular as major players unveil groundbreak-
ing designs. Among these is Wind Catching
Systems, a 2017-established entity situated
near Oslo, Norway. Pioneering the multi-tur-
bine approach, the company focuses on a
“floating wind power plant.” The Dublin-based
Gazelle, a company focused on advanced
offshore wind platforms, has introduced a
new hybrid dynamic mooring system—this
revolutionary platform design promises
unmatched stability, courtesy of its light-
weight and compact structure and flexibility
as it can be seamlessly assembled at global
port facilities. In the realm of US regulation,
the Biden administration has announced a
comprehensive 20-month study of the West
Coast’s burgeoning floating offshore wind
potential. This investigation aims to optimize
transmission networks and connect the grid
flows to cater to the chaotic wind patterns
of urban areas. Available in varied sizes to
suit diverse buildings, O-Wind turbines can
connect to the grid or operate independently
with battery units. From Xenecore, a design
that incorporates I-beam ribs and micro-
sphere structural foam amplifies power gen-
eration with its fan-shaped wind blades. And
because the future needs wind technology
that will work at relatively low wind speeds,
the fastest-growing energy sector across the
globe is bladeless wind energy. It leverages
vortex shedding, a vorticity phenomenon,
through a vertical cylinder affixed with an
elastic rod. It oscillates within the wind’s
range and powers an alternator system
to generate electricity. Prominent players
include Vortex Bladeless, Tyer Wind, Agile
Wind Power, Silent Wind, and Sway Turbine
AS in Europe; Saphon Energy in Africa; and,
Mag-Wind Vertical Axis Turbine, Atmocean,
Enomad, and SheerWind in the US.
Airborne Wind Energy
Kitemill, a Norwegian enterprise, is pioneer-
ing the advancement of airborne wind ener-
gy, which involves attaching a turbine to a
flying device, to revolutionize wind energy’s
effectiveness, versatility, and affordability.
Its latest triumph, the KM2 system, is double
the size of the KM1 prototype, has a 16-me-
ter wingspan, and integrates four propellers
for vertical takeoff. The system can generate
an average power cycle of 100 kilowatts. In
another promising collaboration, EnerKíte and
Volkswagen embarked on a feasibility study
to explore the potential of a mobile e-charging
station. These innovative airborne wind tur-
bines hold the promise of charging electric
vehicles even in the remotest regions, reduc-
ing reliance on conventional power grids and
enhancing green mobility.
with pioneering floating wind projects; it will
be supported by a $100 million public fund.
Vertical Wind Turbines
Vertical wind turbine innovation reached a
milestone with Norway’s March 2022 ap-
proval of a vertical-axis floating wind pilot
project. SeaTwirl and the Marine Energy Test
Centre will test the prototype for five years
near Lauplandsholmenoff. The project’s
progress stalled due to appeals from envi-
ronmental and fishing groups, but Norwe-
gian regulators’ rejection of the appeals en-
sures SeaTwirl’s S2X pilot can move forward
without further challenges. Additionally,
3D-printed vertical wind turbines are grow-
ing, with several companies experimenting.
Going Bladeless
New innovation in wind turbines includes
evolving designs beyond traditional blades.
O-Wind has pioneered an omnidirection-
al design that simultaneously captures
winds from all directions, a breakthrough
that uniquely blends horizontal and vertical
NEW WIND
ENERGY CLIMATE

TECH
AI Wind Farms
Like most other industries, AI is also reshap-
ing renewable energy. Israeli company vHive
has introduced an advanced tool for wind
turbine inspections, leveraging autonomous
data collection and a novel digital twin plat-
form. This innovation empowers wind farm
operators to strategically digitize assets, ele-
vate operational efficiency, and curtail output
decline. UK’s Cognitive Business is working
with RWE’s Robin Rigg offshore wind farm, de-
livering an AI-powered package that encom-
passes pattern recognition and production
forecasting to ensure precise maintenance
predictions and optimal performance. AI has
also proven pivotal in the setup, upkeep, and
enhancement of offshore wind farms. The
Dhalion Inspection System by Perceptual
Robotics is an interesting example with a
fully autonomous solution for wind blade
inspection, encompassing data acquisition,
insights, and decision-making.
Dispatchable Wind Energy Production
As scientists find more ways to store wind
energy, it will grow as a dispatchable elec-
tricity solution from which power grids can
demand electricity based on market needs.
One way to ensure such storage is by fusing
hydrogen with the grid, like what’s being
done in the Netherlands. There, an alliance
of prominent partners led by Shell is using
hydrogen from offshore wind energy plants
to stabilize the electricity grid, and aims to
maintain 70% renewable electricity by 2030.
The FlexH2 consortium, which includes TNO
and other industry players, collaborates
closely on joint research and pioneering
technology development, expediting integra-
tion into the energy matrix.
NEW WIND
ENERGY CLIMATE
AI helps to make Windfarms more efficient and maintain and optimize performance.

TECH
Geothermal
Unlike other renewables, geothermal ener-
gy represents a stable source of power that
doesn’t fluctuate with time of day or weather
patterns. This energy can be harnessed by
tapping into hot water and steam reservoirs
beneath the Earth’s crust. Once accessed, it
can be used for direct heating or to generate
electricity through geothermal power plants.
Supercritical Geothermal
Supercritical geothermal energy production
requires going deeper into the Earth’s crust
than conventional geothermal, tapping into
the extreme temperatures and pressures
present there (in excess of 374 degrees Cel-
sius and 221 bar, respectively). While con-
taining enormous power generation poten-
tial, these conditions also present extreme
technological challenges. Some of those may
be alleviated by recent research, which has
focused on tapping into supercritical geo-
thermal fluids at sites adjacent to volcanic
activity. At these sites, extreme temperatures
and pressures can be found closer to the
of electrical batteries, which require large
quantities of rare earth elements and are not
as efficient. Multiple research efforts have
recently focused on using carbon dioxide as
the storage medium, which has the added
benefit of sequestering greenhouse gases.
In the US, the National Renewable Energy
Laboratory has begun efforts to identify high
storage potential sites, and in Germany, the
Karlsruhe Institute of Technology is evaluat-
ing how geothermal energy storage might be
integrated into the existing grid.
Ocean Thermal Energy Conversion
Ocean thermal energy conversion generates
power from the temperature difference that
exists in ocean water. Because this differ-
ence is unaffected by weather and climate
conditions, it represents a reliable source of
sustainable energy. Furthermore, the sea-
water output from the process is usable in
commercial applications such as fisheries,
agriculture, and air conditioning. Japanese
shipping company Mitsui O.S.K. Lines has
announced a pilot program that would
pump water from a depth of 600 meters at
near-freezing temperatures to be used in a ti-
tanium heat exchanger; the process is expect-
ed to generate 1 megawatt of electric power by
2026.
Hydropower
Accounting for more than 15% of global elec-
tricity generation in 2022, hydropower con-
tinues to be the largest source of renewable
power, contributing more than wind, solar, and
biofuels combined. New opportunities for hy-
dropower manifest in efficiency, storage, and
small scale (but widespread) deployments.
Digitalization
Hydro plants have invested significantly in
digital management for their equipment and
systems, all with the aim of optimizing energy
output and safety. Utilizing cloud computing
and big data, operators hope to build machine
learning models to guide automated deci-
sion-making in these complex environments.
Given the risk involved with such a digital
transformation, some operators have even
Earth’s surface, greatly reducing the risk
and investment required to utilize them. In
Japan, the New Energy and Industrial Tech-
nology Development Organization has been
experimenting in volcanic regions of Hokkai-
do. Meanwhile, new geothermal research and
startups push the boundaries of existing
technologies. Researchers are testing work-
ing fluids composed of supercritical carbon
dioxide to transfer and pump heat back to
the Earth’s surface. Companies are develop-
ing new tools, like Thermochem’s probe and
logging tool rated to operate at temperatures
up to 400 degrees Celsius. Barriers to devel-
opment remain, however, mainly in the form
of policy and international cooperation.
Using Geothermal for Energy Storage
Geothermal energy storage techniques hold
energy in the form of increased temperature
and pressure which is “pumped” into the
earth and released as needed. Research-
ers and government agencies are looking
closely at this option for storing sustainably
generated energy in lieu of massive banks
OTHER RENEWABLES
ENERGY CLIMATE

TECH
opted to create digital twins of their plants,
enabling a safe environment for training the
algorithms of the future. Spain has made sig-
nificant investment so far, and officials plan
to onboard 160 of its power plants to a digital
management platform. China, too, is making
strides, and officials have put all data from
the Three Gorges Dam (the largest hydroelec-
tric project in the world) into the cloud of
Chinese company Huawei Cloud.
Small Scale Hydro
Small scale hydro projects are allowing en-
ergy suppliers across the globe to add more
renewable energy generation to their portfolio
without the need for massive, upfront invest-
ments of capital. In California, a pilot program
led by startup Emrgy places small, modular
turbines into irrigation canals to produce
modest amounts of electricity, between 2
and 10 megawatts. The turbines operate in
a manner that does not require damming
the water. This, paired with the fact that all
the canals used are pre-existing, means the
impact to the local environment is minimal.
ity by 2031. Spain started building a 200
megawatt plant in the Canary Islands. China,
meanwhile, continues to lead the world in
pumped storage capacity, with 51 gigawatts
currently in operation and more planned to
come.
New Turbine Design
Companies continue to innovate on the de-
sign of hydro turbines. Turbulent, a Belgian
engineering company, has developed an un-
derwater vortex turbine capable of operating
in remote locations. With low maintenance
and water flow requirements, it is ideal for
rural communities with simple irrigation.
Similarly, Emrgy is working on hydrokinetic
turbines that can be dropped into canals
and other low pressure, low flow environ-
ments. Voith Hydro in Germany made ad-
justments to the centuries-old Pelton Wheel
design, which allows its turbines to operate
in a horizontal configuration, as opposed to
vertical. More injectors can be utilized in its
operation, increasing its overall output. Addi-
tionally, companies such as Natel are work-
ing to minimize the environmental impact
of hydropower. Its latest turbine implements
a curved blade design aiming to drastically
reduce the rate of marine life fatality for crea-
tures passing through.
The European Union has also funded pilots
for small scale hydro in sites across Central
Asia. Though this region has the world’s sec-
ond largest potential for hydroelectric power,
larger hydro developments there have been
slow due to political, economical, and legal
factors, such as water rights claims and con-
cerns about environmental impact. Small
scale hydro has the ability to simultaneously
provide a significant source of power while
avoiding many of those pitfalls.
Hydro as a Water Battery
Hydro storage, or “pumped storage,” involves
pumping water into uphill reservoirs when
energy is cheap (or when renewables are op-
erating) and then allowing that water to flow
downhill and generate energy as needed.
Recently, large investments in the field have
been made across the globe. Switzerland
just opened a 20-gigawatt-hour plant in the
Swiss Alps that can transition from energy
storage to energy generation in less than 10
minutes. In Utah, a $2.5 billion project broke
ground that could provide 9GWh in capac-
OTHER RENEWABLES
ENERGY CLIMATE

TECH
Hydrogen
Hydrogen fuel technology has been around
since the 1950s, and for the past 75 years, it
has served as the main propulsion source for
spacefaring vehicles. Recent developments,
including updated designs for hydrogen fuel
cells, have reignited conversations of using
the fuel source closer to home. In particular,
“green hydrogen” (hydrogen fuel produced us-
ing renewable energy), has been viewed as an
emissions-free alternative for long distance
naval, aerial, and ground transportation. How-
ever, issues of economics and scale remain.
Reducing the Cost of Hydrogen Production
Researchers are making progress in reducing
the cost of green hydrogen generation. Many
of these efforts focus on improving the ma-
terials coating the electrodes used to extract
pure hydrogen from water. Separate research
teams, both in Korea, have perfected pro-
tective titanium oxide coatings that reduce
corrosion in the electrodes and have experi-
mented with using cheaper iron nitride as a
coating alternatives. An Oxford-based team
Biofuels
Biofuels encompass any fuel made or de-
rived from organic matter—typically corn,
sugar cane, or soy. These fuels include etha-
nol, biodiesel, and biogas. While biofuels rep-
resent an alternative to fossil fuels, concerns
have arisen over the amount of farmland
needed to provide significant amounts of
energy and the impact large scale operations
might have on food systems.
Biofuels From Hemp
Hemp represents a unique opportunity for
biofuels. It can be turned into multiple types
of fuel (e.g., ethanol, methanol, biodiesel)
and boasts one of the highest energy den-
sities of any land-grown crop. However, the
plant’s association with cannabis is often
seen as hindering widespread adoption
(industrial hemp has only a fraction of the
THC associated with recreational marijuana).
A bipartisan bill in the US hopes to change
that by deregulating hemp and legally
decoupling it from cannabis. Meanwhile,
researchers at Texas AM are breeding new
has pursued similar coating-based improve-
ments, focusing on alkaline electrolyzers.
Meanwhile, American company TFP Hydrogen
has announced plans to scale up its electro-
lyzer coating capacity threefold over the next
year, up to 600 megawatts annually. Cana-
da-based Loopflow has developed a new fuel
cell design with a unique geometry and flow
field properties that increase the stability of
internal conditions and efficiency.
New Base Materials
Currently, hydrogen is typically extracted
from treated freshwater. As issues of water
scarcity continue to grow, hydrogen power
companies have sought to harness alter-
native sources. Efforts in Guam, the Nether-
lands, and Australia have focused on using
seawater, both treated and untreated, for hy-
drogen generation. A large EU pilot program
will explore using wastewater as a base ma-
terial. Other efforts are looking into an even
more direct method: accessing “geologic hy-
drogen,” or sources of pure hydrogen trapped
in the earth, in gaseous or other forms.
CLEAN FUELS
ENERGY CLIMATE
Hemp’s high energy density makes it a great source
for biofuel.

TECH
strains of hemp that are both suitable for
the state’s dryer climates and fall within THC
compliant ranges.
Biofuels From Algae
Through photosynthesis, certain kinds of
algae can produce biofuel. As added bene-
fits, they absorb carbon dioxide and grow
organically. In spite of this, scaling biofuel
extraction from algae is slow and expen-
sive. Researchers in Brazil have conducted
experiments with microalgae in which they
“stress” the culture to induce it to make
more lipids, the key component in biofuel
production. Other experiments have focused
on genetically modifying algae to optimize
their efficiency and survivability characteris-
tics. United Airlines has invested $5 million
in biofuel startup Viridos, which focuses on
producing sustainable aviation fuel from
algae in seawater.
Syngas From the Sun
Synthesis gas, or syngas, is a mixture of
hydrogen and carbon monoxide that can be
used as fuel and in the production of meth-
anol. Researchers at the University of Cam-
bridge have devised a solar-powered reactor
that extracts carbon dioxide from industrial
waste, or even from the atmosphere, and
converts it into syngas using what they de-
scribe as “artificial leaves.” By infusing ceria
(a common material used in syngas produc-
tion) with a nickel catalyst, researchers at
the University of Florida have demonstrated
the ability to create syngas at lower tempera-
tures—700 degrees Celsius instead of 1,000
degrees—a reduction that provides economic
benefit to producers.
CLEAN FUELS
ENERGY CLIMATE
Scientists are exposing algae to stressors to increase lipid production, the key component for biofuels.

TECH
Rededication of Contaminated Land
Abandoned coal mines could be the solution
to warming homes in Europe, where homes
across the continent are sitting on top of old
mines filled with warm water. The water gets
warmer the deeper it goes and can be brought
up through boreholes, run through heat
pumps and extractors to increase the tem-
perature even more, and then sent through
heating networks to warm homes. Once the
water is used, it can be redirected back into
the mines where it is heated again. The UK
is currently exploring this project beginning
with 12 preliminary boreholes in Glasgow.
Reducing the Carbon Footprint of Fossil Fuels
In many countries, regulators and research-
ers are attempting new ways to reduce fossil
fuels’ impact on the climate crisis. The US
Environment Protection Agency has proposed
new regulations on limits for pollutants from
American coal and gas power plants, forcing
facilities to track and report emissions. Ko-
rean companies have successfully operated
a gas turbine with a cleaner 60% hydrogen
US. Oil and gas companies will pay a fee if
they emit more than 25,000 tons of carbon
dioxide equivalent per year into the atmo-
sphere, thus spurring innovations in meth-
ane emission monitoring to avoid the fee.
A team of Princeton University researchers
has developed extremely agile drones with
remote-sensing lasers to detect gas leaks
up to 25 times smaller than can be detected
using traditional methods. The drones only
require a small mirror and laser and can
be outfitted to measure other gases such
as carbon dioxide and ammonia as well as
methane. Swiss-based Distran, an innovator
in ultrasonic camera technology to detect
gas leaks, has recently closed $8.3 million
in funding to diversify their products and
tap into new markets to expand the safety of
industrial plants and reduce their environ-
mental impact.
blend fuel. Japan is turning coal into “clean
hydrogen” through the Hydrogen Energy Sup-
ply Chain project, which uses carbon capture
and storage technology and has attracted
$2.35 billion in investment. Botswana is
working with South African-based Sasol’s ex-
tensive coal resources to produce synthetic
fuels to support Europe’s energy crisis. And
in the UK, Net Zero Teesside Power is on track
to be the world’s first commercial-scale
gas-fired power station with carbon capture.
The project will drive the UK government to-
ward its decarbonization goals and produce
enough electricity to power 1.3 million homes
per year.
Methane Emissions Reductions
Methane, a potent greenhouse gas, plays
a major role in trapping heat in the Earth’s
atmosphere. Recognizing its importance
as a driver of climate change, regulators
and scientists are turning their attention
to finding ways to reduce methane emis-
sions. The Inflation Reduction Act contains
the first ever greenhouse gas fee in the
FOSSIL FUEL INNOVATION
ENERGY CLIMATE
Abandoned coal mines could support warming
homes in Europe.

TECH
Fusion
Long considered the “holy grail of energy
production,” nuclear fusion technology would
allow humans to emulate the atomic process
that powers the stars. Most experimental fu-
sion reactors today seek to fuse hydrogen at-
oms into helium, the same reaction our own
sun has been conducting for billions of years.
This reaction creates substantial amounts
of energy in the form of extremely fast mov-
ing atomic particles, which can be converted
to heat and then used to power generators.
The only byproducts are heavier (potentially
useful) elements and trace amounts of ra-
diation at harmless levels. Given the cosmic
abundance of hydrogen, nuclear fusion would
provide a virtually limitless source of energy.
Fusion Momentum
Following the Lawrence Livermore Nation-
al Laboratory’s landmark achievement in
2022, when researchers created the first
human-controlled, net-positive energy-pro-
ducing fusion reaction, conversations around
the technology have reignited, capturing the
German startup Proxima Fusion has raised
$8.6 million to pursue a twisting, new reac-
tor design, one drastically different from the
widely used, toroidal tokamak design.
Molten Salt Reactors
Although the idea of using molten salt as
a coolant for nuclear fission reactors has
been around since the 1950s, it was more or
less abandoned as most commercialized
nuclear installations opted for water cooling
instead. Now, companies and researchers
are revisiting the concept, seeing it as a way
to increase efficiency (by making it easier
to extract thermal energy), generate addi-
tional fissile material (which can then be
used in subsequent reactions), and reduce
radioactive waste. Researchers at MIT plan to
explore the practicality of molten salt reac-
tors while companies such as Kairos Power
and TerraPower are developing commercial
models.
Laser-Driven Fusion
Laser-driven fusion, also known as “inertial
confinement fusion,” is a means of triggering
fusion reactions by firing high energy laser
volleys at small pellets of deuterium and triti-
um to increase temperature and pressure. Fol-
lowing the landmark net positive fusion reac-
tion from Lawrence Livermore in 2022, which
utilized laser-driven ignition techniques,
additional funding is finding its way into
the space. Marvel Fusion, a German startup,
plans to build a $150 million high-power laser
and fusion research facility at Colorado State
University. With an expected completion date
of 2026, the facility will be the most techno-
logically capable research site for laser fusion
energy and high-energy density physics.
Small Modular Reactors
Early in 2023, the US Nuclear Regulatory Com-
mission certified the first design for a small
modular reactor capable of nuclear fission,
meaning utilities can now select it when
building a new power plant. This could repre-
sent a major opportunity for new nuclear proj-
attention of both private and public sectors.
Lawrence Livermore has already repeated—
and improved upon—its initial experiment.
Startups funded by tech millionaires have
begun to enter the field. However, large ob-
stacles remain—namely the massive amount
of investment and lengthy timelines needed
to develop the technology into something
meaningful. Even the most generous esti-
mates place large scale nuclear fusion over
30 years away, a sobering reality check given
the planet’s immediate need for sustainable
sources of energy.
Privately led Fusion Projects
The US Department of Energy injected
capital into a growing ecosystem of private
fusion projects, by distributing $46 million
in funding to eight companies. Incumbent
players, such as TAE Technologies, continue
to iterate and produce new reactor designs
while newer ventures, such as the Sam Alt-
man-backed Helion, aim to bring a Silicon
Valley-esque sense of disruption to the
industry by targeting small scale reactors.
NEW NUCLEAR
ENERGY CLIMATE

NEW NUCLEAR
TECH
ects, especially as a complementary solution
to less reliable emissions-free generation,
such as solar and wind. Conventional nuclear
power plants are bespoke, heavily site-de-
pendent, and must be constructed on-site.
These small modular reactors can be factory
made and integrate much more easily into
existing grids. The approved design is rated
at 50 megawatt output though the company
behind it, Nuscale, is hoping to get higher
capacity models approved soon.
ENERGY CLIMATE
Extraordinary sums of money continue to flow into nuclear fusion, however, timing for scaling remains elusive.

TECH
Energy From Thin Air
New technology is being developed to create
clean energy from the materials in the Earth’s
atmosphere. At KU Leuven University, Belgian
researchers have created solar panels that
produce hydrogen from the water vapor in the
atmosphere and convert it into electricity via
a hydrogen fuel cell. The hydrogen panels will
be on the market through the startup Solhyd;
they can produce up to 250 liters of hydrogen
a day, and the gas can be stored in a tank
to use later. At the University of Massachu-
setts Amherst, researchers have developed a
device that harnesses the same technology
as lightning, by using electricity generated
from water droplets passing through the
generator’s porous material. The “air gener-
ator,” or Air-gen, can make clean electricity
almost anywhere and runs only on ambient
humidity, providing a completely sustainable
energy source. And in Australia, an enzyme
has been found by scientists that can do
something similar, by turning air into ener-
gy. The enzyme, commonly found in soil, can
create electrical currents using low amounts
of hydrogen in the atmosphere.
aquaculture, and renewable energy resourc-
es to maritime communities.
Tidal Turbines
Underwater, companies are turning the
movements of the tide into electricity. Scot-
land’s MeyGen tidal power project, an array
of four underwater turbines about 66 feet
below the ocean’s surface, has generated 50
gigawatt hours of electricity as of February
2023, the first of its kind to do so. This is
roughly equivalent to the annual electrici-
ty consumption of 4,700 US homes. Off the
coast of Eastport, Maine, Ocean Renewable
Power Co. is beginning to test turbine gener-
ators that could be commercially viable. The
turbine will have a 12 month test period, and
then the company plans to implement a full
scale four turbine system further along the
coast. There are also many other tidal energy
startups that have secured substantial fund-
ing, showing a trend in attention returning
to tidal power, including Sustainable Marine
Energy, which raised $11.3 million; Orbital
Marine Power, which raised $11.4 million; and
Verdant Power, which raised $8.73 million.
Mimicking Photosynthesis
Researchers at the University of Cambridge
have made a groundbreaking discovery in the
photosynthesis process that could change
how we generate renewable energy and clean
fuel. Photosynthesis powers a majority of
life on Earth, and it was previously believed
that the chemicals that can extract electrons
occurred later in the photosynthesis process.
However, researchers discovered it takes place
much earlier and pathways exist to move
electrons that were previously unknown. This
new discovery opens up new ways of using
power and the ability to mimic photosynthesis
to create clean fuels from water and sunlight.
This ability to regulate photosynthesis could
also increase crops’ ability to tolerate sunlight
in a world facing climate change.
Wave Power
Wave power, or using wave energy to cre-
ate electricity, is a growing industry that is
expected to increase 4.7% to $30.44 million
by 2030. In Australia, Wave Swell Energy has
done this with floating devices. The com-
pany uses the concept of oscillating water
columns to power its turbines: As waves
push water up and down into a hole at the
bottom of the floating devices, the air in the
space is pushed out of the device and in
turn spins a turbine that creates electricity.
Closer to shore, Eco Wave Power uses“float-
ers” that can be attached to existing man-
made structures. These floaters rise and fall,
moving a piston, accumulator, and hydraulic
motor to create power in a generator. The
company’s floaters were added at the Port of
Los Angeles and are currently being scaled
up; they could generate up to 69% of Cali-
fornia’s electricity production if fully scaled.
And on the support side, AWS Ocean Energy,
who is currently working with Wave Energy
Scotland, is providing technology and ser-
vices for those in the marine energy industry.
The company services oilfield infrastructure,
EMERGING FORMS OF ENERGY PRODUCTION
ENERGY CLIMATE

TECH
Local Sourcing
Chinese mines account for about 60% of
the global supply of germanium and 80% of
gallium—minerals used in everything from
smartphones to electric car batteries. Last
year, in what appeared to be retaliation for
American technology trade limitations, China
enacted heavy export restrictions on these
materials, sending many nations scrambling
to find alternate sources. These actions have
raised concerns that restrictions on rare
earths, which China also dominates (70% of
global production), might not be far behind.
As a result, in recent years Australia, Canada,
the US, and the EU have all invested in finding
domestic sources of rare earths. For exam-
ple, mining company Northern Minerals has
invested $80 million in the exploration of the
Browns Range in Western Australia and plans
to use the site to add dysprosium and terbi-
um to its current offerings of neodymium and
praseodymium. The demand for rare earths
has also kicked off a heated debate over deep
sea mining and which countries can lay claim
to the rich resources on the ocean floor; how-
ever, international law does not yet permit
rare earths. Vitesco Technologies, an electric
vehicle parts manufacturer, has developed
an “externally excited synchronous ma-
chine” (ESM) motor design that can replace
the permanent magnet synchronous motor
(PSM) design that currently dominates the
EV market. In high-performance vehicles,
which often require larger or more powerful
magnets, the ESM design offers an econom-
ic advantage over the PSM design due to its
reduced dependence on costly rare earth ele-
ments. Tesla has likewise announced a pow-
ertrain design that reduces the company’s
dependence on rare earths, citing economic,
stability, and environmental motivations.
such large scale operations, and the environ-
mental impact is not well understood.
Alternative Materials
As another strategy to decrease dependen-
cy on rare earths, researchers have begun
developing alternative materials to replace
them. A number of Japanese companies
claim to have developed valid substitutes
in the magnet space. Proterial has produced
a “high performance ferrite magnet” made
from iron that claims to replace neodymium
magnets in electric vehicle motors. Automo-
tive parts manufacturer Denso has purport-
ed a similar breakthrough, making a magnet
using iron and nickel. Others have developed
a design using iron and samarium. It is per-
haps not surprising that these companies
are all based in Japan; that nation was the
first to feel the brunt of China’s rare earth
trade restrictions in 2010.
Rare Earth Avoidance
Certain manufacturers have attempted to
strengthen their supply chains by reducing,
or completely eradicating, their need for
ALTERNATIVES TO MINERALS AND RARE EARTHS
ENERGY CLIMATE
Countries are looking to diversify their sources for
rare earth and minerals.

SCENARIOS
SCENARIO YEAR 2035
What If the World Runs on Thermal Energy?
In 2035, Thermal Symphony Networks power our world with renewable energy. This technology’s dual capabilities
are leveraging temperature contrast for clean electricity generation and provide heating and cooling with no mov-
ing parts. Initially designed to enhance urban energy production and efficiency where wind and solar were not as
easily integrated, Thermal Symphony Networks quickly transcended their purpose: sidewalks paved with thermo-
electric tiles captured footfall heat to power nearby buildings, and architectural facades lined with thermoelectric
panels balanced indoor temperatures while generating electricity.
As the networks gained traction, industries experienced a paradigm shift. Farmers established “Energy Orchards,”
where temperature-controlled environments enclosed by thermal tiles optimized crop growth while generating
power. Global maritime transport saw the implementation of thermoelectric ship coatings, converting tempera-
ture differences between ocean water and cargo into energy. Electric vehicles equipped with thermoelectric sys-
tems harnessed the heat generated during braking to recharge their batteries, extending their range. Airports
showcased runways embedded with thermoelectric materials, converting the temperature contrast between
asphalt and air into additional power for terminals.
The benefits of Thermal Symphony Networks even extend underwater: The technology’s heat exchange mecha-
nisms promote coral reef health, reversing the damage caused by rising sea temperatures.
TECH
ENERGY CLIMATE

TECH
ENERGY
INFRASTRUCTURE
ENERGY CLIMATE

TECH
TPV Batteries
Given the inherent intermittency of wind and
solar power, electricity generated from these
sources needs a more dependable energy
storage system that would work irrespective
of weather patterns. TPV batteries are one
such solution to bottle and store renewable
energy.
Experiments are being conducted across
the world to find storage alternatives that
are affordable, eco-friendly, and scalable. In
Italy, Enel X and Magaldi Group are building
a 13-megawatt-hour thermal energy storage
using patented fluidized sand bed technol-
ogy. This stores heat that is then used to
release steam at 120-400 degrees Celsius. In
Spain, researchers created a battery that uses
renewable power to melt metals like silicon,
storing latent heat that powers a thermopho-
tovoltaic generator to produce electricity. In
the US, Antora Energy has innovated on TPV
(thermophotovoltaic) technology to convert
carbon block light back into electricity, offer-
ing cost-effective and emissions-free heat
and power innovation. Using wind and solar
technology has also bridged the perfor-
mance gap with materials like nickel and co-
balt. Our Next Energy, a startup, is launching
production of lithium iron phosphate (LFP)
batteries in Michigan. It plans to expand
via a new $1.6 billion plant that will supply
200,000 electric vehicles with LFP batteries
by 2027. Among the larger companies, Ford
has chosen to license LFP battery technolo-
gy from China-based CATL to provide more
cost-effective options for customers, while
General Motors is also exploring using LFPs
to cut costs. Delta has introduced an outdoor
lithium-iron battery system meticulously
tailored for megawatt-level energy storage
applications, addressing the pressing re-
quirements for grid ancillary services, solar
plus storage, and backup power assurance.
Solid State Batteries
Introducing a groundbreaking leap in mak-
ing battery-powered flight a reality, NASA’s
Solid-state Architecture Batteries for En-
hanced Rechargeability and Safety (SABERS)
team has unveiled a sulfur selenium pro-
totype battery with a high energy density of
500 watt-hours per kilogram. With a storage
capacity twice that of traditional lithium-ion
batteries, these batteries offer the rapid dis-
charge capabilities necessary to be used in
aircraft. Toyota has set its sights on launching
solid-state batteries in its mass-produced
EVs by 2027. The company has discovered a
new way to simplify the production of sol-
id-state batteries and enable shorter charge
times with an extended driving range. Nio, an
EV company, in an update to its user manu-
als, has introduced a new 150 kilowatt-hour
semi-solid state battery pack alongside its
existing battery lineup. The Nio ES6 model in
China will incorporate these solid-state bat-
teries. Factorial Energy, a solid-state battery
developer, has forged collaborative partner-
ships with Hyundai, Kia, and Mercedes-Benz
to implement its vision of scaling solid-state
technology by 2026. The company has expand-
ed beyond the US to South Korea, Japan, and
Germany. Furthermore, the realm of 3D-print-
ed solid-state batteries holds transformative
potential, promising enhanced energy storage
solutions for various applications.
energy to heat solid carbon blocks over 1,500
degrees Celsius, Antora can also provide its
technology to industrial processes requiring
high levels of heat. In line with the idea of
ensuring high-temperature stability, a novel
nanophotonic material—embodying struc-
tural variations on a scale comparable to the
wavelength of light—has shown promising
potential for efficient production and control
of thermal radiation. Scientists at the Law-
rence Berkeley National Lab and Scripps Re-
search are also exploring new ways to handle
high energy levels in extreme conditions.
Iron Batteries
With its abundance, iron continues to be
a reliable source for energy storage, and
companies are devising new ways to use iron
batteries to do this on a large scale. Form En-
ergy, a startup, will supply iron-air batteries
to Xcel Energy’s 1-gigawatt-hour project that
is slated for operation in 2025. Lithium-iron
batteries are also gaining popularity. While
their environmental and geopolitical signif-
icance has always been central to growth,
ENERGY STORAGE
ENERGY CLIMATE

TECH
Gravitational Energy Storage
In exploring innovative energy storage solu-
tions, gravitational energy is gaining trac-
tion as a way to create kinetic energy. Swiss
company Energy Vault is nearing completion
of gravity battery installations in the US and
China that are projected to produce 36 and
100 megawatts, respectively. Australian start-
up Green Gravity has joined forces with min-
ing contractor RUC to fast-track the adoption
of its gravitational energy storage technology.
Their technology moves ultra-heavy weights
in mine shafts to turn turbines and create
electricity. Meanwhile, UK-based Gravitricity
is set to initiate pilot demonstrations of its
gravity energy storage systems in India, with
plans for broader deployment in the future.
Flow Batteries
An emerging idea, flow batteries leverage the
flow of special liquids to generate electric
current. Pacific Northwest National Laboratory
is using β-cyclodextrin, a basic sugar extract-
ed from starch, in a flow battery formulation.
This innovative sugar can be synthesized
pled with an impressive life cycle of 50,000
cycles and higher safety standards. Addition-
ally, a research group at Japan’s Osaka Met-
ropolitan University has developed a highly
deformable solid electrolyte. Being touted
as the world’s first bulk-type all-solid-state
capacitor, this capacitor can function at
high current densities and promises higher
efficiency and performance.
Compressed Air Storage (CAES)
These systems use high pressure air to spin
turbines and generate electricity. Current-
ly, only two commercial CAES plants exist
globally: the Huntorf plant in Germany and
the McIntosh plant in Alabama. They utilize
diabatic processes, where off-peak electrici-
ty compresses air for storage, later mixing it
with natural gas for combustion during peak
demand. Hydrostor, a Toronto-based develop-
er, has devised an innovative plan scheduled
for completion in 2028; it involves drilling
three deep shafts that are about 100 yards
high and as long as two football fields. The
company will use excess renewable energy
to compress air into them, later releasing
high-pressure air to generate power. Stan-
ford University researchers have developed
a model to gauge the required compressed
air storage for deep decarbonization of power
systems. Testing their model on California’s
energy grid, the group has highlighted the
cost-effectiveness of compressed air storage
on a dollars-per-kilowatt-hour basis as com-
pared with other sources of energy.
within laboratory settings, offering a sus-
tainable and environmentally friendly alter-
native to currently used materials. The US
Army collaborated with Lockheed Martin to
test a new flow battery that aims to store en-
ergy for a longer duration and at scale. Called
GridStar Flow, this rechargeable flow battery
will use engineered electrolytes to charge
itself. Energy Dome, an Italian startup, is us-
ing “CO2 Battery” to store energy. With fresh
funding secured, the tech company plans to
operationalize two standard 20MW-200MWh
frames by the close of 2024.
Capacitors
Supercapacitors store and release electro-
chemical energy using a flow of electrons
between two conductive plates separated by
an electrolyte. Skeleton Technologies, an Es-
tonian company working on energy storage
technology, has introduced the SuperBattery,
which combines the attributes of super-
capacitors and batteries. This innovation
boasts an extraordinary charging speed 100
times faster than lithium-ion batteries, cou-
ENERGY STORAGE
ENERGY CLIMATE

TECH
UHV Power Lines
Ultra high voltage (UHV) power lines can
efficiently carry electricity over long distances
with minimal loss, enabling energy optimi-
zation across vast areas. While China leads
UHV use, other countries and alliances across
the world are exploring projects to use this
technology. One example is the North Sea
Wind Power Hub, a collaboration from nine
European countries to generate 120 gigawatts
of wind power by 2030 and 300GW by 2050.
The project aims to support Europe’s transi-
tion to a low-carbon energy system through
wind power and uses UHV transmission lines
to interconnect the electricity grids of these
different countries. Beyond Europe, India, Bra-
zil, and Russia are also exploring UHV imple-
mentation.
Superconductors
Superconductors can unlock high-speed en-
ergy transportation without resistive loss. In
a feat first hailed as groundbreaking, scien-
tists at the University of Rochester seemed to
have crafted a superconductor from nitro-
gen-doped lutetium hydride that was able to
transmit electricity at low temperatures and
pressure. However, the experiment couldn’t
be replicated. If successful, superconductors
could scale technologies such as levitating
high-speed trains and make long-distance
energy transport (including wind and solar
energy) a feasible alternative. The SCARLET
initiative, supported by the EU, unites 15
partners from seven countries to develop
superconducting cables, enhancing cost-ef-
fective and efficient power transmission of
renewable energy along with hydrogen in the
same pipeline. The project took off in Sep-
tember 2022 at the Institute for Advanced
Sustainability Studies in Potsdam, Germany.
Focused on further improving long-distance
energy transmission through supercon-
ductors, SuperNode, an Irish renewables
technology company, has partnered with
CERN. CERN, a leader in superconductivity re-
search, will bring its cryogenic and vacuum
expertise to analyze sample materials and
subsystems.
ENERGY TRANSPORT
ENERGY CLIMATE
Ultra high voltage power lines are needed to effectively distribute energy from renewable resources.

TECH
Dynamic Line Rating (DLR) Systems
With climate change, energy grids across the
globe are under more pressure, and grid oper-
ators are increasingly using new technologies
like dynamic line rating (DLR) systems to
enhance the grid. Using sensors, the technol-
ogy delivers real-time information on factors
affecting grid performance, such as wind
speed and temperature. In the US, PPL Corp.
has been sending hourly forecasts to PJM
Interconnection, the regional transmission or-
ganization, since 2022. UK’s National Grid has
collaborated with LineVision, the only compa-
ny specializing in noncontact power line mon-
itoring, to deploy sensors and a DLR platform.
This advancement aims to optimize grid
performance by accurately assessing power
line conditions. In Australia, the Renewable
Energy Agency granted Infravision $732,000
to support the development, testing, and trial
of the “Next Generation Line Monitoring Sys-
tem,” thus enhancing electrical transmission
grid performance.
ing based on this data. The project explicitly
highlights the importance of local and re-
gional decision-making in decarbonization.
In Ireland, the transmission system operator
EirGrid has partnered with Smart Wires to
implement advanced power flow control
devices that automate and optimize the
grid and increase the use of new renewable
energy.
Balancing the Flow of Power Within the Grid
Balancing the flow of power within the
grid is important to ensure that electricity
demand is met in real time with supply. The
University of Applied Sciences of Western
Switzerland has pioneered an optimization
algorithm that can identify the coordinates
of electric current surges in power grids
without knowing the grid’s overall structure,
thus reducing outage costs. Another grid-en-
hancing technology is the Advanced Power
Flow Control, where devices can rapidly push
or pull power from over or under-utilized
lines within a transmission network. The Na-
tional Grid Electricity Distribution, a project
aimed at decarbonization, has launched the
Planning Regional Infrastructure in a Digital
Environment (PRIDE) project to bring togeth-
er key stakeholders on a unified platform to
analyze data and make decisions on energy
systems. Collaborating with the West Mid-
lands Combined Authority and Advanced
Infrastructure, PRIDE explores the potential
of digital twins to understand the regional
energy demands and enable decision-mak-
THE GRID
ENERGY CLIMATE
As the percentage of renewable energy in the grid
increases, we need smart technologies to ensure
stable distribution and supply.

TECH
ENERGY CLIMATE
EMISSIONS
REMOVAL

TECH
Natural CSS
One method of naturally removing carbon
dioxide from the atmosphere is the mass
farming of seaweed, which sucks the gas
from the air. However, a new study has re-
vealed that the amount of seaweed needed to
make an impact (a million square kilometers)
may be too much for turning this solution
into a widespread strategy for combatting
climate change. Mosses are showing sub-
stantial promise after a new study revealed
that moss-covered soil can store 6.43 billion
metric tons more carbon in the soil beneath
it than soil can on its own. The total poten-
tial amount of carbon dioxide that could
be stored by mosses is about six times the
annual global emissions caused by activities
such as deforestation, urbanization, and min-
ing. A French startup is trying to grow geneti-
cally modified house plants to better filter the
air around them. In Georgia, foresters have
begun planting acres of the first genetically
engineered trees, which are designed to grow
faster and capture more carbon than their
traditional counterparts.
anese utility company Tokyo Gas is investing
in Global Thermostat in pursuit of its net-ze-
ro goal for decarbonization.
CO2 Storage
Denmark made history last March with
Project Greensand, which captured carbon
dioxide at a site in Belgium, transported
the liquid gas to the Danish North Sea and
then injected it into the seabed at a depth
of about 1,800 meters. Project Greensand is
also working with Danish company Resen
Waves, which created buoys that generate
power through the ocean waves and can be
used to monitor the seabed-injected carbon
dioxide, detect leaks, and also work as Wi-Fi
hotspots at sea. The project 44.01, involving
partners Abu Dhabi National Oil Co., Fujairah
Natural Resources Corp., and Abu Dhabi Fu-
ture Energy Co., eliminates carbon dioxide by
turning it into a solid rock formation. It is the
first mineralization project to use seawater
and the first carbon-negative project by a
Middle Eastern energy company.
Direct Air Capture
Directly capturing carbon dioxide from the
atmosphere is expensive and requires a lot
of energy. But because the industry is so
visible and young, it’s attracting new innova-
tors such as ex-Tesla Director Douglas Chan
who founded Climeworks with the goal of
being able to remove carbon dioxide by the
gigaton by 2050. Climeworks has announced
plans to scale up its US operations after suc-
cessfully running several European plants.
California-based HolyGrail uses electrical-
ly powered modular scrubbers to remove
carbon dioxide from the atmosphere and
turn it into mineralized cubes, which can be
stacked on top of each other above ground.
Global Thermostat is at the forefront of car-
bon removal technology and has unveiled its
new Direct Air Capture machine—one of their
largest yet, the unit can remove 1,000 tons a
year. The company provides customers with
the main components and associated de-
sign plans of the machine, so that the actual
building and implementation can be done by
a construction company of their choice. Jap-
CARBON CAPTURE STORAGE
ENERGY CLIMATE
Seaweed captures carbon, but it’s questionable if
there is enough to make an impact.

TECH
CCS-as-a-Service
As the demand for carbon capture and
sequestration services increases, CCS-as-a-
service business models continue to spread.
EnQuest, a UK oil and gas company, secured
carbon storage licenses from North Sea Tran-
sition Authority in the UK’s first-ever carbon
sequestration licenses. The company plans
to ship carbon dioxide in liquid form to the
Sullom Voe Terminal in Shetland, a 1,000-acre
future carbon storage hub, before sending it
via an existing pipeline for injection and per-
manent offshore storage. Due to the flexibility
of shipping carbon dioxide in liquid form, this
should make it easier for EnQuest to service
more isolated carbon emitters that may not
otherwise have access to this type of stor-
age. Ørsted’s Asnæs biomass power station
and Northern Lights JV have signed a carbon
dioxide Transport and Services Agreement
to store 430,000 tons per year of biogenic
carbon dioxide from two Danish power plants.
This is a major milestone for both the Euro-
pean commercial CCS market and Northern
Lights JV. Charm Industrial, a company that
turns agricultural remnants into bio-oil, will
be getting $53 million from carbon-removal
credits to turn its agricultural waste into oil
that can lock up carbon dioxide for a million
years and be stored underground.
CARBON CAPTURE STORAGE
ENERGY CLIMATE
First carbon storage licenses have been secured for the North Sea.

TECH
Carbon-Based Food
Traditional farming methods cause mass
damage to our environment in the forms of
habitat loss and deforestation—using up
land, drying up water sources, and releasing
a third of all greenhouse gas emissions. Solar
Foods wants to change that by making food
directly from carbon dioxide. The company is
aiming to start production in 2024 in the first
commercial-scale factory in Finland. Solar
Foods’ technology involves bacteria that use
hydrogen as their energy source; it creates a
yellow powder that is 70% protein and can re-
place animal sourced proteins, such as eggs
in noodles and pasta. The process will use
200 times less land, 600 times less water,
and emit up to 200 times less carbon dioxide
than traditional protein sources.
Diamonds
Diamonds come with a shady history that
can frequently involve corrupt governments
and exploitative working conditions. Aether is
providing a completely conflict-free option by
making diamonds out of atmospheric carbon
dioxide. The US company captures carbon
Chemical Production
The research team at RWTH Aachen has
developed a new chemical reaction that uses
carbon dioxide in the creation of a chemical
compound called aromatic carboxylic acids.
This compound has a variety of uses from
herbicides to plastics—but most important-
ly, it is used in medicine. In the pharmaceuti-
cal industry, a common type of aromatic car-
boxylic acid is salicylic acid, which is used in
aspirin. The RWTH Aachen team’s discovery
not only opens the door to many new indus-
tries but creates the opportunity to utilize
waste carbon dioxide and more sustainably
turn it into a chemical compound that can
be used globally.
dioxide from the atmosphere with a ther-
mochemical process, purifies it, adds green
hydrogen to create atmospheric methane,
and then puts it in specialized chambers
where diamond material can begin to form.
The diamond material is cut and polished
with traditional methods and can be placed
in a variety of jewelry from engagement rings
to earrings. The diamonds themselves are
carbon-negative.
Hair Care, Soap, and Laundry Detergent
Theoretically, carbon can be sequestered
in household products, and CleanCO2 has
found a way to do that. The company cap-
tures carbon dioxide from building heating
systems using a device called CarbinX and
processes it into potassium carbonate (a
non-toxic pearl ash). This pearl ash is used
in the soap formula to create a sudsy lather
when used, and ends up in everyday products
such as hair care, soaps, and laundry deter-
gents. Founded in Alberta, Canada, the com-
pany has now expanded to the US and Japan
and is deploying its CarbinX units globally.
CARBON UTILIZATION
ENERGY CLIMATE
US Company Aether is manufacturing
carbon-negative diamonds.

TECH
Funding Influx
The rapidly growing climate technology
industry is facing an overall mass increase
in funding initiatives. Microsoft’s Climate
Innovation Fund has pledged to distribute $1
billion by 2024, and Amazon’s Climate Pledge
Fund has pledged $2 billion; these projects
vary from concrete that can trap atmospher-
ic carbon to portable batteries that could
replace diesel-burning generators. Venture
capitalists are investing more than $140 mil-
lion in startups, resulting in unprecedented
funds for climate technology: Persefoni raised
$114 million, Watershed raised $85 million,
SINAI raised $36 million, and Sphera raised
$21 million.
Tracking Platforms
Large, established companies are seeing the
value in carbon tracking software, and many
are creating their own or expanding their
services to cover carbon data. SAP’s Sustain-
ability Data Exchange will allow companies
to securely exchange sustainability data in
a standardized format with suppliers and
sion to decarbonize federal buildings, and
nZero, a carbon management and account-
ing platform, is at the forefront. The company
will track federal buildings’ electricity usage
in real time in an effort to make them net-ze-
ro. The agency, along with the US Department
of Energy, will invest $30 million toward this
net-zero buildings goal. The EU enacted its
Corporate Sustainability Reporting Directive,
which requires companies to report scope
3 (indirect emissions) and double materi-
ality—the implications on the company’s
financial value as well as impact to the
environment and world overall. This directive
also includes stricter rules on corporation’s
social and environmental disclosures.
Individual Consumer Tracking
A variety of new tools are encouraging indi-
vidual consumers to consider carbon foot-
print when making decisions. Reewild, a UK
company, has created an app for consumers,
food brands, and retailers to see the carbon
footprint of a variety of food products. They
simply need to scan an item’s barcode while
using the app, which is still in its beta phase.
While it is only in the UK, the company plans
to scale the app to the EU and the US. Ameri-
can Express and MasterCard have introduced
an emissions tracker for consumers to see
how their actions, habits, and spending be-
haviors impact their personal carbon footprint
based on purchases. Google Flights and Uber
have also rolled out emissions data tools for
individual consumer tracking. Australian bank
Westpac developed a tool with Cogo, a car-
bon footprint management fintech, that will
provide customers with personalized carbon
emissions trackers based on their spending.
The bank hopes the transparency will help
close the knowledge gaps that are preventing
individuals from making more sustainable
decisions. Consumers can also track their
waste using Bintracker, a software company,
that uses QR codes to track waste streams
down to individual tenants and then analyzes
and reports on trends and data down to the
source, day, and composition.
partners in order to quickly reduce carbon
emissions in supply chains. EY has released
the beta version of EY OpsChain ESG on its
EY Blockchain SaaS Platform; the product
will provide verifiable and accurate carbon
dioxide emissions information and allow
clearer tracking of an enterprise’s carbon
footprint, providing companies and regula-
tors with a transparent, trusted platform for
carbon emission and credit traceability. Mi-
crosoft has introduced the Microsoft Cloud
for Sustainability to help organizations
more easily track their emissions and is
continuing to add capabilities and updates
to the platform, including a way to track
indirect carbon emissions that frequently
go untracked. Amazon Web Services has
customizable solutions that use AI, machine
learning, data analytics, and the Internet of
Things to capture, analyze, and manage a
company’s sustainability data.
Government Investment and Regulation
The US General Services Administration’s
Green Proving Ground program is on a mis-
CARBON TRACKING
ENERGY CLIMATE

SCENARIOS
SCENARIO YEAR 2050
What If CO2 Is the Foundation for Consumer Products?
In the year 2050, consumer goods conglomerates such as Walmart and Amazon have utilized the scaling of nan-
otechnology to not only effectively integrate carbon capture, utilization, and storage into their operations but also
to provide the delivery of personalized products at scale. A network of specialized nanobots, which the compa-
nies release into the atmosphere en masse, attract and bind carbon dioxide molecules from the atmosphere and
transport them to central processing units. Within these central processing units, the carbon dioxide is convert-
ed into versatile carbon-based raw materials that can be tailored to various applications. The processing units
receive customer orders within a 30-mile radius in real time and are equipped with hundreds of 3D printers, as
well as a wide range of other raw materials and chemical components. Based on the order of each customer, the
carbon dioxide raw materials are integrated into the appropriate material concoction and printed into everything
from highly specific building components to carbon-based foods and intricate consumer goods.
TECH
ENERGY CLIMATE

TECH
ENERGY CLIMATE
EMISSIONS
REDUCTIONS

TECH
CONSTRUCTION
Carbon Neutral or Negative Building
Materials
From plant-based building materials to more
sustainable concrete, companies are exper-
imenting with new ways to construct build-
ings. Plantd, a sustainable building materials
company, creates strong, moisture resistant
carbon-negative building materials from
fast-growing perennial grass. With $10 million
from recent funding, Plantd’s products will
provide a direct substitute for traditional
home construction materials and lock in 80%
of the atmospheric carbon dioxide the plants
captured in the field. Elsewhere, a new type of
engineered wood that traps carbon dioxide,
strengthens the material for use in construc-
tion—the natural material goes through a
process that makes it carbon dioxide-sorbent
and stronger than its more natural state. The
company MAA’VA is creating eco-concrete,
a sustainable carbon negative construction
material, by transforming nonplastic and
plastic waste. This eco-concrete is adaptable
to both conventional and 3D printing con-
partnership between Eco Material Technolo-
gies—a cement alternatives company—and
Hive 3D—an automated construction com-
pany. The result will be a building process
which will emit 92% fewer emissions and
cost 30%-40% less than a traditional con-
struction project of the same size.
struction, which can create housing for one-
tenth the cost and half the waste.
3D Printed Houses
The use of 3D printing methods to build
housing is expanding rapidly, and the use
of natural materials is expanding with it
thanks to designers Ronald Rael and Virgin-
ia San Fratello. They created Potterware, a
browser-based application that allows for 3D
designing without the need to understand
3D modeling software; it also allows the
use of natural materials such as clay, salt,
mud, sawdust, or Chardonnay grape skins
in designing and printing. In Europe, the
continent’s largest 3D printed building is
being built in Heidelberg, Germany, to house
a data center. The construction is expected
to take 140 working hours of robots apply-
ing layers of concrete, instead of dozens of
human workers. In Round Top, Texas, the
seasonal tourist town is about to see five
vacation rental homes that are being called
the world’s first “near-zero-carbon, 3D-print-
ed homes.” These homes are the result of a
GREEN PROCESSES
ENERGY CLIMATE
Companies are increasingly using materials such
as clay and sawdust for the 3D printing of houses.

TECH
HEAVY INDUSTRIES (Steel, Chemicals)
Steel and Iron
Steel production is responsible for 7%-9%
of the world’s total carbon emissions, and
investments in green steel companies are
ramping up. While more resources are com-
ing into the industry than product rolling out,
that may soon change. Hydrogen is playing
a role in advancing the steel refinement
process: The first large-scale green steel
production plant is being built by H2 Green
Steel in Sweden, aiming to cut greenhouse
gas emissions by 95% during the production
process. Using hydrogen technology instead
of blast furnaces, the plant plans to start
shipping its first commercial batches of
steel by 2025. Across Europe, companies like
France’s GravitHy, Germany’s Thyssenkrupp,
and Spain’s ArcelorMittal are constructing hy-
drogen-based plants. Another advancement
to soon enter the steel industry is the use of
Molten Oxide Electrolysis, a process that uses
electricity to separate oxygen from iron ore,
leaving oxygen instead of carbon dioxide as a
byproduct. Boston Metal hopes to bring this
Physics Laboratory has applied “nanowhis-
kers,” aluminum-based membranes that
attract PFAS contaminants and are designed
with a cost-effective coating. Northwestern
University chemists have found a process
that uses low temperatures and common,
inexpensive reagents can remove two major
types of PFAS chemicals and leave them as
benign products—this powerful solution
from a simple technique could be the key
to removing the chemicals at a large scale.
University of British Columbia has also dis-
covered a scalable solution in the form of a
filter that uses a unique absorbing material
to trap and hold the chemicals.
technology to the steel industry within the
next two to three years. The US has a goal of
a net-zero steel sector and is on its way to
that, claiming to have the cleanest global
emissions footprint due to the steel it pro-
duces being roughly 70% made from recycled
scrap.
Chemicals
As US regulators move to restrict two forever
chemicals in drinking water, companies are
experimenting with the technology to do it.
Also known by their proper name of perfluo-
roalkyl and polyfluoroalkyl substances, PFAS
are in everything from waterproof clothing to
dental floss and can cause cancer, liver dam-
age, fertility problems and more. While the
removal of these chemicals with convention-
al filtering techniques is nearly infeasible
and can be costly, researchers at Fraunhofer
Institute for Interfacial Engineering and Bio-
technology IGB have found a way to use plas-
ma-based technology to reduce the chemi-
cal’s molecular chains and remove them at a
low cost. Johns Hopkins University’s Applied
GREEN PROCESSES
ENERGY CLIMATE
A variety of research institutions have come up
with promising filtering systems to remove forever
chemicals from our ecosystem.

TECH
GREEN PROCESSES
ENERGY CLIMATE
AGRICULTURE
Plants Without Plants
Novella is creating plants without the whole
plant by growing botanical ingredients with
necessary macronutrients through nutrient
cultivation—no plant needed. This will help
address the increasing demand for neces-
sary macronutrients, help overcome supply
chain disruptions, expand plant’s life cycles,
and reduce climate impact. While typically a
majority of a plant could be discarded, just to
get to the specific bioactive compounds, this
avoids that waste. This process costs less,
wastes less, and provides more safe, natural
botanicals without chemical additives than
current processes, rising to meet the global
demand.
CRISPR Edited Crops
CRISPR is a technique that allows for the
editing (such as addition or deletion) of an
organism’s DNA—it’s a different technology
from what’s used in genetically modified or-
ganisms, or GMOs. While many international
governments have strict GMO regulations,
they’ve responded differently to crops that
have genes edited with CRISPR. Under a new
law, the UK will allow for gene-edited crops
to be planted, Canada will not regulate crops
that have gone through changes, and the
European Commission is considering next
to no regulation for genetically modified
crops. This is a dramatic change from the
European Commission’s previous stance,
which put in place an intense and expensive
approval process for such crops. But regula-
tors see a lot of possibilities in gene-edited
crops, including saving many from famine
due to climate change and high demand
for food. Already, gene editing is beginning
to help expand the yield and temperature
resistance of staple foods like rice and cow-
peas to meet demand.
Waste
Organic waste used to be something to be
disposed of, but now opportunities are arising
to utilize this material in new ways. Wasted,
a Vermont-based company, created portable
toilets that transport the human waste it
captures to nutrient recovery facilities that
process the waste and turn it into fertilizer.
Mill Industries is doing something similar but
with household waste. The company offers a
food waste bin and service for $33 a month
where the bin will collect, grind, and dehy-
drate the organic matter. The company then
notifies the owner when the material, in the
consistency of coffee grounds, needs to be
shipped to a processing facility to be turned
into chicken feed.
Fertilizer Innovation
New developments in fertilizer technology
are helping ensure stronger, healthier
plants—and experimenting with new uses
for waste material. EnGeniousAg has re-
ceived a grant of $1 million to create soil
nitrogen sensors that are low cost, provide
instant readouts, and can help farmers
measure nitrogen levels in their crops in a
matter of seconds. Along the same lines,
startup Phospholutions recently won the
Africa AgTech Startup Showcase by show-
ing fertilizer efficiency can be increased to
maintain crop yield by reducing the phos-
phorus in it by 50%. Tracegrow Oy, a Finnish
cleantech startup set to expand operations,
is creating certified organic fertilizer from
alkaline batteries. Taking the used batteries,
extracting micronutrients needed by crops,
and placing them in fertilizer has not only
enhanced crop productivity but also reduced
carbon dioxide emissions compared to tra-
ditional fertilizers.

TECH
DIGITAL EMISSIONS
ENERGY CLIMATE
Reducing Carbon Emissions
The increased interest in machine learning
has also raised questions about the carbon
footprint of developing and using such tech-
nology. Google published the “4Ms”—Model,
Machine, Mechanization, and Map Optimiza-
tion—four practices the company believes can
reduce emissions involved in machine learn-
ing by 1,000 times and energy by 100 times.
Model refers to selecting model archetypes
for machine learning that can produce quality
while reducing computation by 3-10 times.
Machine refers to using processors specifi-
cally for machine learning that can improve
energy efficiency by 2-5 times. Mechanization
refers to using cloud computing in data cen-
ters customized for that type of processing
which use higher efficiency servers resulting
in less emissions. Finally, Map Optimization
refers to allowing customers to select loca-
tions with the cleanest energy, which will
increase demand and thus the growth of such
green data centers, reducing carbon footprint
by 5-10 times.
Measuring Emissions
With Google employing the “4Ms,” the rest of
the tech giants are also stepping up to ad-
dress emissions from devices. Taken togeth-
er, all devices globally linked via the internet
have a similar electric consumption to that
of the entire country of France. Amazon, Meta,
Microsoft, Samsung, and Sky have teamed
up with the Carbon Trust to set an industry
standard for measuring and cutting carbon
emissions from their devices while they are
being used by consumers. The group aims
to find a baseline to report energy efficiency
improvements, apply technology to reduce
energy use of connected devices, and drive
toward its ultimate goal of industry-wide net
zero emissions. Researchers from the Lough-
borough Business School have also created
a new tool that helps businesses determine
the carbon footprint of their data and allows
them to create data strategies that are envi-
ronmentally sustainable.
Google hopes to reduce emissions caused by machine learning by 1000 times through its 4M practice.

TECH
RECYCLING
ENERGY CLIMATE
AI Waste-Sorting Robots
AI’s accuracy is being used to improve the
sorting of recycling in order to cut costs.
After a two year trial, Google’s AI-driven recy-
cling-sorting robots have showcased a high
degree of accuracy. The reinforcement learn-
ing system used on the robots in the study
increased accuracy by providing feedback
through rewards and penalties, and the AI
improved the robots’ decisions to maximize
the amount of rewards received. A UK start-
up, Recycleye, has developed an AI-based
waste-sorting robot, which recently won $17
million in funding. The robot uses “vision”
to sort waste into plastics, aluminum, card-
board, and paper with greater accuracy than
humans. Recycleye also announced a part-
nership with Il Solco, an Italian company
that plans to use the AI-based waste-sorting
robots in that country.
Plastic
The startup Norbite uses the greater wax
moth (Galleria mellonella), which naturally
digests plastic, to get rid of the material. It
also uses the moth’s larvae for a variety of
products, such as protein for animal feed,
and the moth’s feces for biofertilizer. Re-
searchers at Leipzig University have discov-
ered an enzyme that can degrade lightweight
PET plastic packaging. The enzyme is able
to compost this plastic type at a higher
efficiency than previously seen in other en-
zymes, including those discovered in Japan
in 2012 that were dubbed “plastic eaters.”
Enzyme PHL7 was discovered to be the
fastest enzyme to break down PET plastic by
90% in 16 hours, and the byproduct of this
enzyme can be rebuilt into new plastic.
Food
Organic waste in landfills eventually builds
up, breaks down, and produces methane,
a powerful greenhouse gas. The company
Divert recently received $1 billion to increase
the use of microbes to break down this or-
ganic waste; this process is called anaerobic
digestion, and it could help reduce the 40%
of methane released from landfills by avoid-
ing sending the waste there in the first place.
Another method is biohydrogen production,
which turns waste into a renewable, clean
energy source. Many processes can be used
in this production but all use microorgan-
isms to break down organic waste materials
and create hydrogen gas. Water is the only
byproduct.
The greater wax moth digests plastic, the feces can
be used for biofertilizer.

TECH
Cross-Laminated Timber
Cross-Laminated Timber is a minimum of
three layers of timber glued together with
grains that cross. Typically, timber can be-
come structurally unsound if enough force is
applied along its grain, but with cross-lam-
inated timber this is far less of an issue.
The material is lightweight, has increased
strength, and has even shown to have a
higher resilience to seismic forces than tra-
ditional timber. But the product is also a way
to use smaller trees, the types of trees that
are usually removed from forests in order to
make them less prone to fire. Mercer Mass
Timber, a company that specializes in CLT, is
preparing for an increase in demand due to
building codes in Oregon, Washington, and
California now classifying it as a new class of
construction. OPAL Architecture has created
a unique, all-electric home using cross-lam-
inated timber and wood-fiberboard insula-
tion to achieve the goal of creating a home
with the lowest-possible carbon output. The
Maine-based company cited the new timbers
as the only material that would work due to
its structural capacity, dimensional stability,
Mushroom Plastic
New research has focused on the fungus
Fomes fomentarius and the way its multi-
ple layers could soon replace a multitude of
plastic products. The mushroom has three
layers: a hard outer layer, a soft middle layer,
and an inner layer similar to the texture of
wood. These multiple layers can potentially
be used in many different products, from
windshield impact-resistant coating using
the hard outer layer to leather-like materials
using the soft middle layer. Led by a scien-
tist from VTT Technical Research Centre of
Finland, a research team was able to create
prototype headphones using the mush-
room’s structure. The applications for the
different parts of this mushroom are vast,
and the researchers’ hope is that this will
soon be the more sustainable alternative to
many plastics.
Self-Healing Materials
There’s a lot of historical precedent in this
field—ancient Roman concrete structures
have long been known to be self-healing.
The structures contain lime clasts, which
were originally believed to be an unfortunate
byproduct. But reevaluations of the durability
of Roman concrete structures have concluded
that this was an intentional choice: As cracks
formed in concrete, the liquid that reached
these lime clasts would reactivate calcium
sources and create calcium carbonate to refill
the cracks and keep the structure sound. Now,
researchers are experimenting with engineer-
ing different materials to similarly heal them-
selves. While performing experiments on how
cracks form and spread, researchers at New
Mexico’s Sandia National Laboratories saw
cracks in copper and platinum spontaneously
heal. The team believes this self-healing could
be engineered into metals to react this way to
stress and cracks. At North Carolina State Uni-
versity, researchers have created self-healing
composite material that allows its structures
to be repaired in place without having to be
removed from service. There is also research
underway to create self-healing plastic at the
University of Michigan, and scientists have
successfully created self-healing solar pan-
els for use in outer space using the mineral
perovskite.
and carbon-storing capacity. Besides being
strong, the cross-laminated timber panels
provided such air sealing that heating or
cooling the interior space would only require
one-third of the typical amount of energy.
Mycelium
Mycelium, a natural fiber from mushrooms,
can be used as a plastic alternative in items
such as textiles, building materials, pack-
aging, and health care products. Completely
biodegradable, this material can be used to
replace plastics such as polyester, building
polymers, concrete, packing materials, and
threaded wound covers. Mushroom Inc. is fo-
cused on finding new ways to use mycelium
hyphae to reduce toxic plastics and waste
while creating more carbon-neutral prod-
ucts. Researchers at Newcastle University
have been able to grow construction mate-
rials using mycocrete, a composite paste of
mycelium. Using a knitted framework, the
composite is fed and allowed to grow until
it reaches the needed density, and then it is
dried out and used as an eco-friendly alter-
native to plastic, timber, or foam.
GREEN MATERIALS
ENERGY CLIMATE

TECH
GREEN MATERIALS
ENERGY CLIMATE
Lab Grown Leather
Vegan leather is becoming more popular as a
sustainable alternative to traditional leather,
but the material typically uses polyurethane
which is toxic. The material innovation com-
pany Von Holzhausen has created Liquidplant,
a top coating for textiles that can help make
vegan leather more durable without using
harmful chemicals. Other companies are ex-
perimenting with using different materials for
vegan leather, including TômTex’s work with
shrimp shells. Using shrimp shells ground
into flakes and then turned into a liquid, the
company is creating vegan leather for wallets,
mimicking leather and its durability without
any petrochemicals. ProjectEx, a Singapore
lab-grown exotic leather producer, is aiming
to raise $1 million in order to create a sustain-
able, cruelty free exotic leather for the luxury
market. The first leathers are expected within
two years of completed funding with plans of
scaling afterwards. The project is a partner-
ship between designer Adrian Furstenburg
and deep-tech startup Cellivate Technologies.
And 3D Bio-Tissues has successfully grown
leather “skin” in a lab without any need for
animals. The company believes that lab-
grown leather technology could be mar-
ket-ready in five years.
A Singapore project aims to create cruelty free exotic leather.

TECH
Micromobility
Micromobility refers to compact personal
transportation, like bicycles, skates, and
scooters, and companies are making it easier
for people to use these devices for greener
daily commutes. French-based startup At-
mosGear has created the first set of electric
inline skates; they have a range of 20 miles
on a fully charged battery that is stored in a
fanny pack, bringing back the Rollerblading
aesthetic. The skates connect to the battery
via a wire that travels down the back of the
leg and allows the wearer to go up to 20 mph.
Another startup increasing micromobility
options is electric bike and scooter company
Yulu. Its app shows rentable scooters’ avail-
ability in real time. Lime is doing something
similar by focusing on electric scooters and
bike-sharing. The California startup is a lead-
er in the space, and is operating in 250 cities
across 30 countries. It has backing from Uber
and other investors that could help this start-
up expand even more.
Heavy Trucks
Truck platooning, where one driver and truck
are connected to a convoy of autonomous
trucks via a network and drive close behind
each other, could ease passenger traffic on
motorways. This use of autonomous vehicles
would increase fuel efficiency through lesser
air resistance and would save costs for long-
haul trucking. This technology is in the trial
phase in several countries and could soon
see the road, with an expected cost savings
of up to 45% compared to today’s trucks
and drivers. Wireless charging is another
technology that so far has been mainly ex-
plored within freight transport. Electreon, a
company specializing in charging electrical
vehicles in motion, has an agreement with
France to equip a portion of a southwest
Paris motorway with its Wireless Electric
Road System. Currently pursuing similar
projects in Sweden, Germany, Italy, and
the US, the company says its product will
increase power transfer capacity and include
real-time monitoring software for all types
and classes of vehicles. However, for now
the focus remains on commercial EV fleets.
Tesla’s purchase of Wiferion, a German-based
wireless charging company, further signals
that the technology is prone to enter passen-
ger driving sooner rather than later.
Trains
Sun-Ways is making trains green by going be-
yond the locomotives themselves. The compa-
ny is using the spaces between railway tracks
to lay out solar panels “like carpet.” Sun-Ways
estimates that a terawatt-hour of solar energy
per year could be produced from the national
rail network in Switzerland, equivalent to 2%
of the country’s total energy consumption.
Meanwhile, Polish company Nevomo is ex-
ploring hyperloop-inspired technology for the
potential future of rail freight. Nevomo uses
MagRail technology, magnetic propulsion that
could be a traction enhancer. While capacity,
reliability, and frequency are more important
factors than speed for rail freight, this tech-
nology could have a revolutionary impact on
the industry—especially considering it would
allow trains to work on both the company’s
MagRail system and traditional tracks. Nev-
omo also signed a deal with the French rail
Cars
Until now, charging electric vehicles (EV) has
been the primary way of keeping them on the
road. But startup Ample has a battery-swap
system that can swap out an empty EV
battery for a fresh one in 5 minutes, far less
than batteries take to charge. With many
other companies pursuing the idea, battery
swapping could bring the ease of a gas sta-
tion to EVs on the road. For combustion cars,
there may be a new alternative that can keep
them on the road in the EU despite the 2035
deadline to phase out polluting vehicles—
running on e-fuels. E-fuels are synthetic fu-
els made from hydrogen and carbon dioxide,
and can be processed in today’s combustion
engines. While e-fuels do still emit carbon
dioxide when they are burned, the amount
taken out of the atmosphere to produce
the fuel offsets the amount that is created
when the fuel is burnt. Luxury car company
Porsche has also created its own synthetic
fuel made of carbon dioxide and water that
can be used in its current cars. The Porsche
e-fuel is among the many seeing the poten-
tial in the fuel market.
GREEN TRANSPORT
ENERGY CLIMATE

TECH
GREEN TRANSPORT
ENERGY CLIMATE
operator SNCF to evaluate if its technology
can increase train efficiency and capacity.
Air
When it comes to air travel, any part, no
matter how small, can make a big difference.
SWING has its focus on the front flaps of
planes, creating them out of thermoplas-
tic polymers and decreasing their weight
by about 20%. While this is just a first step
with this type of material, SWING hopes that
eventually the polymers could be used to
craft entire aircrafts, reducing emissions by
up to 20%. Autonomous aircrafts could first
see their application with unmanned cargo
planes. Dronamics, a UK-based company,
released of one of the first unmanned car-
go aircrafts at the end of 2023. The aircraft
takes less time to operate, saves money, and
emits less carbon dioxide than traditional
cargo freight, catering to underserved areas
and multiple industries. Startup ZeroAvia has
completed a record test flight in a 19-seat air-
craft half powered by hydrogen fuel cells. With
investment from commercial airlines, it’s now
looking to move into commercial tests.
Ocean
With shipping being such a large part of
global trade, many ships are looking to
reduce their carbon footprint due to new car-
bon efficiency regulations. To do this, many
are looking to wind and developing wind pro-
pulsion technologies to decrease fuel usage.
US food company Cargill announced it will
install two folding sails in its bulk carrier’s
deck, and container shipping group Maersk
installed two 30 meter-high rotor sails on
one of its tanker ships. In an effort to make
fuel cleaner, a viscosity meter has been de-
veloped to test the oil used in ship engines,
which is typically full of impurities. The ultra-
sound device is meant to improve shipboard
monitoring and lower the cost.
Tankers might soon be equipped with sails to make them more fuel efficient.

TECH
ENVIRONMENTAL
MANIPULATION
ENERGY CLIMATE

TECH
EARTH
ENERGY CLIMATE
Rewilding: Animals/Nature
Countries across the world are initiating
rewilding efforts, an ecological restoration
approach that involves restoring natural
ecosystems and habitats by reintroducing
native plant and animal species. A NASA-sup-
ported initiative in Idaho uses remote sens-
ing data to forecast the streams that could
support beavers and monitor the biodiversity
shift once beaver populations are introduced.
In the Netherlands, Marker Wadden is a
1,300-hectare archipelago built with the mud
and sand of the lake that now houses diverse
plants, fish, insects, and breeding birds.
Meanwhile, Scotland, aiming to be the world’s
first rewilded nation, has created countrywide
wild lands and natural corridors. This re-
building of ecosystems in Scotland has been
enabled by land ownership laws that allow a
few to own most of the land. (In contrast, in
Ireland, where land ownership is limited to
several acres, natural woodlands make up
only 1% of the island, compared to 80% long
ago.) In Mexico, tequila company Tromba
strives to rewild and reforest land damaged
by blue agave over-cultivation, targeting
immense financial prospects. At the Nation-
al Black-Footed Ferret Conservation Center
in Colorado, Dr. Della Garelle of the US Fish
and Wildlife Service is spearheading the
revival of America’s endangered ferrets. Over
4,000 genetically similar ferrets have been
released into the wild since 1991; this helps
ecosystem restoration, but the limited gene
pool makes them vulnerable to disease. To
maximize breeding success, biologist Robyn
Bortner picks and matches ferrets based on
their genes. In this context, the San Diego
Frozen Zoo is essential, as it leads the de-ex-
tinction field with the world’s largest living
animal cell bank, enabling DNA collection
and storage for future restoration efforts.
For more on rewilding in cities and municipal-
ities, please see our the Rewilding trend in our
Built Environment book.
1,000 hectares in a decade. Citizen-driven
initiatives are also growing. Designer Kiki
Grammatopoulos has introduced “Rewild the
Run,” featuring bristly running shoe out-
soles that aid plant and seed dispersion in
cities. Agricultural rewilding has also grown,
combining rewilding with agriculture to
maintain food self-sufficiency and promote
sustainable and ethical high-quality meat
production.
Bioengineering
Scientists predict that almost a quarter of
all species on Earth are at risk of being lost
within the next few decades. De-extinction
is a scientific method to bring back and
resurrect extinct or close to extinct spe-
cies. Organizations like nonprofit Revive
Restore and for-profit Colossal Bioscience,
are pioneering these efforts. For instance,
Colossal is using gene editing to revive
the dodo bird, while Revive is focused on
the passenger pigeon. In both cases, the
newly created animal will be a hybrid and
not a replica of the predecessor, qualifying
for patent protection and thus producing
The black footed ferret is one of the species that is
being rewilded in the US.

TECH
SKY
ENERGY CLIMATE
Geoengineering
Geoengineering refers to environmental
manipulation—manipulation of Earth’s
resources including oceans, rivers, soil, and
atmosphere. Although it’s in an early stage of
development, geoengineering technology is
already being commercialized, leading to crit-
icism from scientists across the world. The
regulation around it is also mired in conflict,
with different countries adopting contrasting
approaches. In Mexico, the company Make
Sunsets is sending gas-filled balloons into
the atmosphere and selling “cooling credits”
without much scientific validation. The lack
of regulation is allowing such companies to
function despite low credibility. On the other
hand, the European Union is grappling with
the complexities of this issue and consid-
ering potential regulations. Meanwhile, in
China, geoengineering is being considered as
a tool for hybrid warfare strategy. Especially in
the realm of regional geopolitics, China could
amplify its gray zone capabilities and use
weather modification for military purposes,
obstructing river water flow, and even creat-
ing artificial islands.
ing up with innovation, MIT scientists have
proposed a unique approach to use a fleet
of “space bubbles” to reflect sunlight from
space instead of injecting particles into the
atmosphere, potentially reducing harmful
effects. Meanwhile, startups are considering
iron particle spraying above the ocean to
combat climate change by breaking down
methane, a natural phenomenon that may
have influenced ice ages. However, scientists
stress that more fundamental research is
needed before considering large-scale imple-
mentations.
Cloud Seeding
Cloud seeding is a decades-old weather
modification technique that enhances pre-
cipitation by dispersing specialized particles
into the atmosphere. With increased focus
on climate change, new and improved cloud
seeding techniques are being developed and
embraced across the world, including in the
US, China, Russia, and parts of the Middle
East. India recently made strides in this
field by embarking on its first cloud-seed-
ing initiative where an aircraft released
chemical powder to stimulate rain in clouds.
Experimentation continues in other countries
as well with diverse seeding materials. The
United Arab Emirates, at the forefront of inno-
vation, has been using cloud seeding for more
than two decades, leading to a 25% annual
increase in optimal precipitation. Researchers
are now employing nanotechnology and har-
nessing algorithms to optimize cloud seeding
conditions, and they are exploring the use of
drones to amplify the efforts. In Texas, Dan
Martin from the Department of Agriculture’s
Agricultural Research Service is investigating
the use of electrically charged particles to trig-
ger cloud condensation. In an experiment, an
aircraft with tanks of water released a spray
of electrically-charged water particles into the
cloud to see its effect on precipitation. The
United Kingdom, meanwhile, is pioneering the
use of electrical pulses as a potential seeding
technique.
Solar Geoengineering
Still in a nascent stage, solar geoengineer-
ing is very controversial. It theorizes that by
reflecting more sunlight into space, global
warming can be controlled, but countries
and international institutions are grap-
pling with how to regulate the space. The
US government, recognizing the need for
further research, has proposed a study of
risks associated with deploying solar geoen-
gineering techniques. The EU, while formally
disqualifying solar geoengineering as a
climate solution, also recognizes the need to
deepen understanding of the implications
associated with the technology. More recent-
ly, the United Nations convened a panel of
climate experts to deliberate international
regulation of the stratosphere, stressing
the need to manage risks associated with
spraying aerosols to reflect sunlight. In the
UK, a first-of-its-kind solar geoengineering
test flight has been conducted, showcasing
ongoing exploration of the concept. As a part
of it, scientists launched a high-altitude
weather balloon into the stratosphere that is
low-cost, controllable and recoverable. Keep-

TECH
OCEAN
ENERGY CLIMATE
Microplastics
Innovative ideas are surfacing in the battle
against microplastic pollution, offering hope
for a cleaner ocean. One such initiative, GoJel-
ly, harnesses the mucus produced by jellyfish
species to craft filters that effectively trap
plastic particles from wastewater, preventing
their entry into the ocean. The project is inno-
vatively using the growing jellyfish population
to curb microplastics. Another breakthrough
comes in the form of sound wave technology.
Recent research reveals that pulsing sound
waves can efficiently dislodge microplastics
from the ocean’s depth including tiny small
specks that may otherwise be easy to miss.
The team experimented with a prototype
comprising sturdy 8 millimeter steel tubes
and a powerful transducer. With the sound
waves, even the small particles vibrated and
accumulated.
Living Breakwater
Living Breakwater is an innovative approach
to coastal protection and ecological resto-
ration that utilizes nature-based solutions. A
remarkable milestone has been achieved in
water. Researchers at University of California,
Santa Barbara have recently been exploring
a proposal to enhance ocean alkalinity for
accelerated carbon sequestration. By enrich-
ing the ocean with minerals and increasing
alkalinity, they aim to stimulate geologic
processes that efficiently remove carbon
dioxide from the atmosphere. Promisingly,
their study reveals that crucial plankton
groups in the marine food chain respond
positively to this treatment, encouraging
further research in this climate change inter-
vention. Despite the positive outlook, con-
cerns have increased about using minerals
such as basalt to increase the alkalinity of
seawater: These methods may disrupt nu-
trient cycles and affect marine ecosystems.
Acknowledging the significance of marine
geoengineering in climate change mitiga-
tion, the Australian Labor government has
taken a proactive step by introducing a bill
to regulate “marine geoengineering” activi-
ties. The proposed legislation seeks to gov-
ern and control interventions in the ocean
environment. Listed marine geoengineering
activities would require permits, ensuring
proper oversight and safeguarding against
ecological consequences.
Gene Editing
Gene editing coral refers to the scientific
practice of using advanced genetic engineer-
ing techniques to modify the DNA of corals.
The goal is to introduce specific changes to its
genetic code, which can enhance its resilience
to environmental stressors, such as rising sea
temperatures, ocean acidification, and dis-
ease. Scientists, including Carnegie Science’s
Amanda Tinoco, have employed genome
editing tools to unlock a pivotal discovery in
coral development. Their research highlights
the significance of a specific gene known as
SLC4γ, which is essential for the growth of
skeletons in young coral colonies. This gene,
exclusive to stony corals, encodes a protein re-
sponsible for transporting bicarbonate across
cellular membranes, a crucial process in coral
skeleton formation. This breakthrough offers
opportunities for further research into coral
resilience and conservation strategies.
the Port of Rotterdam, where 17 Reefy Reef-
Blocks have been successfully installed in
the river Meuse. The project experiments
with innovative nature-based wave barriers
to safeguard the shores from the impact of
large ship waves while preserving and re-
storing regional biodiversity. By reviving the
intertidal environment, the living breakwater
provides a vital sanctuary for migratory fish
species like sturgeons and European eels,
facilitating their journey between the Atlan-
tic Ocean and major European rivers. In New
York, the “Living Breakwaters” project off the
south shore of Staten Island aims to protect
coastal communities and promote ecologi-
cal restoration by creating habitats for ma-
rine life and fostering biodiversity. Encom-
passing a series of eight in-water structures
stretching 2,400 linear feet, the project was
initially slated for completion by the end of
2024 but is currently ahead of schedule.
Ocean Chemistry
Ocean chemistry is a multidisciplinary field
that delves into the chemical processes and
composition of Earth’s biggest bodies of

SCENARIOS
SCENARIO YEAR 2045
What If a Sunshield Leads to a New World Order?
As the effects of climate change became more dire and governing bodies noticed that conventional climate mit-
igation efforts would not be sufficient to prevent catastrophic impacts to their populations, the most impacted
nations launched “Project Celestial Shield.” The core of this initiative involved launching an array of solar reflec-
tors into space, positioned to intercept a portion of the sun’s energy and redirect it away from Earth to counter
rising global temperatures. Initially celebrated as a revolutionary solution, Project Celestial Shield successfully
attenuated the planet’s temperature rise. However, with the reduced influx of solar energy, regions that once relied
on specific temperature patterns faced abrupt shifts in climate. Agricultural cycles even in thriving regions have
been disrupted, leading to decreased crop yields and global food shortages. Ecosystems accustomed to specific
temperature ranges experienced upheaval, resulting in mass migration of species and a ripple effect throughout
food chains.
As atmospheric dynamics adjusted to the altered energy balance, unforeseen weather patterns have emerged.
Rainfall distribution has dramatically shifted, causing both unexpected droughts and deluges in areas previously
not affected by these extreme weather events. Diplomatic tensions are high as the geopolitical power dynamics
fundamentally shifted in response to the newly defined ecological realities, resetting the political stage for the
coming decades.
TECH
ENERGY CLIMATE

TECH
ENERGY CLIMATE
EFFECTS OF
CLIMATE CHANGE

TECH
Emissions
With increasing recognition of greenhouse
gases (GHGs) as a significant driver of
climate change, there is a pressing need to
reduce these emissions. The United Nations’
World Meteorological Organization (WMO)
is taking the lead in ensuring standard-
ized, real-time tracking of GHGs that can be
used to influence policy making. WMO uses
weather prediction and climate analysis to
monitor greenhouse gases in an integrated
framework. Another initiative is the Global
Greenhouse Gas Watch, where WMO is collab-
orating with the global greenhouse gas mon-
itoring community to ensure sustained and
regular monitoring of GHGs. In this context,
methane, a potent greenhouse gas, is import-
ant. Scientists posit that 60% of atmospheric
methane results from human actions. To
better monitor methane emissions, Harvard
researchers, private space companies, activ-
ists, and philanthropists are launching the
MethaneSAT project. Scheduled for 2024, the
satellite will revolutionize methane moni-
toring by providing a comprehensive view of
emissions across vast areas, surpassing ex-
and Huawei have introduced machine-learn-
ing methods capable of predicting weath-
er patterns as accurately as conventional
methods and with much greater speed. How-
ever, given the unpredictability associated
with future weather events, it remains to be
seen if these AI-powered forecasting models
trained on historical weather data can make
proficient predictions. For extreme heat, Goo-
gle is taking a proactive approach by part-
nering with the Global Heat Health Informa-
tion Network to push notifications to users
in affected locations. Addressing flooding
risks, companies like C2S are leveraging
technology to help insurers better under-
stand and underwrite flood risk and monitor
flood events in real time. Using machine
learning, the company analyzes data from
satellites, historical flood maps, and on-the-
ground intelligence to provide insights into
flood extent and impact.
In another initiative to decrease flooding
risks, the National Digital Twin program is
creating an interactive demonstrator tool to
showcase how connected digital twins can
enhance infrastructure resilience to flooding.
Additionally, real-time flood sensors devel-
oped by the FloodSense project at NYU and the
CUNY Advanced Science Research Center aim
to provide hyperlocal street-level flood event
information to stakeholders, including policy-
makers, government agencies, emergency re-
sponse teams, and citizens. In the face of hur-
ricanes, the National Oceanic and Atmospheric
Administration is equipping forecasters with
a new model called the Hurricane Analysis
and Forecast System. The model offers ear-
lier warnings of rapid intensification and
improved predictions of impacts like rainfall
and storm surge, empowering communities to
better prepare for and respond to hurricanes.
Crops
In Africa, scientists are using artificial in-
telligence to predict the yields of vital crops
by harnessing satellite remote sensing and
machine learning. The tool can analyze vast
agricultural landscapes, providing real-time
predictions of crop yields and offering farmers
and policymakers valuable insights into crop
productivity.
isting aerial and ground-based methods. The
core instrument of MethaneSAT is a spec-
trometer, a sophisticated device capable of
precisely detecting methane concentrations.
By monitoring methane emissions at a
global scale with higher speed and efficien-
cy, MethaneSAT can shed light on emissions’
sources, pinpointing areas for targeted cli-
mate mitigation strategies.
Extreme Weather
The increasing frequency and severity of ex-
treme weather events is causing significant
challenges for insurers and communities
worldwide. In states like Florida and Califor-
nia, insurance companies are withdrawing
from offering homeowner insurance due to
the growing risks associated with extreme
weather events like hurricanes and wildfires.
In Germany, insurers are demanding that
house-building cease in flood-prone areas
to mitigate potential damages. More accu-
rate weather prediction could be helpful in
such cases. Recent advancements in artifi-
cial intelligence have shown promise so far:
Researchers from Nvidia, Google DeepMind,
MONITORING CLIMATE CHANGE
ENERGY CLIMATE

TECH
Community-based Efforts
With climate change becoming a global
phenomenon, community movements have
grown. In an attempt to make climate mod-
eling easy, researchers from Northwestern
University and Argonne National Laboratory
have deployed Nvidia’s Jetson-driven Wag-
gle devices worldwide. Waggle is an open-
source platform, empowering scientists and
researchers globally to join climate studies
using edge computing and sensors. The
platform collects hyperlocal climate data, en-
abling communities to access crucial infor-
mation for climate resilience. Parallely, a new
collaborative, is blending indigenous knowl-
edge with earth science to bolster climate re-
silience in vulnerable coastal communities by
bringing indigenous knowledge-holders and
university-trained scientists together. In In-
dia, a crowdsourced initiative brought togeth-
er community scientists and professional
researchers to develop a meteorology-based
landslide prediction system called “Satark”
for India’s Western Ghats mountain range.
With an impressive 76.5% accuracy, Satark
predicts landslides along the southwestern
egy, furthering the goals of green and digital
transitions and contributing to a sustain-
able future.
coast a day in advance, safeguarding lives
and livelihoods.
Digital Twins of Earth
To tackle climate change, researchers are
experimenting with digital twins of Earth as
part of a strategy to combat warming. The
Nvidia Earth-2 initiative’s digital twins could
revolutionize weather predictions, climate
projections, and the development of effec-
tive mitigation and adaptation strategies
using data-driven analysis at the global
scale. Powered by machine learning and
accelerated computing, this groundbreaking
project uses the largest supercomputing
systems to make unprecedented advance-
ments in weather information and climate
emulation. The European Commission’s flag-
ship initiative, DestinE, is equally ambitious.
Using technology to ensure high accuracy,
local detail, access-to-information speed,
and interactivity, this digital model of the
Earth will monitor, simulate, and predict in-
teractions between natural phenomena and
human activities. It aligns with the European
Commission’s Green Deal and Digital Strat-
MONITORING CLIMATE CHANGE
ENERGY CLIMATE
A digital twin of the earth is aiding with strategies
against global warming.

TECH
Floating Cities
The Institute for Economics and Peace pre-
dicts that by 2050, more than 1 billion people
may have to deal with inadequate infrastruc-
ture due to sea-level rise. Floating cities will
thus become a part of our new living reality.
Projects like Oceanix City in Busan, South
Korea, Baltic Sea islands developed by Blue21,
and Maldives Floating City are great early ex-
amples of this. Oceanix, in fact, is built from
biorock with the ability to self-sustain and
self-repair. Another example is Waterbuurt,
or the Water District, in Amsterdam, where
100 floating houses have been built by Dutch
architect Marlies Rohmer on Lake Eimer.
Other Concepts for Cities
Smart Forest City, an ambitious project by
Italian architect Stefano Boeri, was revealed in
2019 near Cancún. This visionary metropolis,
deeply rooted in the region’s Mayan heritage
and culture’s reverence for nature, is designed
to house 7.5 million plants, creating a harmo-
nious blend of urban living and green spaces.
The company behind the project says the
goal is to create a “perfect balance between
contract, and crack. Recently, scientists from
Northwestern University have verified this
phenomenon based on data from sensors
in urban basements, subway tunnels, and
underground parking garages. The study
predicts that underground climate change in
cities poses long-term challenges for urban
structures, and researchers anticipate these
issues will persist for years.
Indonesia’s New Capital
Acknowledging climate change as an essen-
tial issue, Indonesia is building a new cap-
ital—Nusantara—that aims to transcend typi-
cal planned cities. The intent is to fulfill the
Indonesian president’s vision of “a new work
ethic, new mind-set, new green economy.” It
will stand as a green metropolis powered by
renewable energy, free from suffocating traf-
fic jams, offering serene paths for strolling
and biking amid lush greenery. Embodying
adaptation to a warming planet, the high-
tech city, also known as I.K.N., intends to
attract digital nomads and millennials, who
will embrace cryptocurrency to invest in chic
apartments. Nusantara seeks to set a par-
adigm for a sustainable and forward-looking
urban lifestyle.
Domed Cities
Cities with sealed bio-domes that control air
temperature, air composition, and air quality
are also being explored as a potential solution
for cities affected by climate change. Saudi
Arabia’s Prince Mohammed bin Salman has a
new vision for the country’s capital in line with
this idea: it will include a redesigned down-
town with a 400-meter-high cube or “Mukaab.”
The cube will offer its residents immersive
experiences, including changing landscapes
and holographic technology. The project is
called “New Murabba” and is scheduled to be
ready by 2030. The city Riyadh will expand by
19 square kilometers as a part of the project.
The country also has other futuristic projects
underway, including Neom City, which plans to
include robot maids, flying taxis, and a giant
artificial moon and The Line, a huge project
that will house 9 million people.
the amount of green areas and building
footprint.” Equitism is the idea of an eco-
nomic model empowering citizens with land
ownership. Inspired by this idea, billionaire
Marc Lore aims to establish a new American
city called Telosa to “set a global standard for
urban living, expand human potential, and
become a blueprint for future generations.”
While the exact location is yet to be deter-
mined, the city’s design by Bjarke Ingels
Group places pedestrians and bikers at the
forefront, supplemented by a few “slow-mov-
ing autonomous vehicles.” Sustainability
is at the city’s core, with renewable energy,
protected green spaces, and on-site water
recycling. Targeting a population of 5 million
by 2050, Telosa is poised to redefine the par-
adigm of urban development.
Underground Climate Change
Underground climate change is a silent
hazard. In general, cities and urban areas
release and hold heat more than rural areas.
As cities heat up, the ground also traps more
heat, affecting building foundations and, in
some cases, leading the ground to expand,
LIVING IN A NEW REALITY
ENERGY CLIMATE

TECH
ENERGY CLIMATE
CLIMATE
ECONOMY

TECH
Investment From Traditional Banks
Interest in carbon credits and green investing
has increased as companies strive toward
their net-zero emissions goals. Nine global
banks have invested $45 million toward scal-
ing a new platform called Carbonplace that
will be able to handle more transactions of
voluntary credits and make it easier for bank
customers to participate. Carbonplace will
connect credit buyers and sellers through the
banks. The investors, UBS, National Australia
Bank, BNP Paribas, and Itaú Unibanco among
them, have invested $5 million each in the
interest of accelerating corporate climate
change through visible and secure means.
Verification Methodology
With increased interest in carbon credits,
it’s necessary to make sure the methods are
sound to avoid greenwashing. One compa-
ny, Isometric, is planning to do just that by
launching a platform that can vet carbon
removal companies and review new ones.
It will include a list of fully verified carbon
removal companies and the ability for scien-
tists and researchers to view and comment
one that is simply greenwashing. Sylvera, a
UK-based startup, is on a mission to provide
transparency and trust to carbon offset proj-
ects by using data and assigning ratings to
the projects. Its goal is to ensure the projects
with the best credibility receive investments
in order to make a real climate impact. The
company’s recently closed Series B funding
of $57 million will be used for US expansion.
Blockchain Integration
CarbonKerma is a platform that is embrac-
ing blockchain technology and combining it
with a carbon capture marketplace in order
to provide companies with trackable, quality,
measurable carbon credits. The platform’s
listed credits are highly regulated and vetted
through a stringent process. CarbonKerma
offers a deeply needed value to the market:
transparency. Each credit can be tracked,
traded and, once removed from circulation,
retired to never be used again. The visibility
and auditability of this platform provides full
transparency along with reputable carbon
removal companies to ensure the quality of
carbon removal.
Measuring Carbon Sequestration
The New Acre Project, which recently partnered
with Albo Climate and ALUS, is an investment
platform for corporations that is identifying
carbon stocks and sequestration for trees
using a remote sensing-based platform. This
AI-powered product will be directed at four
Canadian provinces: Alberta, Ontario, Quebec,
and Saskatchewan. The participating sites are
private lands managed by ranchers and farm-
ers in the ALUS program. The credits will be
“ecosystem credits” and will allow investment
in any project that involves carbon removal,
biodiversity, and other environmental benefits
to the provinces.
on data provided by new companies, namely
startups, that enter the industry. The Integ-
rity Council for the Voluntary Carbon Market
and the Voluntary Carbon Markets Integri-
ty Initiative are joining together to craft a
market integrity framework. This framework
is built to instill confidence in investors
using the Voluntary Carbon Market for their
carbon credits and ensure top quality of all
companies on the market. Governments are
beginning to create their own frameworks as
well, with the EU creating a Carbon Removal
Certification Framework, which will influ-
ence the rules set in the Voluntary Carbon
Market. Xilva, a Swiss company, is pursuing
a similar verification but specifically for
forest projects. The company provides an
evidence-based assessment of a project’s
impact by considering multiple criteria such
as economic viability, ecological integrity,
and social equity in order to provide a holis-
tic assessment on their platform.
Focus on Asset Management
Investors are frequently unsure if their
credits are going to a verified company or
CARBON CREDITS
ENERGY CLIMATE

TECH
Blue Economy
The term “blue economy,” originally champi-
oned by developing small-island countries
including Fiji, Palau, and the Bahamas, was
coined to describe the benefits of ocean
industries. This includes establishing and
supporting socially equitable ocean spaces
and industries, ensuring they are environ-
mentally sustainable, and that they can have
an economic profit. The US highlighted a
newly launched global initiative at the Our
Ocean Conference, which would total more
than $800 million to protect its ocean and
support developing countries. The initiative
focuses on securing and protecting marine
areas and improving the resilience of coast-
al areas that could be impacted by climate
change. European ports have received fund-
ing from the European Maritime and Fisheries
Fund through the Atlantic Smart Ports Blue
Acceleration Network to support the growth of
new and sustainable business at more than
40 ports to grow the blue economy. The goal is
to reduce the current carbon footprint of the
port and increase the diversity of activities.
Africa’s blue economy is receiving attention
used. In a space that is used by so many, so
frequently, and is home to so much wildlife,
conflicts of interest are bound to arise when
it comes to ocean activities. MSP serves
as a clearly defined line for any disputes to
ensure that all activities remain sustainable.
It is frequently used to detect marine life or
habitats and restrict building to the oceans’
least impactful locations. MSP is frequent-
ly used by Manna Farms, a fish farming
company, to ensure their practices remain
sustainable.
Foundational Data US
The National Oceanic and Atmospheric
Administration is providing free and open-
source ocean and coastal data that pro-
vides a perfect foundation for the future
blue economy. This data is used by a broad
variety of organizations, from federal, state,
and local governments, to academia and
business and especially offshore aquacul-
ture and wind farms. Open Ocean Robotics, a
Canadian company, received $1.75 million to
create an uncrewed, solar-powered surface
vehicle that can capture information via
sensors and cameras, and relay it instantly to
researchers. The company’s robot vehicles can
travel on the surface of the ocean for months
without stopping, producing carbon dioxide,
making noise, or risking oil spills.
Data Platforms
Cognizant, a technology services and con-
sulting company, has partnered with Tidal
to make the company’s ocean information
platform available to the wider aquaculture
market to allow companies to make more
informed decisions. Tidal uses machine learn-
ing, artificial intelligence, underwater percep-
tion, and automation innovations to gather
and analyze data sets. True Ocean, a German
company, is also looking to provide compa-
nies with more maritime information and has
already become a prominent maritime data
platform. The platform offers solutions for
data processing and management, empower-
ing organizations to see the value of the data
they hold by providing a suite of services with
the goal of increasing efficiency and sustain-
able practices, and facilitating more educated
decision-making.
after venture launcher Triggering Exponen-
tial Climate Action awarded $55,000 to
seven startups to make an economic impact
at a local level.
Blue Carbon Offsets
Blue Carbon Offsets are the management of
marshes, mangroves, kelp, and seagrasses
in ocean ecosystems for carbon offsetting.
Indonesia and the World Economic Forum
have signed an agreement aimed at im-
proving the country’s efforts by supporting
a blue carbon roadmap. The blue carbon
credit financing will help ocean conserva-
tion and restoration efforts. Researchers in
China called the “Blue Carbon Catchers” are
a collaboration between Tencent’s Carbon
Neutrality Lab and Xiamen University. The
group is dedicated to understanding the
ocean’s capacity to act as a carbon sink and
the future of its capacity.
Ocean Land Mapping
Marine Spatial Planning (MSP) is the ocean
equivalent of land zoning planning and is
helping to ensure the ocean is sustainably
BLUE ECONOMY
ENERGY CLIMATE

TECH
Fishing
GAIN (Green Aquaculture Intensification in
Europe) is an EU project that has monitors
and sensors located around fish farms and
placed into fish to help track and under-
stand the activities in the farms. Paired with
machine learning and IBM analytics, this
real-time information will help farmers make
more informed decisions when it comes
to protecting fish, as well as detecting and
preventing problems such as fish escapes,
environmental issues, and excess feed. There
is also an increase in transparency in the life
journey of fish, from the farm to the table.
Atea, with the partnership of the Norwegian
Seafood Association and IBM, is bringing this
transparency by using the blockchain to track
Norwegian salmon’s journey including buyers
and trips through customs. They believe that
transparency is the key to sustainability and
engaging society’s awareness.
Desalination
Ocean Oasis, a Norwegian company, has
found a way to use the oceans’ natural wave
power to produce fresh water. The company
uses waves off Norway’s coast to power their
desalinator units and then fresh water is
sent to shore through pipes along the ocean
floor. This method of producing fresh water
is sustainable and creates no emissions.
EVOVE in the UK has developed technology
to make the desalination process of water
easier. Its Direct Lithium Extraction system
removes the lithium from highly salinated
water which increases the ease of the desali-
nation process. The company is aiming to
scale its technology to provide more efficient
desalination processes on a global scale.
BLUE ECONOMY
ENERGY CLIMATE
North Sea waves are aiding Nowergian company Ocean Oasis to desalinate water without causing emissions.

AUTHORS
CONTRIBUTORS
ENERGY CLIMATE
TECH

Co-authors
EMMA MILLER
NICHOLAS IMBRIGLIA
MAHAK CHHAJER
AMY WEBB
Managing Director
MELANIE SUBIN
Creative Director
EMILY CAUFIELD
Editor
ERICA PETERSON
Copy Editor
SARAH JOHNSON
CHERYL COONEY
ENERGY CLIMATE
TECH
392
392
CHRISTINA VON MESSLING
Senior Manager
Christina von Messling leads our European client portfolio and our
Life Sciences practice area at Future Today Institute. She is re-
nowned for her expertise in strategic foresight and an unparalleled
ability to navigate complex industry landscapes. With a career
spanning over two decades, she has guided multinational corpora-
tions through transformative strategies, leveraging his deep understanding of market dynamics and
future trends.
Christina’s diverse experience across sectors enables her to craft visionary scenarios and strate-
gies, driving sustainable growth and innovation for clients worldwide. Having split her time between
Europe, the UK, Brazil, and the US, she combines a global perspective with in-depth, holistic expertise
of the main drivers of change: technology, geopolitical, economic, and societal developments. She
has advised leading pharmaceutical and healthcare companies, as well as beauty, entertainment,
media, and retail brands.
Christina is a coach in the strategic foresight MBA course at the NYU Stern School of Business. She
holds a Masters in Law from the Freie Universität Berlin, where she graduated within the top 10% of
the nation. She works from offices in New York City, Berlin, and London.

TECH
ENERGY CLIMATE
SELECTED
SOURCES

TECH
“3D Printing Cement Alternative
Boasts 92% Fewer Emissions.”
Professional Builder, July 18,
2023. www.probuilder.com/uto-
pia/3d-printing-cement-alterna-
tive-boasts-92-fewer-emissions.
“44.01 Expands Into UAE With Fu-
jairah Pilot—4401.” Accessed July
18, 2023. https://4401.earth/44-
01-expands-into-uae-with-fu-
jairah-pilot/.
“A Strategy to Improve the
Efficiency and Carrier Lifetimes
of All-Inorganic Perovskite Solar
Cells.” https://techxplore.com/
news/2023-04-strategy-efficien-
cy-carrier-lifetimes-all-inorgan-
ic.html.
“About.” FloodNet. Accessed May
16, 2023. www.floodnet.nyc/.
“About Green Proving Ground
(GPG).” Accessed July 18, 2023.
www.gsa.gov/climate-ac-
tion-and-sustainability/
center-for-emerging-build-
ing-technologies/about-green-
proving-ground.
Administrator. “New Study Finds
That 43 000 ‘Excess Deaths’ May
Have Occurred in 2022 from
the Drought in Somalia.” World
Health Organization—Regional
Office for the Eastern Medi-
terranean. www.emro.who.int/
somalia/news/new-study-finds-
that-43000-excess-deaths-may-
have-occurred-in-2022-from-
the-drought-in-somalia.html.
“Air Pollution Could Kill ‘Equiv-
alent of More than 40 People’ a
Day by the next Decade.” BBC
Science Focus Magazine. www.
sciencefocus.com/news/air-pol-
lution-could-kill-equivalent-of-
more-than-40-people-a-day-by-
the-next-decade/.
Allen, Patricia. “Finnish Startup
Tracegrow Bags Fresh Capital
to Turn Batteries into Fertil-
izers.” EU-Startups, August
26, 2022. www.eu-startups.
com/2022/08/finnish-start-
up-tracegrow-bags-fresh-capi-
tal-to-turn-batteries-into-fertil-
izers/.
Andajani, Rezkia Dewi, et al.
“Tracking Supercritical Geother-
mal Fluid Distribution from Con-
tinuous Seismic Monitoring.”
Scientific Reports, vol. 13, no. 1,
May 1, 2023, p. 8370. https://doi.
org/10.1038/s41598-023-35159-8.
“Apple Announces Major
Progress toward Climate Goals
Ahead of Earth Day.” Apple
Newsroom. www.apple.com/
newsroom/2023/04/apple-an-
nounces-major-progress-to-
ward-climate-goals-ahead-of-
earth-day/.
“AspBAN—Transforming the EU
Atlantic Ports into Blue Economy
Hubs.” June 30, 2023. https://
oceans-and-fisheries.ec.europa.
eu/news/aspban-transform-
ing-eu-atlantic-ports-blue-econ-
omy-hubs-2023-03-01_en.
Asscheman, Ed. “AgTech Sensor
Startup EnGeniousAg Wins $1
Million Award from NSF.” Future
Farming, May 12, 2023. www.
futurefarming.com/smart-farm-
ing/tools-data/agtech-sensor-
startup-engeniousag-wins-1-
million-award-from-nsf/.
“Austrian, Georgian Profes-
sionals Working to Imple-
ment Digital Management for
Georgia’s Largest Hydropower
Plant.” Azernews.az, March 28,
2023. www.azernews.az/re-
gion/208037.html.
Avenue, 677 Huntington, et al.
“The Supreme Court Curbed
EPA’s Power to Regulate Carbon
Emissions from Power Plants.
What Comes Next?” News, July
19, 2022. www.hsph.harvard.edu/
news/features/the-supreme-
court-curbed-epas-power-to-reg-
ulate-carbon-emissions-from-
power-plants-what-comes-next/.
Berkowitz, Rachel. “New Com-
pound That Withstands Extreme
Heat and Electricity Could Lead
to Next-Generation Energy
Storage Devices.” Berkeley
Lab News Center, February 15,
2023. https://newscenter.lbl.
gov/2023/02/15/new-com-
pound-could-lead-to-next-gener-
ation-energy-storage-devices/.
Berkowitz, Rachel. “Pinpoint-
ing Problems in Transmission
Power Grids.” Physics, vol. 16,
June 2023, p. s83. APS Physics.
https://doi.org/10.1103/PRXEner-
gy.2.023009.
Blankenship, Grant. “Genetically
Engineered Trees in a Georgia
Forest Mark a First in the Na-
tion.” Georgia Public Broadcast-
ing, April 13, 2023. www.gpb.org/
news/2023/04/13/genetically-
engineered-trees-in-georgia-for-
est-mark-first-in-the-nation.
“‘Blue Carbon Catchers’ Find Na-
ture-Based Solutions from the
Ocean to Offset Carbon Emis-
sions.” Tencent. www.tencent.
com/en-us/articles/2201615.
html.
Burgos, Matthew. “Turbulent
Electrifies Remote Locations
with Low-Maintenance Vor-
tex Turbines.” Designboom.
January 13, 2023. www.design-
boom.com/technology/turbu-
lent-vortex-turbines-hydropow-
er-plants-01-13-2023/.
CABI. “Impacts of Climate
Change on Animals Will
Be Multi-Faceted, Reveals
Study.” https://phys.org/
news/2023-07-impacts-cli-
mate-animals-multi-faceted-re-
veals.html.
“CFTC Division of Enforcement
Creates Two New Task Forces.”
CFTC. www.cftc.gov/PressRoom/
PressReleases/8736-23.
“China’s Dominance of Stra-
tegic Resources.” DW.com,
April 13, 2021. www.dw.com/en/
how-chinas-mines-rule-the-
market-of-critical-raw-materi-
als/a-57148375.
“Climate Tech VC Market Map.”
Pitchbook. Accessed May 30,
2023. https://pitchbook.com/
news/articles/Q4-2022-market-
map-climate-tech-carbon-cap-
ture.
“Climeworks Confirms Expan-
sion of U.S. Team.” https://clime-
works.com/news/expanding-
our-us-team.
“CO2を食べる自販機！？.” Asahi
Soft Drinks, Note.com, May 10,
2023,. https://note.com/asahiin-
ryo100yg/n/n87818f9665b0.
SELECTED SOURCES
ENERGY CLIMATE

TECH
“Cognizant and Tidal Partner on
Ocean Information Platform for
Aquaculture.” Global Seafood
Alliance, June 15, 2023. www.
globalseafood.org/advocate/
cognizant-and-tidal-part-
ner-on-ocean-information-plat-
form-for-aquaculture/.
Colthorpe, Andy. “Solar-Powered
‘Sand-Based Battery’ Thermal
Energy Storage Project Under-
way in Italy.” Energy-Storage.
News, March 31, 2023. www.
energy-storage.news/solar-pow-
ered-sand-based-battery-ther-
mal-energy-storage-project-un-
derway-in-italy/.
“Commercial Concentrated Solar
One Step Closer.” Australian Re-
newable Energy Agency, February
15, 2023. https://arena.gov.au/
blog/commercial-concentrat-
ed-solar-one-step-closer/.
“Commission Outlines European
Hydrogen Bank to Boost Renew-
able Hydrogen.” https://energy.
ec.europa.eu/news/commis-
sion-outlines-european-hydro-
gen-bank-boost-renewable-hy-
drogen-2023-03-16_en.
“Compressed Air Energy Stor-
age.” Accessed July 17, 2023.
www.eseslab.com/ESsensePag-
es/CAES-page.
Constantinoiu, Marina. “Cap-
tured CO2 from Air Can Be Made
into Green Fuel.” ISRAEL21c, No-
vember 20, 2022. www.israel21c.
org/captured-co2-from-the-air-
can-be-made-into-green-plas-
tic/.
“CRISPR/Cas9 Reveals a Key
Gene Involved in the Evolution
of Coral Skeleton Formation.”
https://phys.org/news/2023-06-
crisprcas9-reveals-key-gene-in-
volved.html.
Dasenbrock-Gammon, Nathan,
et al. “Evidence of Near-Ambient
Superconductivity in a N-Doped
Lutetium Hydride.” Nature,
vol. 615, no. 7951, 7951, March
2023, pp. 244–50. https://doi.
org/10.1038/s41586-023-05742-
0.
Destination Earth | Shaping
Europe’s Digital Future. February
15, 2023, https://digital-strategy.
ec.europa.eu/en/policies/desti-
nation-earth.
“Document Shows Nio ES6
With 577-Mile Range, 150-kWh
Solid-State Battery.” InsideEVs.
https://insideevs.com/
news/676131/nio-solid-state-
battery-577-miles-range/.
“E-Fuels: Are These Synthet-
ic Fuels a Good Alternative to
Electric Cars.” Euronews, April
13, 2023. www.euronews.com/
green/2023/04/13/what-are-e-
fuels-and-can-they-really-make-
europes-cars-emissions-free.
Eldridge, David J., et al. “The
Global Contribution of Soil
Mosses to Ecosystem Services.”
Nature Geoscience, vol. 16, no. 5,
May 5, 2023, pp. 430–38. https://
doi.org/10.1038/s41561-023-
01170-x.
Environment Act 2021. King’s
Printer of Acts of Parliament,
www.legislation.gov.uk/ukp-
ga/2021/30/enacted.
“EPA Releases Proposed CO2
Regulations for Power Plants.”
www.dentons.com/en/insights/
alerts/2023/may/16/epa-re-
leases-proposed-co2-regula-
tions-for-power-plants.
“EU Blue Bioeconomy Report Is
Out.” https://oceans-and-fish-
eries.ec.europa.eu/news/
eu-blue-bioeconomy-re-
port-out-2023-01-13_en.
European Chips Act. https://
commission.europa.eu/
strategy-and-policy/priori-
ties-2019-2024/europe-fit-digi-
tal-age/european-chips-act_en.
“European Countries Sign
Declaration to Make North Sea
Wind Power Hub.” Euronews,
April 25, 2023, www.euronews.
com/my-europe/2023/04/24/
european-countries-sign-decla-
ration-to-make-north-sea-wind-
power-hub.
“European Energy Firms
Dominate Landmark Califor-
nia Offshore Wind Auction.”
Offshore, December 8, 2022,
www.offshore-mag.com/renew-
able-energy/article/14286829/
european-energy-firms-domi-
nate-landmark-california-off-
shore-wind-auction.
“European Sovereignty Fund.”
European Commission. https://
ec.europa.eu/commission/
presscorner/detail/en/STATE-
MENT_22_5543.
“‘Fit for 55’: Council Adopts
Regulation on CO2 Emissions
for New Cars and Vans.” www.
consilium.europa.eu/en/press/
press-releases/2023/03/28/
fit-for-55-council-adopts-regula-
tion-on-co2-emissions-for-new-
cars-and-vans/.
“Floating Wind Is Making Great
Strides.” WindEurope, May 12,
2023. https://windeurope.org/
newsroom/news/floating-wind-
is-making-great-strides/.
“‘Forever Chemicals’ Destroyed
by Simple New Method.”
https://news.northwestern.
edu/stories/2022/08/forev-
er-chemicals-destroyed-by-sim-
ple-new-method/.
Fraunhofer-Gesellschaft. “Using
Plasma against Toxic PFAS
Chemicals.” https://phys.org/
news/2023-05-plasma-toxic-
pfas-chemicals.html.
Gain, Vish. “SuperNode Teams
up with CERN to Improve
Energy Transmission.” Silicon
Republic, April 24, 2023, www.
siliconrepublic.com/innovation/
supernode-cern-superconduc-
tivity-cables-energy-transmis-
sion-partnership.
Gately, James A., et al. “Coc-
colithophores and Diatoms
Resilient to Ocean Alkalinity
Enhancement: A Glimpse of
Hope?” Science Advances, vol.
9, no. 24, June 2023. https://doi.
org/10.1126/sciadv.adg6066.
GeoEnergy, Think. “First Com-
mercial High-Temp Logging Tool
Introduced by Thermochem.”
August 25, 2022. www.thinkgeo-
energy.com/first-commercial-
high-temp-logging-tool-intro-
duced-by-thermochem/.
George, Sarah. “Tech Giants
Team up to Tackle Emissions
from Connected Devices.” Edie.
www.edie.net/tech-giants-team-
up-to-tackle-emissions-from-
connected-devices/.
SELECTED SOURCES
ENERGY CLIMATE

TECH
“Germany—Insurers Demand
a Stop to Building New Homes
in Flood-Risk Areas.” FloodList.
https://floodlist.com/europe/
germany-stop-building-houses-
flood-risk-areas.
Gipson, Lillian. “NASA’s Sol-
id-State Battery Research
Exceeds Initial Goals.” NASA,
October 6, 2022. www.nasa.gov/
aeroresearch/nasa-solid-state-
battery-research-exceeds-initial-
goals-draws-interest.
“‘Give Us the Tools,’ Says In-
dustry, as Europe Ramps up
Offshore Wind Targets.” Riviera.
www.rivieramm.com/news-con-
tent-hub/news-content-hub/
give-us-the-tools-says-industry-
as-europe-ramps-up-offshore-
wind-targets-75916.
“Global Greenhouse Gas Mon-
itoring Infrastructure Takes
Shape.” January 31, 2023.
https://public.wmo.int/en/
media/news/global-green-
house-gas-monitoring-infra-
structure-takes-shape.
“Global Greenhouse Gas Watch
(GGGW).” July 14, 2022. https://
public.wmo.int/en/our-man-
date/focus-areas/environment/
greenhouse-gases/global-green-
house-gas-monitoring-infra-
structure.
“Good News About the Carbon
Footprint of Machine Learn-
ing Training.” February 15,
2022. https://ai.googleblog.
com/2022/02/good-news-
about-carbon-footprint-of.html.
“Green Gravity Links with Miner
to Fast Track Gravity Energy
Storage Tech.” PV Magazine
International, July 3, 2023. www.
pv-magazine.com/2023/07/03/
green-gravity-links-with-min-
er-to-fast-track-gravity-energy-
storage-tech/.
House, The White. “FACT SHEET:
President Biden Announces New
Initiatives at COP27 to Strength-
en U.S. Leadership in Tackling
Climate Change.” The White
House, November 11, 2022,
www.whitehouse.gov/
briefing-room/state-
ments-releases/2022/11/11/
fact-sheet-president-biden-
announces-new-initiatives-at-
cop27-to-strengthen-u-s-lead-
ership-in-tackling-climate-
change/.
“How Megacities Could Lead
the Fight against Climate
Change.” MIT Technology Re-
view, www.technologyreview.
com/2021/04/28/1023236/
how-megacities-fight-cli-
mate-change/.
www.facebook.com/anine.
kilian. South Africa: Redstone
Solar Power Park Set for 2023
Commissioning. 19 July 2022,
https://energycapitalpower.
com/redstone-solar-pow-
er-2023-commissioning/.
Hunt, Kara. “Superhot Rock En-
ergy Is Growing, but Needs More
Support to Reach Its Full Po-
tential.” Clean Air Task Force, 15
Nov. 2022, www.catf.us/2022/11/
superhot-rock-energy-growing-
needs-more-support-reach-full-
potential/.
“In a First, Caltech’s Space Solar
Power Demonstrator Wirelessly
Transmits Power in Space.” Cal-
ifornia Institute of Technology,
1 June 2023, www.caltech.edu/
about/news/in-a-first-caltechs-
space-solar-power-demonstra-
tor-wirelessly-transmits-power-
in-space.
In Greening Air Travel, Small
Parts Can Make a Big Differ-
ence | Research and Innova-
tion. https://ec.europa.eu/
research-and-innovation/en/
horizon-magazine/greening-air-
travel-small-parts-can-make-
big-difference.
“Innovative Viscosity Meter
Opens the Way for Smarter
Use of Fuel Oil at Sea | Imperial
News | Imperial College London.”
Imperial News, 5 Jan. 2023, www.
imperial.ac.uk/news/242311/
innovative-viscosity-me-
ter-opens-smarter-fuel/.
International, Power Engineering.
“Frame 7E Gas Turbine Operates
with Hydrogen Blend at 60%.”
Power Engineering International,
14 July 2023,
www.powerengineeringint.com/
hydrogen/frame-7e-gas-turbine-
operates-with-hydrogen-blend-
at-60/.
Italian Startup Raises $60M
in Total to Boost Energy Stor-
age. https://carboncredits.
com/italian-startup-rais-
es-60mto-boost-energy-stor-
age-for-renewables/. Accessed
May 16, 2023.
Jack. “Electreon to Equip First
French Motorway with Wireless
Charging - Electrive.Com.” www.
Electrive.Com/, 15 July 2023,
www.electrive.com/2023/07/14/
electreon-to-equip-first-french-
motorway-with-wireless-
charging/.
Jeffay, John. “Gene-Editing An
Ancient Crop To Help Feed The
World.” NoCamels, 19 Apr. 2023,
https://nocamels.com/2023/04/
gene-editing-an-ancient-crop-
to-help-feed-the-world/.
Johnson, James K., et al. “Re-
moving Forever Chemicals via
Amphiphilic Functionalized
Membranes.” Npj Clean Water,
vol. 5, no. 1, 1, Oct. 2022, pp. 1–6.
www.nature.com, https://doi.
org/10.1038/s41545-022-00193-y.
jostapiuk@siadvance.com,
Joseph Ostapiuk |. “See Rock
Structures Rising off Tottenville?
Project to Blunt Waves Nears
Completion.” Silive, 11 May 2023,
www.silive.com/news/2023/05/
see-rock-structures-rising-off-
staten-island-breakwaters-proj-
ect-nears-completion.html.
Juvenile Alewife Passage
through a Compact Hydropower
Turbine Designed for Fish Safety
- Watson - 2023 - North Ameri-
can Journal of Fisheries Man-
agement - Wiley Online Library.
https://afspubs.onlinelibrary.
wiley.com/doi/full/10.1002/
nafm.10866.
Kareem, Shefy Manayil. “What’s
New with Microsoft Cloud
for Sustainability.” Microsoft
Industry Blogs, 12 Oct. 2022,
www.microsoft.com/en-us/
industry/blog/sustainabili-
ty/2022/10/12/driving-innova-
tion-for-esg-progress-with-mi-
crosoft-cloud-for-sustainability/.
SELECTED SOURCES
ENERGY CLIMATE

TECH
Karthikeyan, Smruthi, et al.
“Wastewater Sequencing Re-
veals Early Cryptic SARS-CoV-2
Variant Transmission.” Nature,
vol. 609, July 2022. Research-
Gate, https://doi.org/10.1038/
s41586-022-05049-6.
Kemper, Gregor, et al. “Pd(II)-Cat-
alyzed Carboxylation of Aromatic
CH Bonds with CO 2.” Science
Advances, vol. 9, no. 5, Feb. 2023,
p. eadf2966. DOI.org (Crossref),
https://doi.org/10.1126/sciadv.
adf2966.
Kraemer, Susan. “Add Nickel to
Ceria for Solar Syngas at Just
700°C.” SolarPACES, 29 Mar.
2023, www.solarpaces.org/add-
nickel-to-ceria-for-solar-syngas-
at-just-700c/.
Kulkarni, Jeevan R., et al.
“‘Satark’: Landslide Prediction
System over Western Ghats
of India.” Land, vol. 11, no. 5, 5,
May 2022, p. 689. www.mdpi.
com, https://doi.org/10.3390/
land11050689.
L, Jennifer. “Revolutionizing
Carbon Credits: ICVCM and VCMI
Team Up to Create High-Integ-
rity Voluntary Carbon Market.”
Carbon Credits, 20 June 2023,
https://carboncredits.com/
vcmi-icvcm-market-integri-
ty-framework/.
---. “WEF and Indonesia Join
Hand to Boost Blue Carbon Cred-
its.” Carbon Credits, 2 Feb. 2023,
https://carboncredits.com/wef-
indonesia-join-hand-to-boost-
blue-carbon-credits/.
“Latent Heat Thermophotovol-
taic Battery for Renewables
Storage.” Pv Magazine Inter-
national, 23 Mar. 2022, www.
pv-magazine.com/2022/03/23/
latent-heat-thermophotovolta-
ic-battery-for-renewables-stor-
age/.
Laws, Liza. “Nutri-Tech Startup,
Novella, Grows Botanical Ingredi-
ents without Using the Plants.”
Labiotech.Eu, 30 Nov. 2022,
www.labiotech.eu/more-news/
novella-grows-botanical-ingredi-
ents-without-using-the-plants/.
Le Page, Michael. “A Factory Will
Soon Start Making Green Food
from Air and Electricity.” New
Scientist, December 28, 2022.
www.newscientist.com/article/
mg25634192-700-a-factory-will-
soon-start-making-green-food-
from-air-and-electricity/.
“‘Least-Cost’ Model for Com-
pressed Air Energy Storage.”
Pv Magazine International, 5
Apr. 2023, www.pv-magazine.
com/2023/04/05/least-cost-
model-for-compressed-air-ener-
gy-storage/.
Liu, Yueliang, et al. “An Integrat-
ed Framework for Geothermal
Energy Storage with CO2 Se-
questration and Utilization.” En-
gineering, March 2023. https://
doi.org/10.1016/j.eng.2022.12.010.
Loria, Rotta, and Alessandro F.
“The Silent Impact of Under-
ground Climate Change on Civil
Infrastructure.” Communica-
tions Engineering, vol. 2, no. 1, 1,
July 2023, pp. 1–12. https://doi.
org/10.1038/s44172-023-00092-1.
Magazine, A. E. C. “Climate Resil-
ience Planning through Digital
Twins.” AEC Magazine, August
4, 2022, https://aecmag.com/
features/climate-resilence-plan-
ning-through-infrastructure-dig-
ital-twins/.
Muniz, Ricardo and FAPESP. Re-
searchers Cultivate Microalgae
for Biofuel Production.
https://phys.org/
news/2023-04-cultivate-mi-
croalgae-biofuel-production.
html. Accessed 14 July 2023.
National Grid Granted Nearly £1m
from Ofgem to Fund Eight Net
Zero Innovation Projects | Nation-
al Grid Group. www.nationalgrid.
com/national-grid-granted-near-
ly-ps1m-ofgem-fund-eight-net-
zero-innovation-projects.
Net Zero Industry Act. https://
single-market-economy.ec.euro-
pa.eu/publications/net-zero-in-
dustry-act_en.
Oka, Daisuke, et al. “Conceptual
Model of Supercritical Geother-
mal System in Shiribeshi Re-
gion, Hokkaido, Japan.” Geother-
mics, vol. 108, February 2023, p.
geothermics.2022.102617.
Pinkus, Arin. “Mapping Climate
Risks by County and Commu-
nity.” American Communities
Project, February 17, 2021.
www.americancommunities.org/
mapping-climate-risks-by-coun-
ty-and-community/.
“U.S. Department of Energy An-
nounces $40 Million as Part of
New ARPA-E Program Focused on
Improving Reliability, Resiliency,
and Security of Power Infrastruc-
ture.” Arpa-e.Energy.Gov, March
30, 2023. http://arpa-e.energy.
gov/news-and-media/press-re-
leases/us-department-energy-
announces-40-million-part-new-
arpa-e-program.
Procedure File: COM(2023)0165
| Legislative Observatory | Euro-
pean Parliament. Accessed July
27, 2023. https://oeil.secure.
europarl.europa.eu/oeil/pop-
ups/ficheprocedure.do?refer-
ence=COM(2023)0165l=en.
“QA on a New Outlook on the
Climate and Security Nexus.”
European Commission. https://
ec.europa.eu/commission/
presscorner/detail/en/qan-
da_23_3525.
“Researchers Develop Coating
That Prevents Synthetic Fabrics
from Shedding Harmful Micro-
plastics in the Wash.” University
of Toronto News. www.utoronto.
ca/news/researchers-devel-
op-coating-prevents-synthet-
ic-fabrics-shedding-harmful-mi-
croplastics-wash.
Richter, P. Konstantin, et al.
“Structure and Function of the
Metagenomic Plastic-Degrad-
ing Polyester Hydrolase PHL7
Bound to Its Product.” Nature
Communications, vol. 14, no. 1,
April 1, 2023, p. 1905. https://doi.
org/10.1038/s41467-023-37415-x.
Salisbury, Mark. “Platooning:
Several Trucks but Only One
Driver.” FleetPoint, July 13, 2022.
https://fleetpoint.org/autono-
mous-vehicles/platooning-sev-
eral-trucks-but-only-one-driver/.
Saravanapavanantham, Ma-
yuran, et al. “Printed Organic
Photovoltaic Modules on Trans-
ferable Ultra-Thin Substrates as
Additive Power Sources.” Small
Methods, vol. 7, no. 1, 2023, p.
2200940.
SELECTED SOURCES
ENERGY CLIMATE

TECH
https://doi.org/10.1002/
smtd.202200940.
“Scientists Have Invented a
Method to Break Down ‘Forever
Chemicals’ in Our Drinking Wa-
ter. Here’s How.” World Econom-
ic Forum, May 19, 2023. www.
weforum.org/agenda/2023/05/
forever-chemicals-pfas-drink-
ing-water/.
“Scottish Startup to Begin Grav-
ity Energy Storage Pilot in India.”
PV Magazine India, November
29, 2022. www.pv-magazine-in-
dia.com/2022/11/29/scot-
tish-startup-to-begin-gravity-en-
ergy-storage-pilot-in-india/.
Service, Purdue News. “A Simpli-
fied Way to Turn Food Waste into
Hydrogen Energy.” www.purdue.
edu/newsroom/releases/2020/
Q1/a-simplified-way-to-turn-
food-waste-into-hydrogen-ener-
gy.html.
Shaw, Al, et al. “New Climate
Maps Show a Transformed Unit-
ed States.” ProPublica, Septem-
ber 15, 2020, https://projects.pro-
publica.org/climate-migration.
Singh, Mrityunjay, et al.
“Comparative Study on Heat
Extraction from Soultz-Sous-
Forêts Geothermal Field Using
Supercritical Carbon Dioxide
and Water as the Working Fluid.”
Energy, vol. 266, March 2023, p.
energy.2022.126388.
“Startup of the Month: How
Estonia’s Skeleton Technologies
Is Changing the Energy Storage
Game Beyond Batteries.” Silicon
Canals. March 31, 2023. https://
siliconcanals.com/news/start-
ups/startup-of-the-month-skel-
eton-technologies/.
Stites, Katy. “CSU Researcher
Part of Indigenous Climate
Resilience Science Hub.” Warner
College of Natural Resources,
November 18, 2022.
https://warnercnr.source.
colostate.edu/csu-research-
er-part-of-indigenous-cli-
mate-resilience-science-hub/.
“Sustainable Cloud Computing.”
Amazon Web Services, https://
aws.amazon.com/sustainability/.
“The Climate Case—Saúl vs.
RWE.” Accessed July 27, 2023.
https://rwe.climatecase.org/en.
“The Dawn Of A New Era In
Superconductor Materials.”
OilPrice.com. https://oilprice.
com/Energy/Energy-General/
The-Dawn-Of-A-New-Era-In-Su-
perconductor-Materials.html.
“The Future of Renewable
Energy: Genetically Modified
Algae Biofuel.” EnergyPortal.eu,
July 12, 2023. www.energyportal.
eu/news/the-future-of-renew-
able-energy-genetically-modi-
fied-algae-biofuel/61227/.
“The United States Releas-
es Signed Memorandum of
Understanding with the Dem-
ocratic Republic of Congo
and Zambia to Strengthen
Electric Vehicle Battery Value
Chain.” United States Depart-
ment of State. www.state.
gov/the-united-states-releas-
es-signed-memorandum-of-un-
derstanding-with-the-democrat-
ic-republic-of-congo-and-zam-
bia-to-strengthen-electric-vehi-
cle-battery-value-chain/.
“Tracking Waste in the Cloud:
Aussie Startup’s Innovative Data
Platform Helps Organisations
Reduce Their Carbon Footprint
and Save on Costs.” Microsoft
Australia News Centre. https://
news.microsoft.com/en-au/
features/tracking-waste-in-the-
cloud-aussie-startups-innova-
tive-data-platform-helps-organ-
isations-reduce-their-carbon-
footprint-and-save-on-costs/.
Trang, Brittany, et al. “Low-Tem-
perature Mineralization of
Perfluorocarboxylic Acids.” Sci-
ence, vol. 377, no. 6608, August
2022, pp. 839–45. https://doi.
org/10.1126/science.abm8868.
Tubeuf, Carlotta, et al. “Increas-
ing the Flexibility of Hydropower
with Reinforcement Learning on
a Digital Twin Platform.” Ener-
gies, vol. 16, no. 4, 4, Jan. 2023, p.
1796. www.mdpi.com, https://doi.
org/10.3390/en16041796.
University, Southern Cross.
“Advanced Imaging Technology
Helps Monitor Baby Corals on
Great Barrier Reef.”
https://phys.org/
news/2023-07-advanced-imag-
ing-technology-baby-corals.html.
“US Startup Begins Producing
40%-Efficient Thermophotovoltaic
Cells.” Pv Magazine International,
January 25, 2023, www.pv-mag-
azine.com/2023/01/25/us-start-
up-begins-producing-40-effi-
cient-thermophotovoltaic-cells/.
Viaintermedia.com. “Biofuels—Ve-
olia Launches Innovative Solution
to Produce CO2-Neutral Biofuel.”
Renewable Energy Magazine,
Apri. 6, 2022. www.renewableen-
ergymagazine.com/biofuels/veo-
lia-launches-an-innovative-indus-
trial-solution-to-20220406.
“Watch ‘Europe’s Largest 3D-Print-
ed Building’ Come into Shape.”
Euronews, May 12, 2023. www.
euronews.com/next/2023/05/11/
only-140-hours-needed-to-put-to-
gether-europes-largest-3d-print-
ed-building.
“Water Battery: Utah Com-
pany Advances Wyoming
Pumped-Storage Project.” www.
sltrib.com/renewable-ener-
gy/2023/01/28/water-bat-
tery-utah-company/.
“World First: Researchers Create
CO2 Measurement Tool to Calcu-
late Emissions Caused by Digital
Data.” Loughborough University,
July 1, 2023. www.lboro.ac.uk/
news-events/news/2023/july/
researchers-create-co2-mea-
surement-tool/.
YARA. “EnerKíte Und Volkswagen
Starten Machbarkeitsstud-
ie Für Mobile E-Tankstelle.”
EnerKíte Flugwindkraftanlagen,
April 7, 2023, https://enerkite.
de/en/enerkite-und-volkswa-
gen-starten-machbarkeitsstud-
ie-fuer-mobile-e-tankstelle/.
“Zukunft für frühere Kohlege-
biete: Es werde grün.” March
3, 2022, www.profil.at/wis-
senschaft/zukunft-fuer-frue-
here-kohlegebiete-es-werde-gru-
en/401918623.
SELECTED SOURCES
ENERGY CLIMATE

MOBILITY • ROBOTICS • DRONES

TECH
TABLE OF CONTENTS
MOBILITY, ROBOTICS DRONES
402 Top Headlines
403 State of Play
404 Key Events
Developments
406 Why Mobility, Robotics
Drones Trends Matter
407 When Will Mobility,
Robotics Drones Trends
Impact Your Organization?
To Consider
410 Central Themes
412 Ones To Watch
413 Important Terms
415 Mobility Trends
416 Electrification Transforms
Mobility Ecosystems
416 Decarbonizing Mobility
416 Automaker Restructuring
424 CarOS
425 Scenario: What if expanded
vehicle capabilities
redefine the nature of home
ownership?
426 Data Collection Enables Safety
and Autonomy
426 Mobility Simulation
426 Self-Aware Vehicles
426 Pilot and Passenger
Observation
427 Mobile Weather Stations
427 Mobility Superapps
427 Utilizing Mobility Data
427 Relying on ADAS
428 Pedestrian Concerns
428 AV Viability
429 Local AV Regulations
429 Robotaxi Growth
430 Scenario: Personal Everything
Mobility Platforms
431 Robotics Drones Trends
432 Cobots Become Coworkers
432 Accelerated Adoption
432 General Purpose Robots
433 Robots in the Home
433 Robots Coexisting with
Creative Applications
434 Space Robots and Drones
435 Robot and Drone
Infrastructure
435 Robot Compiling and Training
435 Robot and Drone Swarms
435 Drone Fleets
436 Drone Traffic Management
437 Scenario: Drone Harvests
438 Moving People, Pets
And Objects
438 Last-Mile Delivery
438 Expanded Payload Capacity
439 Flying Taxis (eVTOLs)
439 Ocean-Faring Drones
440 Blurring The Human-Machine
Line
440 Natural Exoskeleton Movement
440 Superhuman Abilities
440 Personal Mobility
441 Humanoid Robots
441 Soft Robotics/Getting a Grip
442 Robot “Eyes”
443 Scenario: The Evolution of
Disaster Response Robotics
444 Taking Cues From Nature
444 Quadrupedal Robots
444 Necrobotics
444 Using Live Organisms
(Ethically)
444 Insect-Like and Animal-Like
Designs
445 Fluid Movement
446 Moving Across Modalities
447 Scenario: Self-Regulating and
Repairable Robots
448 Authors
416 Mass Market Goes Custom
417 Incentive-Driven Investments
417 Global Battery Belts
417 Battery Recycling
418 Better Batteries
418 Solar Vehicles
418 Shifts in the Servicing Model
419 Electrification Expands to
Other Vehicles
420 Vehicle Charging Scales
420 Charging Gets A Roadmap
420 Charging Standardization
420 Redefining the Roadside
421 Electrifying Cities
421 EVs At Home
422 Bidirectional Charging
423 Immersive Vehicles Connect
to Other Ecosystems
423 Livable Cabins
423 Simulated Driving Experience
423 In-Vehicle Connectivity
424 Mobile Entertainment Hubs

The mobility, robotics,
and drones industries
are building on their
substantial gains
of the past year to
increase capabilities
and capacity.
TOP HEADLINES
01
02
03
04
05
Demand for Customization Outweighs Supply Chain Delays
Even as supply chains bounce back from extreme conditions, customers have indicated
they are willing to wait for what they want, impacting traditional business models.
Vehicles Are No Longer a Private Refuge
While vehicles have traditionally functioned as a utility to get us from point A to point B,
they are now following and integrating into our digital lives, eliminating one of the last
spaces that was not immersed in tech.
Battery Capacity Drives Capability
The expanded capacity and shifting form factors of emerging battery technology are
enabling units to go further, faster and take on a higher compute load to enhance their
capabilities and communication.
General Purpose, Yet Impressive, Results
General-purpose robots are moving closer to reality. As bots become more capable,
they will be more versatile, able to execute on a variety of different tasks.
Cobots Merely Supplement Staff, for Now
Assistive robots promise to help humans be more efficient at their jobs and supplement
a shrinking labor force, but as bots improve, inevitably, they will eliminate the need for
humans to conduct certain tasks.
TECH

STATE
OF PLAY
While the diverse industries of mobility, robotics, and drones have prominent dif-
ferences impacting each domain, they also have common drivers. Electrification
remains a major driver of change for all of them thanks to the support of consumer
and ESG (environmental, social, corporate governance) demands, coupled with glob-
al, legislative shifts bringing an electric future to the forefront.
Fully autonomous systems are still on the horizon, especially for automobiles, but
systems are becoming smarter and more alive. A wealth of data is proving to be
foundational for molding these systems, and virtual and simulated testing environ-
ments are increasingly being used to teach them before they are placed in real-world
environments.
Although these systems are getting smarter, they are still fraught with challenges.
There is a vast dichotomy between those that have been successfully testing and
expanding in select US markets, while others have been forced to halt real-world
operations after too many public failures.
Similarly, in the world of robotics, a general purpose, fully autonomous humanoid
robot has long been a dream of many and presented as a holy grail of innovation.
However, while there are early signals of such robots, they are unlikely to fully super-
sede the human workforce any time soon, with most job replacement coming from
functional robots with a limited task set. Contrastingly, drones have already been
successfully impacted by autonomous systems and are seeing annual capability
enhancements across commercial and consumer sectors.
The bifurcation of these industries will likely continue as the world pushes to bring
about electrified replacements for billions of vehicles and impart autonomy to make
vehicles, robots, and drones smart enough to drive real efficiency gains across the
marketplace.
Overcoming challenges to
electrify and bring about the age
of smarter, autonomous systems.
TECH MOBILITY, ROBOTICS DRONES
403

MAY 25, 2023
Ford Teams Up with Tesla
The automakers align to use
the NACS plug, promoting
industry standardization for
charging.
AUGUST 04, 2023
Ukrainian Drone Attack
A Ukrainian drone with 450kg
TNT hits a Russian Navy base
in the Black Sea, damaging a
docked warship.
SEPTEMBER 19, 2023
AI Masters Complex Tasks in Hours
Toyota Research Institute engineers
use generative AI to teach robots new
skills quickly and with dexterity.
AUGUST 14, 2023
Self-driving Cars Block Ambulance
Two autonomous cars prevent an
ambulance from getting to a crash
site, delaying an injured pedestrian’s
transport.
OCTOBER 30, 2023
Cruise Suspends Robotaxis
The GM business puts a stop to
operations in Arizona and Texas,
after having to halt operations
in San Francisco.
KEY EVENTS

MORE CERTAINTY IN COMPLEX SYSTEMS
In the immediate future, a wave of
transformative development is poised
to reshape our world across various
domains. From the continued integra-
tion of connectivity in automobiles—
fueling advanced driver assistance
systems and enriched in-car expe-
riences—to the pressing challenges
facing our electrical grids as we pivot
toward an all-electric future and the
changing regulatory landscapes im-
pacting drones and autonomous sys-
tems, these developments underscore
the dynamic nature of technological
progress. They collectively signify an
era of both challenge and opportunity,
where adaptability and forward-think-
ing will be key to navigating the dis-
ruptive forces of technology.
Continued Connectedness
Automobiles are only becoming more con-
nected. This will impact advanced driver
assistance systems as well as infotainment
within the cabin. Automobiles will be less
isolating as drivers and passengers seam-
lessly expand how they connect to their lives
outside the car.
Challenges to the Grid
As we idealistically transition to an all-elec-
tric future, many have speculated our elec-
trical grids will not be able to handle this
adjustment. Others are more optimistic. In
time, we will have more certainty.
Supply Chain Disruptions Persist
Even as the supply of vehicles and chips for
robotics stabilizes, manufacturers and sellers
should still brace for continued supply chain
disruptions. Additionally, chip nationalism
and other geopolitical factors will also threat-
en supplies of goods.
Viability of Drone Traffic Management
The escalating use of drones and eVTOLs has
necessitated advanced traffic management
solutions. As various regions have put mea-
sures in place, the viability of ubiquitous drone
use will be determined in the short term.
New Modalities for Robotics
Researchers have developed many innovative
modalities for robotics and drones, seeking
inspiration from various sources. This innova-
tion will continue, with unexpected inspiration
from nature or animals, and some systems
will incorporate several different modalities in
their designs.
Clarity on Robotaxis
Robotaxis have been the source of much con-
troversy, as different companies have tested
their systems with varying success. The near
future will shed light on long-term feasibility,
especially as regulations shift to account for
recent events and developments.
Media
Telecom
Distribution

Increased connectivity
and semiautonomous
systems have led to a
rise in data collection.
This, combined with
information gathered
from fleets, vehicles, or
bots, can provide valuable
insights for improved
decision-making and
product development.
Transportation is one of
the leading contributors
to carbon emissions. The
regulations are already
affecting the industry,
and this trend is expected
to continue. Although
many manufacturers
have adopted electric
strategies, they still
need to take steps to
proactively offset more of
their emissions.
As businesses explore
ways to enhance
their workforce or
even automate their
operations, it may not
always be beneficial
for them to personally
own their own fleet of
robots. A comprehensive
understanding of robots’
full capabilities can help
businesses determine
whether it is better to own
the robots themselves or
outsource for the service.
As collaborative robots
continue to advance
and become more
sophisticated, businesses
have a chance to
improve the efficiency
of their workforce and
complete tasks quicker.
By incorporating
collaborative robots,
companies can also
safeguard their employees
from harsh working
conditions. Keeping up
with these trends is vital
for optimizing workforces.
The regulatory and
technological landscapes
for last-mile delivery are
evolving. Companies
that can both master
the technology and stay
ahead of regulatory
changes will be well-
poised to serve their
customers effectively,
providing goods and
services quickly and
efficiently.
As systems become
semiautonomous and
fully autonomous,
insurance strategies will
need to adjust. This will
impact automobiles,
robotic systems on
warehouse floors, and
even employees whose
roles change in light
of collaborative and
assistive robots.
A Deluge
of Data
The “Threat” of
Decarbonization
New Applications for
Robotics-as-a-Service
Optimizing Your
Workforce
Leveraging Last-Mile
Delivery
Dramatic Changes
to Insurance
WHY BUILT MOBILITY, ROBOTICS DRONES TRENDS MATTER
TO YOUR ORGANIZATION

TECH
407
WHEN WILL MOBILITY, ROBOTICS DRONES DISRUPT YOUR ORGANIZATION?
Transportation
Infrastructure development
Supply chain management
Energy generation
Warehouse management
Delivery
Last-mile delivery
Prosthetics
Grid management
Energy storage
Manufacturing
Construction
Health care
Eldercare
Space exploration All trends will be relevant

Threats
OEMs, and other automotive service and parts players, are facing uphill
pressures if they fail to adapt to the growing electric market. The longer-term
threat will persist as more customers seek green options and governments
phase out fossil fuel-powered vehicles.
As vehicles gain more ability to observe and control the passengers within
them, the industry is creating entirely new venues for privacy and data
security concerns. With increased levels of data collection, automobiles are
becoming a more attractive cyber-infiltration point, posing risks for OEMs,
third parties, and customers.
The rapid adoption of EVs could potentially lead to an increase in
brownouts, electrical surges, or fires resulting from overloading home or
even commercial electrical systems. As EV adoption scales, real estate is
at a greater risk of damage and increasing insurance premiums, until the
infrastructure modernizes and stabilizes.
As bots become more dynamic and function across multiple modalities,
organizations that fail to adopt general-purpose robotics could be at risk of
falling behind competitors that masterfully incorporate these systems into
workflows.
While the applications for robots and drones become seemingly endless,
organizations that are slow to conceive and execute new ways of using these
systems risk underserving their customers and their employees.
Opportunities
To get more electric vehicles on the road, manufacturers can look for new and
novel partnerships to drive adoption. Partnerships such as ones that assist gig
workers in leasing EVs can serve as the template for new programs.
Technology companies and startups have a significant opportunity to create
seamless experiences across all mobility modalities, providing consumers with
multiple options for arriving at their desired destinations with minimal friction.
As bidirectional charging becomes more pervasive, the traditional role of
cars is transformed. Vehicles can now be the powerplant that runs a home or
business in an emergency and could potentially run full time as capabilities
and capacities increase.
When cobots create efficiencies and robotics begin to automate repetitive and
even dangerous tasks, organizations can upskill their workforce to engage
in more creative and meaningful undertakings, bringing new value to the
organization.
Training robots and drones in virtual and simulated environments allows these
systems to learn complex tasks virtually before transitioning to the real world.
Championing this process will place organizations in enviable positions.

As autonomous systems
become more prevalent,
data will continue to be
widely available. Business-
es can utilize advanced
machine learning and AI
to unlock new insights that
can feed into new products
and services. This leads to
opportunities to surprise
and delight customers
along their journeys.
For transportation industry
leaders, focusing on sus-
tainability is now a basic
requirement. However,
leaders should not become
complacent in their efforts
and must continue to push
for new technologies that
can provide competitive
advantages, such as better
and smaller batteries.
Consider calculated in-
vestments in collaborative
and assistive robots, which
can supplement strained
workforces and create
efficiencies. However, there
is not a blanket call for or-
ganizations to pursue such
solutions. They will have
to balance the threat of
not acting with the capital
expenditures required for
adoption.
Pilot and passenger ob-
servation technologies
might seem invasive to
drivers and passengers, but
investment in this tech can
reduce accidents and save
lives. In a future where au-
tonomous driving becomes
viable, these systems can
be transformed to assess
and anticipate the wants
and needs of passengers,
providing new ways to
reach them.
Electrification is radically
changing the transporta-
tion industry, and there will
increasingly be new ways
to engage with consumers
who experience this new
paradigm. Consider new
products and services as
passengers wait for their
charging vehicles, pro-
viding opportunities for
companies outside the
traditional transportation
industry.
Investing in robot compil-
ing and training technol-
ogies has the potential to
radically accelerate ro-
bots’ abilities to learn and
to adapt to their environ-
ments in real time. While
hardware is very import-
ant to robotics, software
is equally, if not more,
important, potentially re-
sulting in efficiencies that
compound in time.
1 4
2 5
3 6

CENTRAL THEMES
410
Electrification Upends the Industry
As electrification becomes more pervasive in the world
of mobility, the entire industry is being drastically
redefined. While electric vehicles’ range increases,
spurring more adoption of EVs, the roadside experience
is radically changing to include opportunities to en-
gage drivers and passengers in novel ways. The advent
of electrification is having impacts on the traditional
business models associated with dealerships, forcing
significant restructuring of how business is done. Re-
pair servicing is at significant risk, as electric vehicles
require less maintenance and different skill sets for
making repairs. This shift is affecting supply chains
and changing the types of materials needed, forcing
manufacturers to make substantial investments in ar-
eas such as lithium processing. Electrification is also
having an impact on safety standards, as EVs are less
likely to be heard by bystanders. Amid electrification’s
scaling, stakeholders must fully consider its implica-
tions on the future.
Better Batteries Boost Improvements
Electrification of the broader industry will continue
to face a “chicken or the egg” dilemma—battery ca-
pabilities must grow to drive electrification demand,
and electrification demand must grow to drive battery
technology. However, battery development continues
to press forward, with capacity improvements, efficien-
cy gains, and shrinking or altering of form factors to
make them more relevant and useful. Improvements
are helping increase the range of vehicles or the oper-
ational time of drones and robots, reducing the “range
anxiety” of consumers and commercial buyers alike.
These updates also enable these vehicles and units
to focus more power on their compute abilities either
in a trade-off for range or in addition to that extended
range. This is leading to a new class of devices and
vehicles that are more powerful and capable than ever
before over longer time spans and distances.
Expanded Capacity and Capability
In both automobiles and mobility at large, robotics
and autonomous systems are growing in their capacity
to do tasks and support payloads, while their overall
capabilities are increasing, too. These achievements
are due to advances in both hardware and software. A
constellation of vision, audio, and touch sensor sys-
tems, including lidar, radar, 2D and 3D cameras, accel-
erometers, gyroscopes, bump sensors, force sensors,
and temperature sensors are enabling these devices
to better sense the environments around them. Ad-
vanced methods for training and modeling, including
using virtual elements to simulate training scenarios,
are increasing these systems’ capabilities and helping
them achieve new thresholds of what is possible on
even shorter timelines. As these systems become more
adept and dexterous, they also benefit from increased
strength and increased payload capacity. And as they
become more dynamic, they will be better equipped to
tackle new challenges.

CENTRAL THEMES
411
Maximizing Connectivity and Communication
The increased capabilities of vehicles, robots, and
drones mean they can use their additional sensors and
software to know more about their environments. This
also requires these form factors to collect substantial-
ly more data than ever before and, in most instances,
communicate back to their OEM, a third party, the
end user or owner, and even each other. This is forcing
product and vehicle designers to more closely consider
the connectivity needs of these devices and the poten-
tial partnerships required to enable these higher levels
of communication. As levels of autonomy increase
over time, the compute load will as well, in addition to
the vast amounts of data that will need to be collected
in real time, and then streamed back and forth to the
cloud’s future iteration.
Mimicking Your Surroundings
When it comes to designing complex systems, nature
is a meaningful muse for development and execution.
To achieve fluid movement for robots, engineers often
find inspiration from the movements of creatures and
plants like jellyfish, caterpillars, and even vines—these
allow the bots to be dynamic and even reactive based
on external stimuli. Organic material also lends it-
self to the material design of robots, where tissues of
specific organisms become instrumental in optimiz-
ing the functionality of a bot. Organisms, both living
and dead, are increasingly finding themselves as key
components of a robotic system, functioning as hands
or grippers to lift and move objects. Organisms even
factor into the training of robotics, as engineers have
used ants as inspiration for enhancing the navigation
capabilities of robots in challenging terrains. As robot-
ics design remains challenging, nature will continue to
provide valuable design solutions.
Tempered Autonomy
The development of autonomous systems is a major
driving force for scientific advancement across the do-
mains of mobility, robotics, and drones. Although the
promise of fully autonomous vehicles has not yet been
fully realized, autonomous driver assistance systems
have significantly impacted the automotive industry
and provide a strong signal for increased autonomy in
the future. In the field of robotics and drones, varying
degrees of autonomy are required for efficient and ef-
fective operation. The methods for programming these
systems are becoming more sophisticated but are
still in the early stages of refinement. As the technol-
ogy advances, the level of autonomy in these systems
is expected to continue increasing, leading to more
efficient and advanced robotics and drones. The future
of these domains heavily relies on the development of
autonomous systems, and the progress made in recent
years reflects a positive step toward achieving fully
automated systems in the future.

Sertac Karaman, associate professor of
aeronautics and astronautics at MIT, for his
contributions in driverless cars, unmanned
aerial vehicles, distributed aerial surveillance
systems, air traffic control, and certification
and verification of control systems software.
Aaron Becker, associate professor of electrical
and computer engineering at the University of
Houston, for his contributions to swarm robot-
ics, distributed robotics, medical robotics, and
motion planning.
Jonathan How, professor of aeronautics and
astronautics at MIT, for his role in developing
algorithms that keep drones from colliding in
midair.
Bilin Aksun Güvenç, research professor in
the Department of Mechanical and Aerospace
Engineering at The Ohio State University, for
her role in the development of Vehicle-in-Virtu-
al-Environment testing, which allows testing of
driverless cars in a safe virtual environment.
Steven Hartley Collins, associate professor of
mechanical engineering at Stanford University,
for his contributions to versatile prostheses
and exoskeleton design.
Martin Nisser, a Ph.D. candidate in the HCI En-
gineering Group at MIT’s Computer Science and
Artificial Intelligence Laboratory (CSAIL), for his
efforts to democratize robotics and hardware by
creating self-configurable and self-assemblable
systems that address a diverse functionality of
needs.
Chris Anthony and Steve Fambro, co-CEOs of
startup Aptera Motors, for working to leverage
a lightweight chassis, low-drag aerodynamics,
solar cells, and materials science to provide
high efficiency solar electric transportation.
Wei Wang, researcher at the US Energy Depart-
ment’s Pacific Northwest National Laboratory,
for his contributions to using sugar to design
better flow batteries.
Gill Pratt, CEO of Toyota Research Institute, for
his role in using generative AI technology to
quickly teach robots new, dexterous skills.
JB Straubel and Andrew Stevenson, co-found-
ers of Redwood Materials, for pioneering circu-
lar supply chains and recycling pathways for
end-of-life EV batteries.
Dr. Pisak Chermprayong and Dr. Ketao Zhang,
who as researchers at the Imperial College Lon-
don produced innovative work on 3D printing
with drone swarms.
Daniel Preston, assistant professor of mechan-
ical engineering at Rice University, for pioneer-
ing necrobotics.
Jensen Huang, founder of Nvidia, for his vision
in enabling key components for robotics, auton-
omous systems, and AI.
Dr. Robert Playter, CEO of Boston Dynamics,
for pushing the robotics industry forward while
pledging to never weaponize technology, and
other efforts to instill public trust.
Dr. Marc Raibert, founder of Boston Dynam-
ics, for his dedication to the study of dynamic
moving systems, including robots with legs,
simulated mechanisms, and animated figures.
Henry Liu, professor of civil engineering at the
University of Michigan, for his contributions
to the first realistic simulated driving envi-
ronment based on a “crash-prone” Michigan
intersection.
Manoj Raghavan, CEO of Tata Elxsi, for his
vision in leading a company whose advanced
sensors and AI algorithms aim to keep individ-
uals safe from the hazards of driving.
Michael Smith, postdoctoral researcher in soft
robotics at EPFL, for his work in developing flexi-
ble, stretchable pumps for soft robotic systems.
Jocelyne Bloch, neuroscientist and neurosur-
geon at Lausanne University, for her work in
functional neurosurgery.
Giuk Lee, associate professor at Chung-Ang
University, for his work on assistive exoskele-
tons and wearable robots.
Hyung Ju Suh, Ph.D. candidate in electrical en-
gineering and computer science at MIT’s CSAIL,
for his work in enabling robots with human-like
dexterity and intelligence in manipulation.
Russ Tedrake, Toyota Professor of Electrical En-
gineering and Computer Science, Aeronautics
and Astronautics, and Mechanical Engineering
at MIT, for his work in combining systems theo-
ry and robot manipulation.
Zachary Manchester, assistant professor of
robotics at Carnegie Mellon University, for his
efforts to enable robotic systems to match
or exceed the level of agility, efficiency, and
robustness demonstrated by humans and
animals.
Josephine Galipon, associate professor at the
Graduate School of Science and Engineering at
Yamagata University, for exploring the potential
benefits of collaborations between robots and
living creatures.
ONES TO WATCH

IMPORTANT TERMS
413
MOBILITY
ADAS (advanced driver assistance systems)
Technologies that assist drivers by performing
certain functions in a vehicle, such as blind-spot
monitoring, lane departure warning, pedestrian
detection, emergency braking, and traffic sign
recognition.
AMD (assistive mobility device)
A mobility aid such as a wheelchair, scooter, walker,
or orthotic.
Bidirectional charging
A system that enables an electric vehicle to transfer
electricity back to the grid, as well as to charge
using electricity from the grid.
EV charging port
The connector that supplies power to an elec-
tric vehicle when it is plugged in. Of the different
connector types the most common in the US is the
North American Charging Station, or NACS. Tesla
uses it, and more manufacturers are adopting this
connector.
ICE (internal combustion engine)
An engine powered by fuel combustion, most
commonly gasoline or diesel fuel.
• Level 2
Partial Automation
The vehicle has combined automated func-
tions, like steering and acceleration. However,
the driver must always remain engaged and
monitor the environment constantly.
• Level 3
Conditional Automation
A driver is essential. Driver does not need to
monitor the environment, but must be ready to
assume control at any time.
• Level 4
High Automation
The vehicle is capable of performing all driving
functions under specified conditions. The driver
has the option to take control.
• Level 5
Full Automation
The vehicle is capable of performing all driving
functions under all conditions. The driver may
have the option to take control.
LEVELS OF EV CHARGING
There are three levels of charging. The higher the
level, the less time it takes to reach a full battery.
• Level 1
These charging stations use a standard 120v
outlet. The time to charge a vehicle’s battery can
take 60 hours or more.
• Level 2
These are the most commonly available
charging stations and are used most often for
home charging. The time to reach a full battery
is around 11 hours.
• Level 3
There are two types of rapid charging stations:
DC Fast Charging and Supercharging. Level 3
stations can fully charge a battery in under
30 minutes and most closely resemble the
customer experience of gasoline-powered cars.
Tesla’s standard is Supercharging, and the com-
pany has the largest network of Level 3 charging
stations in North America.
V2G (vehicle-to-grid)
Allows bidirectional charging so electric vehicles
can receive electricity from a charging station, or
share their stored electricity with the grid.
V2I (vehicle-to-infrastructure communication)
Enables vehicles to communicate with traffic
lights, RFID readers, cameras, lane markers, and
other parts of the physical world.
V2V (vehicle-to-vehicle communication)
Allows vehicles to exchange information with other
vehicles, sharing data such as speed and location.
Levels of Automation
The Society of Automotive Engineers (SAE) clearly
defines six levels of driving automation:
• Level 0
No Automation
A human driver manually performs all tasks.
• Level 1
Driver Assistance
A driver controls the vehicle, but the vehicle
design may include some driving assistance
features.

IMPORTANT TERMS
414
ROBOTICS
Cobot
A collaborative robot built for interaction with
human workers, assisting with certain tasks, often
those that are repetitive or harmful to humans.
Exoskeleton
A rigid, mechanical robotic structure that encases
the human limb, or envelops the body, and assists
the wearer in motion-based activities, such as
walking or lifting.
Microrobotics
A field of robotics that develops miniature robots,
typically smaller than 1 mm (or .001 meters) in size.
Nanobot
A field of robotics that develops robots at the scale
of a nanometer (or 10-9 meters).
Necrobotics
A field of robotics that utilizes biological material,
such as insect cadavers, as robotic components.
Drone
An unmanned vehicle that can operate in the air, on
land, or in the sea:
• Fixed-wing drone
A drone with one rigid wing, resembling an air-
plane. It is typically capable of remaining in the
air longer and flying longer distances than other
drone types.
• Fixed-wing hybrid VTOL
A drone with a rigid wing, and rotors that are at-
tached to either side, enabling vertical takeoffs
and landings.
• Single-rotor drone
A drone with a single rotor on top, much like a
helicopter.
• Multi-rotor drone
A drone with multiple rotors. The most common
multi-rotor drone is a quadcopter, which has
four rotors.
• Drone swarm
Fleets of networked drones capable of coordi-
nated operations and communication.
• eVTOL (electric vertical take-off and landing)
An electric-powered drone that has the ability to
take off and land vertically, as well as hover.
• Federal Aviation Administration (FAA)
Drone operators are required to comply with FAA
rules. There are a multitude of airspace restric-
tions as well as FAA-Recognized Identification
Areas (FRIAs). In a FRIA, a drone operator may fly
their device without Remote ID.
• UAV
Unmanned aerial vehicle.
Quadrupedal robot
A four-legged robot.
Robotics
The use of a physical, mechanical device capable
of performing tasks at various levels of complexity,
either on command or via preprogrammed instruc-
tions.
DRONES
AGV (automated guided vehicle)
A robot that follows specific lines, lanes, or other
markings, often used in industrial settings.
AUV
Autonomous (or uncrewed) underwater vehicle.
BVLOS (beyond visual line of sight)
Operating an UAV outside of the visual line of sight
of the operator.

MOBILITY
TRENDS

TECH
Decarbonizing Mobility
In the US, the transportation sector gener-
ates the largest amount of greenhouse gas
emissions—primarily from burning fossil fuel
for cars, trucks, ships, trains, and planes. In
efforts to mitigate climate change, this in-
dustry is a prime candidate for impact. Some
states offer proactive examples. California
allocated more than $50 billion in funds to
address climate change and move away from
the use of fossil fuels. In the area of trans-
portation, the state is implementing regu-
lations to reduce toxic freight pollution and
accelerate the deployment of zero-emission
trucks; officials are also allocating emergency
funding to ensure the safety of public transit
riders and workers, in hopes of persuading
more people to choose that form of transpor-
tation. California also enacted a Low Carbon
Fuels Standard to ensure that it aligns with
climate and environmental justice priorities.
In Europe, the transportation sector is also a
major contributor of greenhouse gasses, and
officials are aiming for a 90% reduction to
meet the EU’s target of carbon neutrality by
2050. This transition poses significant chal-
ments in an electric future, with plans and
announcements from the late 2010s now
becoming reality. Western automakers are
even investing in lithium mining companies
to ensure a stable supply of this key com-
ponent of EV batteries, committing billions
of dollars to secure the resource. General
Motors has invested $650 million in Lithium
Americas to develop the Thacker Pass Mine
in Nevada and has entered into supply agree-
ments with lithium companies like Livent.
Ford has arranged lithium supply deals with
Chilean supplier SQM, Charlotte-based Albe-
marle, Nemaska Lithium in Quebec, and the
Argentina mining company Rio Tinto.
Ford and Uber are collaborating through the
Ford Drive program to provide flexible leases
of EVs to Uber drivers. The goal is to support
Uber’s efforts to convert more drivers to EVs
and reduce emissions, while Ford will benefit
from expanding the presence of its Mustang
Mach-E EVs. The initial pilot program offers
one- to four-month leases in San Diego, Los
Angeles, and San Francisco.
School districts are replacing traditional bus-
es with electric models to reduce emissions
and save on fuel costs, with potential orders
up to $1 billion over the next five years. Blue
Bird has opened a new Electric Vehicle Build-
Up Center in Georgia to meet the increased
demand for electric school buses, and aims to
increase production from 100 to 5,000 electric
buses annually. Blue Bird will assemble its
“Vision” and “All American” buses, each with a
155 kWh battery providing 120 miles of range.
Mass Market Goes Custom
During the COVID-19 pandemic, the shortage
of semiconductor chips caused significant
supply chain constraints. The lack of chips,
used in automotive parts such as operating
systems, cameras, sensors, and entertain-
ment systems, led to a decline in production,
resulting in a shortage of vehicles. Dealers
had little to no stock on their lots and were
forced to operate as showrooms, where some
customers placed orders for a custom vehi-
cle with a longer lead time. While supplies
are slowly and tortuously improving, CEOs of
major car brands anticipate that inventory
lenges for European cities, which significant-
ly contribute to these emissions; to support
them in this effort, the EU is offering help
creating Sustainable Urban Mobility Plans.
However, various obstacles, including gover-
nance issues, need to be addressed. Addi-
tionally, the Council of the European Union
has recently passed a new regulation aimed
at facilitating EV travel across the continent
while mitigating the impact of greenhouse
gas emissions. The regulation mandates the
installation of fast charging stations that
offer a minimum of 150 kW of power at a
maximum distance of 60 km (37 miles) from
each other along the Trans-European Trans-
port Network (TEN-T) highway system by the
year 2025. As more electric mobility technol-
ogy is adopted, cost structures and opera-
tions will continue to change. The mobility
industry, including vehicle manufacturers,
is already feeling the impact and rethinking
business models.
Automaker Restructuring
Major car companies and transportation
platforms are making significant invest-
ELECTRIFICATION TRANSFORMS MOBILITY ECOSYSTEMS

TECH
levels will never go back to where they were
pre-pandemic. But this window has revealed
that customers will wait and pay for their
dream car, with the ability to pick their de-
sired color, features, and accessories. Conse-
quently, build-to-order purchases will likely
continue to increase and render big vehicle
inventories less necessary. Ford is extending
this schematic to its Mustang Mach-E, offer-
ing a $1,000 discount in some instances for
customers who pre order. This trend is likely
to continue as more production lines shift
to increased EV manufacturing. Such a shift
does have tremendous implications for the
industry, changing the nature of dealerships
and impacting current commission and
profit-sharing structures. Both the industry
and consumers will have to adjust to such
restructuring, and adequately prepare for
impending shockwaves.
Incentive-Driven Investments
Many vehicle manufacturers and battery
makers are investing heavily in the devel-
opment of electric vehicles and their future
success. A recent analysis by Atlas Public
to purchase EVs, the Inflation Reduction Act
has unexpectedly complicated the practice.
In August 2023, the law restricted the $7,500
tax credit to only EVs assembled in North
America. As a result, automakers that can no
longer incentivize new car buyers this way
are encouraging consumers to lease EVs.
Global Battery Belts
Major players in the automotive and battery
chemistry industries are investing in US bat-
tery production. States that attract electric
vehicle manufacturing and battery plants
highlight the job opportunities these in-
vestments create, not only within the plants
themselves but also in the surrounding
supplier and logistics sectors. The growth in
EV manufacturing is especially prominent in
the Battery Belt: The area, which runs from
Detroit to Georgia, offers lower electricity
costs and strong manufacturing employ-
ment growth, driving industry expansion. As
an example, Hyundai Motor Group and LG En-
ergy Solution announced a joint investment
of $4.3 billion in a new electric battery plant
in southeast Georgia. By late 2025, the plant
expects to be producing batteries for electric
vehicles and aims to accelerate the produc-
tion of electrified Hyundai and Kia vehicles in
North America. But despite this push by US
companies to produce more batteries, most
production is still dominated by China. Last
year, China refined approximately 95% of the
world’s manganese, around 70% of cobalt and
graphite, two-thirds of lithium, and over 60%
of nickel, all of which are vital components
in the production of lithium-ion batteries. In
time, that could change, especially with the
2023 discovery of lithium in a US volcano
along the Nevada-Oregon border, which could
result in a stable and sustainable source of
the metal for the US for decades to come.
Battery Recycling
The debate surrounding the sustainability of
EVs versus internal combustion engine vehi-
cles (ICEs) revolves around the environmental
impact caused by the mining of lithium and
cobalt, which are crucial materials for batter-
ies. However, by using recycled battery com-
ponents, the need for new mineral mining can
be reduced, leading to a more environmentally
Policy reveals that a total of $860 billion will
be invested globally by 2030 toward the tran-
sition to EVs. In the US, investments are ex-
pected to total $210 billion—almost a quarter
of the entire investment. Amid this spending
increase, some of the benefits are trickling
down to consumers. Increased plant capaci-
ty, production scaling efforts, and improved
battery material costs are enabling Ford to
reduce the prices of its F-150 Lightning elec-
tric truck models by up to $10,000 to incen-
tivize sales. The base F-150 Lightning Pro will
be priced at $49,995. Elsewhere, Geely an-
nounced the Galaxy E8 sedan (about the size
of a Honda Accord) will be on sale in the Chi-
nese market for under $25,000. Automakers
are also taking advantage of tax incentives
associated with EV production. The state of
Georgia has offered Rivian $1.5 billion in tax
incentives for the company to build a $5
billion factory east of Atlanta. There, Rivian
plans to produce 400,000 electric vehicles
annually and provide 7,500 new jobs after
officially getting the green light to move
forward with production. While tax credits
have been effective at enticing consumers

TECH
friendly EV market. Currently over 80 compa-
nies worldwide are engaged in the recycling
of electric vehicles, with over 50 startups
receiving at least $2.7 billion in funding from
corporate investors such as automakers, bat-
tery manufacturers, and mining companies.
Industry insiders predict that by 2040, up
to 40% of the battery materials used in new
electric vehicles could come from recycled
sources. A prominent player in this space,
Redwood Materials, has partnered with Ford
and Volvo to establish responsible disposal
and recycling pathways for end-of-life EV bat-
teries. They are also going beyond EVs and col-
laborating with Rad Power Bikes to conduct
the same process for retired e-bike batteries.
Ascend Elements has secured $542 million
in Series D funding along with $480 million
in earlier Department of Energy grants for
recycling lithium batteries into black mass to
be reused in other materials. Additionally, Nth
Cycle has secured a focused Series B funding
round of $50 million, and Green Li-ion has
received $20.5 million in funding to support
its recycling initiatives.
in Mitra Chem to use iron-based cathodes
to make battery technology more accessible
and cost-effective. Researchers at Pacific
Northwest National Laboratory have devel-
oped a breakthrough in flow battery technol-
ogy by using a solution based on sugar to
lead in the development of low-cost, long-du-
ration energy storage systems, which could
impact the source of energy for charging EVs.
Startup Ample is taking a different approach,
focusing on battery swapping as a means to
keep EVs on the road.
Solar Vehicles
Solar-powered EVs require less frequent
charging and can increase efficiency. A
recent study found that solar-powered cars
can travel between 11 and 29 kilometers per
day using solar energy, reducing the need
for frequent charging. These solar cars have
the potential to make electric transportation
cleaner and more affordable by minimizing
pollution from electricity production. Dutch
startup Lightyear saw the advantage of
solar, but despite its efforts, was declared
insolvent in 2023 shortly after releasing its
$250,000 Lightyear 0 solar car. After raising
more capital, the company will try again with
a more affordable Lightyear 2, which has a
starting price of $40,000. First revealed as a
concept, Kia’s EV9 SUV features solar panels
to supplement its 100-kilowatt-hour battery;
the company started taking reservations for
the EV9 in late 2023. In contrast to a large SUV,
California-based Aptera Motors focuses on
producing ultra-efficient EVs through aerody-
namics and weight savings. Its three-wheeled
EV, the Aptera, boasts a highly aerodynamic
design with a drag coefficient of 0.13, offering
up to 400 miles on a single charge, which can
be supplemented through included solar pan-
els. While solar vehicles will not eliminate the
need for charging any time soon, their integra-
tion will help the industry move toward more
sustainable transportation.
Shifts in the Servicing Model
While the mobile service model is not new,
it’s increasingly impacting the mobility and
automobile industry. Now, mechanics can
travel to meet customers in their physical lo-
cations, and over-the-air updates can, in some
Better Batteries
Electric vehicle manufacturers and their
partners are working hard to improve batter-
ies by exploring new battery types, designs,
and materials. The development of sol-
id-state batteries and other innovative bat-
tery solutions will make batteries smaller,
safer, and capable of providing longer rang-
es, much like the Harvard John A. Paulson
School of Engineering and Applied Sciences
(SEAS) research that developed a new way
to make solid-state batteries with a lithi-
um metal anode. Beyond the obvious safety
and range benefits, enhanced batteries will
provide greater versatility in terms of bat-
tery shape, which will allow for more options
in different mobility platforms and provide
more cabin space in vehicles. QuantumS-
cape is striving to make solid-state batteries
available to the public as early as this year.
This technology will allow EVs to travel up
to 400 miles on one charge and recharge in
only 15 minutes. Toyota plans to be using
solid state batteries by 2028, reducing the
size, cost, and weight of its EV batteries by
50%. General Motors is investing $60 million

TECH
instances, resolve digital-based issues. As the
model takes hold, this presents a significant
threat to traditional dealership/servicing
models, and companies are racing to adapt.
Rivian plans to repurpose large commercial
Amazon vans to function as mobile service
vehicles for its consumer and commercial
vehicle fleet. These electric Rivian Service
Vans will offer maintenance, repair, and ve-
hicle-to-vehicle charging. Ford is recognizing
the need to enhance the vehicle service ex-
perience and is moving to expand its remote
service offerings. The company will be provid-
ing complimentary pickups and deliveries of
its vehicles and mobile repair options to more
of its customers. Startups are also taking
advantage of this shift. On-demand car care
startup Spiffy has acquired $30 million in
funding to aid car dealerships in expanding
its mobile service offerings. Through sales
of its software and van upfits, Spiffy plans
to help dealerships and repair shops provide
mobile brake maintenance and oil changes.
Repair servicing is really beginning to go the
extra mile to win customers and foster and
enhance relationships.
the other end of the spectrum, China has un-
veiled its first battery-electric container ship,
the 700 TEU, which has a capacity of carrying
700 20-foot containers. Even micro mobility
is becoming electrified, as the e-bike compa-
ny Cowboy introduces a more affordable Core
line of its e-bikes and startup AtmosGear
has unveiled electric inline skates featuring
a 20-mile range on a full charge.
Electrification Expands to Other Vehicles
The rise in electrification of vehicles is
extending beyond consumer cars. Deliv-
ery carriers like Amazon and the US Postal
Service are investing in the technology:
Amazon aims to deploy 100,000 electric
delivery vehicles by 2030, and the USPS says
it will buy over 66,000 electric vehicles by
the end of 2028. When it comes to aviation,
ZeroAvia has successfully flown the world’s
largest hydrogen-electric aircraft as part of
the HyFlyer II project, a government-funded
initiative aimed at making small passen-
ger planes more environmentally friendly.
Magpie Aviation, a California company, has
proposed using electric aircraft as towing
planes to connect with passenger or cargo
planes that have sufficient battery power
for takeoff, landing, and flight to alternative
airports. Electric boating is impacting both
small and large-scale initiatives. BMW and
Tyde have launched the ICON electric boat, a
43-foot eco-friendly vessel powered by hydro-
foils, six BMW i3 batteries, and two electric
motors. It has a range of more than 50 nau-
tical miles and a max speed of 30 knots. On
Amazon Prime electric delivery vans built by Rivian
on the street in Seattle.
Image credit: 400tmax/istock.com

TECH
Charging Gets A Roadmap
The electric vehicle revolution is well under-
way, but a key bottleneck remains a concern:
charging infrastructure. Rural and suburban
consumers alike have had their desire to
adopt EVs hampered as a result of “range
anxiety”—the fear of running out of charge
mid-journey. Recognizing this anxiety, regu-
lators in the US have approved $5 billion for
EV charging projects over the next five years
and recently allocated an additional $2.5
billion for community chargers. Europe has
gone even further, mandating fast charging
stations within every 60 km length of major
highways by 2026.
The private sector is stepping up its efforts,
too. Automakers have gotten closer to con-
vergence on charging standards with several
top manufacturers adopting the Tesla con-
nector for their vehicles. Major brands like
Walmart, Sam’s Club, Comcast, Ikea, Marriott,
and Hilton have committed to build tens of
thousands of new chargers across all of their
properties, in addition to tens of thousands
in combined charging station commitments
Major automakers including Ford, Rivian,
Fisker, GM, Honda, Volvo, Jaguar, Nissan,
and Mercedes-Benz have pledged to adopt
Tesla’s NACS connector or offer NACS adapt-
ers, making standardization one step closer
in the US. Tesla has also begun to expand its
“Magic Dock ‘’ superchargers in several US
states, adding CCS compatibility in much
the same way it has had to in Europe. These
two developments have not only moved
global charging capacity closer to universal
compatibility but have instigated a renewed
incentive for further charger development.
As EV adoption and universal compatibili-
ty increases, the need for charging station
capacity intensifies. This opens up a new
vector of competition throughout the indus-
try where vendors, partnerships, and enter-
tainment offerings at charging stations may
dictate station utilization rates, and inno-
vations like real-time pricing could allow
for solar-powered stations to out-compete
traditional stations during fluctuations in
solar availability.
Redefining the Roadside
The rise of EVs is transforming our roadside
experience. Extended charge times have led
retailers like Walmart, Ikea, and Macy’s to
integrate charging tech, turning wait times
into retail opportunities. Tesla’s drive-in-diner
concept provides culinary and film experienc-
es while EVs charge, while Juxta offers autono-
mous stores at traditional charging hubs.
With the surging EV demand, innovative
solutions are addressing infrastructure gaps.
Ample swaps batteries in just five minutes.
EV Safe Charge is piloting robots to charge ve-
hicles in non-equipped garages. SparkCharge
delivers on-demand charging where fixed
infrastructure is lacking.
Even in emergencies, adaptations are evident.
Apple’s new satellite function broadens the
coverage for roadside assistance in low-signal
areas. Meanwhile, AAA assists stranded EVs
with mobile charges and is exploring electric
tow trucks for green, on-the-go recharging.
made by BMW, GM, Honda, Hyundai, and
Mercedes. Improvements to existing infra-
structure are also underway with Tesla prom-
ising 40% faster charging through its V4
technology, companies like EVgo launching
initiative’s to “renew” existing stations, and
navigation services like Waze integrating
charging locations into its navigation tech
to improve peace of mind in trip planning for
EV adopters.
Charging Standardization
One of the lingering issues in the develop-
ment of the EV market has been fragmen-
tation of charging standards. Globally, the
Combined Charging System (CCS) has been
the prevailing standard, but the North Amer-
ican Charging Standard (NACS)—commonly
known as the Tesla connector—has become
the standard in North America with Tesla’s
overwhelming EV market share. This diver-
gence has become more complicated since
Tesla began open-sourcing its NACS connec-
tor design at the end of 2022, but progress
has been made toward cross-compatibility.
VEHICLE CHARGING SCALES

TECH
Electrifying Cities
Accelerating frequency and severity of cli-
mate events, along with rising electric vehicle
adoption, has seen cities push their electri-
fication strategies to the top of their agenda.
On the grid, decentralized energy systems
have been a consistent theme. Public utilities
like Vermont’s Green Mountain Power are in-
stalling batteries in customers’ homes to en-
sure power resilience during outages and to
optimize costs by utilizing stored resources
during periods of low supply or peak demand.
In North Carolina, a residential community
called Heron’s Nest is being developed to
include a microgrid independent of the larger
grid. Each home includes a solar energy sys-
tem contributing to a collective network that
not only powers the community but ensures
its resilience and sustainability.
In transportation, cities like New York City
and Los Angeles have considered requiring
rideshares to run completely on EVs by 2030.
To cope with rising demand for charging in
cities, governments are racing to expand
capacity. France, for example, is working with
with initiatives from automakers to try to
bring chargers to the home. Hyundai, for
example, has pushed promotions that would
provide a free EV charger and a reduction in
installation costs to every new customer.
These efforts to transform have paved the
way for more widespread adoption, but the
swift transition has also uncovered poten-
tial challenges for EVs at home. Surveys by
the Electrical Safety Foundation indicate
that the electrical systems of over half of
US homes may be unable to safely carry the
continuous load that EV charging demands.
If the rapid adoption of EVs leads to an
increase in brownouts, electrical surges, or
fires resulting from overloading home elec-
trical systems, more stringent regulations
may become more popular for concerned
local governments and homeowners asso-
ciations. Still, forward-thinking communi-
ties may use the opportunity to collaborate
on more communal solar arrays or shared
charging stations, presenting an efficient
and cost-effective solution.
Electreon to install wireless charging on its
roadways, and the US recently earmarked
$25 billion to expand community charging
capacity. These actions underscore what is
becoming a global shift toward a sustain-
able and resilient urban future, ensuring
cities remain powered, efficient, and ready
for the next era of electrification.
EVs At Home
As the EV market has matured, a significant
shift has been taking place in the home. On
the heels of California’s 2020 mandate for
solar installations in new homes, Illinois has
passed laws requiring new and renovated
properties to have at least one EV-capable
parking space for each residential unit that
has dedicated parking by 2024. A proposal
under consideration by New Mexico legis-
lators would go even further, mandating all
new homes be constructed with a solar-pow-
ered system and EV charger. In Germany, a
grant to help fund home-charging installa-
tions was tapped out in one day, with over
33,000 people applying just hours after it
went live. These regulatory efforts coincide
Electric vehicles taking advantage of chargers on
city streets.

TECH
Bidirectional Charging
Electric vehicles with bidirectional charging
capabilities present an exciting frontier.
These vehicles can not only consume energy
but also supply it for other applications,
transforming the traditional role of cars. No
longer merely used for transportation, bidirec-
tional EVs now have the potential to energize
homes, businesses, and entire communi-
ties. With bidirectional charging, owners
can strategically charge their cars overnight
using affordable grid energy and then utilize
the EV’s stored energy during high-demand
daytime periods, ensuring efficient use of
resources. Growing challenges such as the
energy crises linked to geopolitical events
in Ukraine, escalating climate disruptions
across the globe, and widespread concerns
around outdated infrastructure and cyberse-
curity only underscore the timely importance
of this technology.
The automotive industry has been quick to
embrace this innovation. Ford, Genesis, Hyun-
dai, Kia, Mitsubishi, Nissan, and Volkswagen
have all rolled out vehicles with bidirection-
al charging. GM has gone a step further,
declaring that by 2024, vehicle-to-home
(V2H) charging will be a standard feature
in its vehicles. Other industry leaders like
BMW, Volvo, and Porsche are actively testing
this technology. Even Tesla, which initially
seemed skeptical, has announced plans to
incorporate bidirectionality in all its vehicles
by 2025.
Municipalities are recognizing the value of
this transformative technology in energy
management. Utrecht, a prominent city in
the Netherlands, stands out in its adoption,
actively installing bidirectional charging
stations that allow shared vehicles to con-
tribute energy back to the grid. In the US,
California is at the forefront, contemplating
legislation to make bidirectional charging
mandatory for new EVs. The possibilities are
vast, and the trajectory suggests a future
where our vehicles play an integral role in a
sustainable energy ecosystem.
Rendering of the Nissan Vehicle to Home (V2H) System concept
Source: Nissan

TECH
Livable Cabins
Auto manufacturers are revolutionizing the
way we experience our vehicles. Rather than a
car solely serving as an uninspired vessel to
get us from one location to another, manu-
facturers are emphasizing enhancing cabin
environments. Now, they’re not just for driving
but also for riding, relaxing, working, and
playing. In many instances, automakers at-
tempt to transform these landscapes through
the increased use of screens. In a Peugeot
concept called Inception, a screen-supported
human-machine interface replaces the tra-
ditional steering wheel and displays control
information. Continental is also capsulizing
on the screen-infused future with its ultraw-
ide In2visible: a touchscreen that spans the
full length of the dash resulting in a high-in-
tensity driving experience. Ultimately, such
interfaces are intended for fully integrated
digital experiences. Chrysler has given its
view of a fully electric future where the cock-
pits of its vehicles integrate Stellantis-brand-
ed software, including the STLA Brain oper-
ating system, STLA AutoDrive Level 3 driver
assist, and STLA Smart Cockpit infotainment
ing drivers of auditory and even touch-based
feedback they are accustomed to. This also
affects bystanders. The lack of noise from
electric cars contributes to their high rate
of accidents: EVs are 40% more likely to hit
a pedestrian than a normal car, and for the
visually impaired population, this number
jumps to 93%.
Toyota is adding audible features to its
electric cars—including a simulated gear
stick and artificial “noise” simulating a
combustion engine—aimed at drivers who
prefer a more traditional driving experience.
This simulated manual transmission expe-
rience might even come with the possibility
of stalling in order to retain the charm and
enjoyment of driving a manual transmission
car. Hyundai and Dodge are also exploring
sound design in the execution of their EVs,
not just for preference reasons but because
the addition of artificial engine noise is also
a safety standard, alerting pedestrians to
the presence of electric vehicles. Ultimately,
some manufacturers are hoping that drivers
are more likely to adopt electric cars if they
sound and feel just like their gas-powered
counterparts, while the inclusion of these sim-
ulations is primarily about keeping people safe.
In-Vehicle Connectivity
The role of high-speed, low-latency connectiv-
ity is becoming a prominent factor impacting
the performance of the vehicle along with the
in-cabin experience. While internet access in
cars is not new—Starlink currently provides
connectivity for Subaru—carriers such as
ATT and Verizon have ambitions for exe-
cuting on this modality. ATT, in particular,
sees 5G connectivity in conjunction with edge
computing as a way to enable new functions
and services through this network and cloud
convergence. This framework will not only sup-
port the safety and mission-critical functions
of the car but also intelligent transportation
systems, teleoperations,and autonomous
driving, along with onboard infotainment
systems. More tactically, Verizon launched the
Connected Car by Verizon, which offers select
BMW models features such as voice, data, and
unlimited Wi-Fi hotspot connectivity through
subscription. With regards to more tech
system. This connected experience will sync
with calendars and smart home data, handle
driving tasks within designated areas, and
even offer wellness and fun features, like
meditation and in-car games. But the prolif-
eration of large and clunky screens is being
sharply criticized, as they have impacts on
ergonomic, safety, and aesthetic factors.
If they continue to grow unchecked, they
could ultimately be challenged by regula-
tors. Despite these concerns, as personal
mobility is enhanced with longer ranges and
increased implementation of autonomous
driving features, auto manufacturers will
continue to improve comfort within vehicles
and engage customers through new onboard
touchpoints, impacting how we’ve tradition-
ally engaged with roadside attractions and
amenities.
Simulated Driving Experience
Electric vehicles are known for being in-
credibly quiet, and some view this as prob-
lematic. Specifically, EVs don’t make the
typical shifting noise that drivers are used to
hearing, and this can cause issues by depriv-
IMMERSIVE VEHICLES CONNECT TO OTHER ECOSYSTEMS

TECH
demonstrations or spectacles, General Motors
collaborated with Etisalat to host a gaming
tournament that took place entirely within
connected Chevrolet and GM cars equipped
with in-vehicle Wi-Fi. The tournament fea-
tured 10 gamers who participated in a 12-hour
event covering 450 km of roads in the United
Arab Emirates, organized by OnStar. Cradle-
point partnered with The University of New
South Wales for the Bridgestone World Solar
Challenge race, by providing 5G/LTE wireless
network edge solutions for the Sunswift 7
solar race car and support vehicles. This tech-
nology allowed the team to remotely monitor
the car’s performance and telemetry data. As
connectivity becomes more reliable, more
subscription services like Verizon’s Connect-
ed Car are inevitable, much to the disappoint-
ment of most consumers.
Mobile Entertainment Hubs
Vehicles are becoming increasingly equipped
with multiple entertainment options for
passengers and drivers, including technolo-
gies such as immersive audio, large screens,
and the ability to stream movies, TV shows,
in its vehicles. The service connects to the
cloud to access an extensive gaming library.
While the use of these technologies is limit-
ed for safety, inevitably, these technologies
are intended for broader adoption as we
inch closer to an autonomous driving future.
Ultimately, these new platforms intend to
simplify content discovery and enhance in-
car entertainment.
CarOS
As cars become even smarter and offer
more immersive experiences, navigational
and entertainment systems will continue to
evolve, necessitating a holistic and central-
ized operating system. In the current market,
many options are available, developed by
both OEMs and technology companies. While
tech companies have taken a lead in years
past, some OEMs are attempting to regain
control. Google is making progress in the au-
tomotive industry through Android Auto—an
app that operates on the user’s smartphone
and wirelessly sends navigation, parking,
media, and messaging to the infotainment
system of the vehicle—and Google Built-in,
which directly integrates Google services with-
in the vehicle. Brands like Chevrolet, Renault,
Volvo, Polestar, and Honda are set to showcase
features like YouTube available in cars with
Google Built-in, gaming with GameSnacks,
conferencing with Cisco, Microsoft Teams,
and Zoom, and further integration with Google
Assistant. General Motors is indexing sig-
nificantly on Google, and in the process is
ditching Apple CarPlay for Android Auto in
its upcoming electric vehicles. Relating to
OEMs, Swedish EV manufacturer Polestar has
teamed up with Xingji Meizu to develop an
operating system for its cars in China, as part
of Geely Group’s strategy to tailor vehicles for
the Chinese market. The new system, based on
Flyme Auto, will connect with in-car apps and
mobile phones, and use the latest smart tech-
nologies. Jaguar Land Rover is set to release
its new electrical/electronic (E/E) architecture,
EVA Continuum, with the help of Continental
subsidiary Elektrobit for running its software
platform and operating system. This move
enables faster development and over-the-air
software updates, a strategy that other auto-
makers are also beginning to adopt.
and video games. While Apple and Mer-
cedes-Benz have enabled spatial audio in
vehicles, Jaguar Land Rover is turning to
haptics to enhance the music experience.
Its “Body and Soul Seat” technology, dubbed
BASS, features headrest-mounted mem-
branes and transducers that provide haptic
feedback in response to lower frequencies
in music—and even offer health benefits
through integrated wellness programs. Auto-
makers are continuing to add features to be
enjoyed by passengers while they’re com-
muting or charging their EVs. Xperi and BMW
have collaborated to use TiVo’s video media
platform in cars to offer customers access
to various video content providers, including
linear and on-demand streaming services,
covering country-specific content such as
news, movies, and media libraries. Polestar’s
most recent software update, P2.9, includes
the addition of YouTube for streaming videos
during vehicle charging and a revamped
version of Apple CarPlay that allows Maps
to be projected onto the instrument cluster.
Polestar is also the first automaker to offer
Nvidia’s cloud gaming service, GeForce NOW,
IMMERSIVE VEHICLES CONNECT TO OTHER ECOSYSTEMS

SCENARIOS
SCENARIO YEAR 2038
What if expanded vehicle capabilities redefine the nature of home ownership?
As younger generations grapple with lower real income, higher housing prices, and an expanded tolerance for
remote work, connected vehicles became the new starter home of yesteryear. Risks associated with pandemics,
crime, and higher rates of depression drove people out of cities and closer to nature, and their vehicles stepped
in as an invaluable resource. Mesh networks of 5G and satellite-enabled internet allow people to work, consume
entertainment, and connect with others virtually in the comfort of their vehicle—no matter where it’s located.
Vehicle cabins are redesigned around sleeping, working, and relaxing, while new features that enable cooking,
food storage, and waste management become popular add-ons. AI allows vehicles to assist passengers with
daily tasks like cooking food or supporting work, while providing security, health monitoring, and the ability to
contact emergency services or autonomously drive the vehicle to receive help during situations of duress.

TECH
Mobility Simulation
Simulation has become an increasingly
ubiquitous component across several areas
of mobility and will be a defining characteris-
tic in the development of its future. Tesla, for
example, has been relying on its Dojo super-
computer to simulate billions of miles of driv-
ing for its autonomous driving software since
choosing to rely purely on cameras and neural
nets to provide its vehicles with self-driving
capabilities. Mercedes has also made simu-
lation more central to its strategy, forming a
partnership with Nvidia that only underscores
this trend. Leveraging Nvidia’s Omniverse
Generative AI platform, Mercedes hopes to im-
prove its vehicle designs and craft enhanced
driving algorithms by simulating countless
potential real-world scenarios, to create ve-
hicles that respond optimally across various
conditions.
Universities like Ohio State and the Universi-
ty of Michigan have brought mobility sim-
ulation to academia as well. Ohio State, for
example, uses simulated environments to
test the safety and aptitude of real driverless
sion sources, allowing vehicles to reach as
much as 1,000 horsepower while maintain-
ing equilibrium to prevent risks like disin-
tegration. Indian technology company Tata
Elxsi is developing technologies that would
alert automobiles to approaching emergency
vehicles, and indicate slippery roads, dan-
gerous curves, and potholes or roads under
repair so that they can appropriately adapt
to the approaching conditions. Honda and
Sony’s Afeela cars are being designed to
“feel” their driver’s moods and allow them to
express themselves and interact with others
on the road via external screens on the front
of their cars. Ford has even patented self-re-
porting technology that allows cars to lock
out drivers, disable features like air-condi-
tioning, restrict driving to only certain hours
or locations, and to even utilize self-driving
to return cars to an impound lot. As vehicles
continue to get more connected and aware,
mobility will look strikingly different from
the modern day mobility environment as we
know it.
Pilot and Passenger Observation
Technological advancements have enabled
detailed monitoring of both driver and pas-
senger behaviors, significantly enhancing
safety and operational efficiency within vehi-
cles. Safety sensors and equipment are now
being built to detect distraction, drowsiness,
and substance influence. Smart Eye’s Driver
Monitoring System, for example, gauges driver
attentiveness, providing real-time feedback to
prevent mishaps. Similarly, Magna’s advanced
driver assistance systems use cameras and
interior mirrors to identify distracted behav-
ior and try to reduce accidents. In the US, the
government is pushing to require in-vehicle
breathalyzers by 2026 so that drivers under
the influence cannot take the wheel until the
vehicle confirms their capacity to wield it
responsibly. As vehicles gain more ability to
observe and control those within them, entire-
ly new privacy and data security concerns are
emerging. With monitoring systems becoming
more common in standard vehicle require-
ments and designs, a balanced framework
that addresses both safety enhancements
and privacy concerns will become crucial for
cars, while the University of Michigan has
begun running simulations on historically
crash-prone intersections to try and reduce
incidents. Globally, the Indian city of Chennai
is now using traffic simulation to study and
improve congestion within its borders; while
academics in Egypt and Brazil are relying
on simulation to test and demonstrate new
innovations in traffic signal technology. As
data abundance exponentially expands,
simulation will increasingly be at the heart
of mobility design.
Self-Aware Vehicles
Vehicle connectivity is changing the land-
scape of mobility. The abundance of sensors,
monitors, computing power, and network
availability has enabled vehicles to provide
information and, in some cases, act on it
in ways that would have once been consid-
ered unprecedented. For example, this year,
Goodyear tested its SightLine tires, which
can measure tire-road friction, wear, load,
inflation pressure, and temperature. Soft-
ware-enabled vehicles can integrate with
and monitor multiple powertrains or propul-
DATA COLLECTION ENABLES SAFETY AND AUTONOMY

TECH
getting the best out of these technologies.
Mobile Weather Stations
The Mobile Weather Station (MWS) has be-
come a crucial instrument for the real-time
monitoring of weather and environmental
conditions, offering granular insights crucial
for sectors like transportation, agriculture,
and emergency response. While universities
across the United States, along with com-
panies like Verizon, are creating some of
the technologies that enable the stations,
several states and municipalities are using
them to bring about improvements across
their domains. For example, in New York
City, the FloodNet Initiative utilizes MWS to
monitor water levels, predict flooding events,
and optimize drainage systems to minimize
potential damage to the metropolitan area.
In California, MWS is being used to detect
early indicators of potential wildfires and
changes in algae levels that can ultimately
save lives and preserve ecological health. In
Connecticut, MWS was introduced to gather
air pollution data to ensure public health and
enforce breaches in regulatory compliance.
ers invaluable data that enhances the user
experience in addition to providing insights
that can inspire better urban planning, pub-
lic transit, and infrastructure decisions by
local governments. Companies across Asia
and the Middle East have already proven the
viability of superapps, but now US mobility
and technology companies are striving to
be the first to make superapps the norm
domestically.
Utilizing Mobility Data
Mobility data has become a pivotal asset in
modern transport dynamics. Real-time mon-
itoring and analytics, combined with geo-
spatial data, enables companies to harness
spatial information to create and enhance
products and services across the mobili-
ty spectrum. New open map data sets like
what are provided by the collective efforts of
Amazon, Microsoft, and Meta add to the map
data already offered by Apple and Google;
this additional competitive pressure will
likely lead to further innovation across the
space. More widespread availability of data
and analytics has also allowed companies
like Uber and Lyft to leverage mobility data
to pinpoint underserved areas and identi-
fy burgeoning potential for new markets or
services. This data can also help automakers
and municipalities identify optimal locations
for charging stations or micromobility docks,
to effectively spread charging capacity and
achieve optimal utilization. Richer mobility
data can also lead to improved regulatory
efforts. For instance, data insights can help in-
form regulations for autonomous vehicles and
ride-sharing frameworks, and potentially be-
come the catalyst for standardizing data-shar-
ing protocols among mobility providers. As the
investment in and availability of mobility data
continue to grow and the sources for collect-
ing the data expand, a future where mobility
solutions are seamless, efficient, and tailored
to individual needs will start looking like the
norm.
Relying on ADAS
As the auto industry continues to push toward
fully autonomous self-driving, advanced driver
assistance systems (ADAS) are giving drivers
a glimpse of what that future might look like.
As drones, cameras, sensors, meteorological
instruments, particle detectors, mobile net-
working, and high precision GPS technology
continue to improve, expect to see the MWS
become more ubiquitous across the US.
Mobility Superapps
Superapps are emerging as a powerful
force, amalgamating various transportation
services into a singular mobile platform,
and subtly changing the mobility landscape
for governments, businesses, and con-
sumers alike. They embody a convergence
where ride-hailing, public transit, and even
non-transportation services harmonize,
offering a seamless user experience. Lead-
ing this transition are companies like Uber
and Grab. For example, Uber’s ambition
extends beyond ride-hailing; in the UK, its
app integrates bikes, scooters, trains, buses,
and even planes, aspiring to set a precedent
in global mobility solutions. They’re not just
about consolidating services for consumer
convenience but also about orchestrating a
smarter, sustainable urban mobility frame-
work. And with every transaction, Uber gath-

TECH
New vehicles are increasingly coming with
automated technology to monitor blind spots,
stay within lanes, parallel park, and trigger
automatic emergency braking (AEB). The driv-
er’s experience is evolving, with ADAS reduc-
ing the manual load, creating a more relaxed
yet controlled driving environment. Estimates
suggest that ADAS implementation has the
potential to prevent up to 250,000 deaths
between 2021 and 2050, not only saving lives
but likely also reducing insurance and health
care costs . On the regulatory front, regulators
are already proposing stricter requirements
for AEB to mitigate high-speed collisions
and better protect pedestrians and drivers.
Looking ahead, ADAS applications will con-
tinue to expand. Advancements like adaptive
cruise control and intersection assist can
further mitigate risks and improve traffic
flow. Several collaborations are taking place
across the industry to make these systems a
reality. Porsche and Mobileye, for example, are
pushing the boundaries, developing systems
capable of full collision avoidance and other
sophisticated functionalities. Such advance-
ments portend a future where features of
These real-world incidents have spurred a
mix of public, regulatory, and commercial
responses. In California, legislation requiring
safety operators on autonomous trucks was
a notable step toward ensuring safer road in-
teractions, but in San Francisco, the expan-
sion of driverless taxi services has prompted
protests by city officials and civic groups.
Commercial entities are also pitching in;
Cruise officials in Austin have been training
first responders to foster safer interactions
with their AVs. As AVs and micromobility
solutions continue to meld with urban mo-
bility, addressing pedestrian safety con-
cerns through a combination of legislation,
community engagement, and technological
advancements will be paramount.
AV Viability
Autonomous vehicles are making major
strides in capability and level of adoption,
but full self-driving still remains far out of
reach. Most come standard with autono-
mous features that keep them in lanes and
cause them to brake automatically, and
features that allow automobiles to auton-
omously change speeds, change lanes, and
take advanced actions like parallel park are
becoming increasingly common. Mercedes
has begun to test Level 3 autonomous driving,
and more major auto manufacturers are slat-
ed to begin testing their Level 3 vehicles in the
coming years. Some AV software developers
have been testing robotaxi services in select
municipalities, with some going as high as
Level 4 autonomy. Meanwhile, several auto
manufacturers have started adding features
and services to their vehicles to facilitate pro-
ductivity and entertainment for passengers
in anticipation of autonomous driving freeing
up drivers’ attention. Yet several major hurdles
remain for Level 5 autonomy. Some friendly
legislators have resisted pressure from stake-
holders like truckers and pedestrians, giving
AV manufacturers some breathing room to
begin testing and rolling out their vehicles.
But federal legislation to address the topic
has been stuck in Congress for six years with
no signs of life. With challenges that include
safety concerns, cybersecurity threats, insur-
ance liability risks, and questions regarding
infrastructure readiness, privacy protection,
these kinds are the standard and hopefully
driving is safer, more accessible, and eco-
nomically beneficial as a result.
Pedestrian Concerns
The advent of autonomous vehicles (AVs)
and micromobility solutions like e-scoot-
ers and e-bikes in urban landscapes has
triggered a cascade of pedestrian concerns.
In Austin, Texas, complaints are rife about
dangerous encounters with AVs, one nota-
ble incident being a Cruise vehicle veering
off-road into a small building. Similarly, in
San Francisco, robotaxis have been report-
ed blocking traffic, obstructing emergency
vehicles, and causing nuisances. The micro-
mobility sphere isn’t devoid of issues either;
e-scooters and e-bikes have been associ-
ated with injuries, like in incidents report-
ed where riders navigate recklessly amid
pedestrian traffic or lose control, leading to
accidents. Research from the US Consum-
er Product Safety Commission shows that
micromobility incidents increased by 21%
year over year in 2022, as e-bikes, e-scoot-
ers, and hoverboards grew more popular.

and ethical frameworks to dictate how AVs
make decisions in trade-off situations, full
implementation faces many major hurdles.
That’s why Ford, for example, stopped devel-
oping full self-driving in favor of nearer-term
goals like Level 3 and Level 4 autonomy.
What’s clear is that whether it’s full self-driv-
ing or not, autonomy is here to stay.
Local AV Regulations
As vehicles become more sophisticated, add-
ing semi- or full-autonomous features, local
governments must determine when and how
to authorize their use within city limits. Many
local governments and business communi-
ties eye autonomous vehicles as an engine
for economic growth. Cities and states are
exploring new ways to integrate autonomous
vehicles into their longer-term planning. Phoe-
nix and Los Angeles, among other cities, are
developing innovative approaches to design-
ing, building and testing autonomous vehicle
systems. Several states and cities are consid-
ering legislation to help bring the technology
to market. But regulations intended to spur
development could hit a roadblock—city and
miles, the initial results are encouraging.
Yet, challenges persist. Cruise, projected
to incur over $2 billion in losses in 2023,
has been striving to extend service hours
and geographic coverage to bolster revenue
and meet the high demand. The rollout of
robotaxis in Texas and California has also
spurred public protests, especially follow-
ing a series of incidents relating to traffic
congestion and safety. And despite a string
of approvals to expand service in California,
the state’s DMV recently moved to suspend
Cruise’s permits following a number of new
incidents. Concurrently, a lawsuit challeng-
ing the regulatory body overseeing robotaxi
deployments in the state has started to gain
traction, and federal authorities initiated a
probe into Cruise’s safety practices during
its operational rollout. To achieve broader
public acceptance, autonomous vehicles will
need to demonstrate markedly better safety
outcomes when compared to human drivers.
The enthusiasm from manufacturers and
municipalities to expedite robotaxi adoption
globally is palpable, yet public apprehen-
sions and the accident incidence rate during
the testing phase will influence whether the
pioneering companies can achieve substan-
tial testing benchmarks to draw meaningful
comparisons with human drivers, and the
time frame required to reach this goal.
state governments control their local streets,
but the federal government regulates sur-
rounding highways; for autonomous vehicles
to become ubiquitous and practical, Ameri-
ca’s roads would need to be interoperable.
Robotaxi Growth
Robotaxis are undergoing significant test-
ing and deployment across numerous US
cities, with 23 states having enacted laws
permitting companies to do so. Moreover,
at least eight other countries in Europe,
Asia, and the Middle East have embraced
the robotaxi revolution through testing and
operations. Operating up to Level 4 auton-
omy, robotaxis primarily function within
geofenced urban areas that are well-mapped,
characterized by slower speed limits, and
highly trafficked, which provides ample
data for system refinement. Robotaxis have
now amassed several million miles, and
over 100,000 users wanting to experience a
robotaxi service are still on Waymo’s waitlist
alone. Although assertions that robotaxis are
safer than human-driven vehicles may be
premature pending more extensive service

SCENARIOS
SCENARIO YEAR 2028
Personal Everything Mobility Platforms
While superapps were ubiquitous in Asian countries as early as the 2010s, they didn’t gain momentum in North
America until the mid-2020s. The movement began when Elon Musk pushed for X to become the everything
platform, but his vision did not come to fruition after the company failed due to advertising conflicts and us-
ers abandoning the platform. Instead, where everything platforms began to gain traction was in the travel and
mobility industries. While there’s still competition among these platforms to avoid antitrust laws, one particu-
lar platform has risen to prominence: OmniMoble, which successfully aggregated many major touchpoints and
access points for an individual’s comprehensive mobility needs. By fully integrating into users’ schedules, the
platform removes most frictions, frustrations, and hurdles that present themselves in the tediousness of every-
day life. Now, when a user is told to go on a business trip, OmniMoble will recommend, unprompted, available
livery, flight, and hotel options, accounting for both her personal preferences and company allowances. Once
she’s presented with options meeting those criteria, the user can make her selections with minimal effort. On
the day of travel, she boards her flight with ease. However, while on her flight, she realizes she didn’t pack a belt
for her business attire. From the plane, she consults OmniMoble, and it provides available options closest to her
hotel from her favorite designers. Before landing, the belt is delivered to the hotel.

ROBOTICS
DRONES
TRENDS

COBOTS BECOME COWORKERS
TECH
Accelerated Adoption
Cobots serve to supplement or even replace
workforces, especially those constrained by
labor shortages. This evolution is already
happening in Japan, which is in the midst
of a significant demographic shift, as the
working-age population begins to decline. The
resulting extreme labor shortages are caus-
ing industries in Japan to turn to increased
cobot usage, but the country is also able to
undertake this endeavor because of a highly
computer-literate workforce. With a strong
knowledge base for mechanical and control
systems and IT skills to manage the cobots,
companies are supplementing more work-
places with these bots—and also improving
productivity in the process. In one example,
Fujita Works has successfully incorporated
cobots in welding processes, reducing the
time required to master welding techniques.
In a broader geographic lens, cobots are
expected to have a tremendous impact on
the future of work. According to Grand View
Research, the collaborative robot market is
expected to grow by more than 30% by 2030,
reaching a value of $11.04 billion. Cobot adop-
tion is also being driven by major players, as
Amazon seeks to expand its robotics oper-
ations at fulfillment centers with updated
sorting machines, robotic arms, and mobile
robots. This new system, Sequoia, is intend-
ed to work in collaboration with humans and
is expected to increase delivery fulfillment
speed by 25%. Amazon contends that this
increase in speed will not be at the cost of
eliminating humans from the workforce.
Whether or not that turns out to be true, the
inclusion of new bots in the workforce will
likely have at least one positive outcome for
human workers: They increase safety and
reduce human injuries.
General Purpose Robots
Versatile, general purpose robots are the holy
grail of robotics, promising bots that do not
have to be limited or pigeonholed into single
categories but can be used for vast and di-
verse purposes without requiring extensive
calibration between disparate tasks. Sev-
eral robots are in the works that may bring
this kind of general purpose robot closer to
reality. One, from startup Figure, is a versa-
A “Digit” robot working in an Amazon fulfillment center.
Source: Amazon

TECH
tile, bipedal humanoid robot that can per-
form a range of tasks, from manual labor to
eldercare. Figure has raised $100 million and
hired top talent from leading tech companies,
including Boston Dynamics, Apple, Google,
and Tesla. Its plan includes an unveiling of
the robot this year, with a starting focus on
warehouse and retail applications and poten-
tially a robotics-as-a-service (RaaS) leasing
model. Tech startup Sanctuary AI is pursuing
a similar model, and has created a humanoid
robot named Phoenix that can perform a va-
riety of workplace tasks. Standing at 5’7” and
weighing 155 pounds, Phoenix is equipped
with advanced sensors and human-like
hands with haptic sensors that enable it to
carry out precise tasks. The robot is powered
by Sanctuary’s AI control system, Carbon,
which can be trained to learn new tasks either
by simulation or human demonstration. But
in order for general purpose robots to suc-
ceed, real-world and simulated training data
is necessary to teach bots to adapt to various
tasks. RoboCat, a self-improving AI agent for
robotics, attempts to do just that. In as few
as 100 demonstrations, it can operate various
robot, is bridging the emotional gap for
seniors, offering friendly interaction and
ensuring their safety. The robot proactively
engages in conversation, offers medication
reminders and issues emergency alerts;
these aren’t just features but a leap toward
combating loneliness among the elderly.
Robosen’s Grimlock transformer toy similar-
ly provides companionship, but for kids and
young adults. Their toy not only transforms
but is capable of understanding a variety
of commands and can communicate with
those who play with it. In what is maybe the
most extreme example of robotic compan-
ionship, individuals known as iDollators
are even using AI-augmented synthetic
dolls to form sexual relationships at home.
And with new ideas like Bopeep’s robotic
furniture—carefully designed to overcome
robots’ typically overbearing presence—or old
ideas brought to life like Prosper Robotics’
robot butler, efforts are underway to blend
household capabilities and companionship
together. The fusion of AI with robotics is
blurring the lines between the mechanical
and the emotional, pushing the boundaries
of what robots can offer in the home. As tech-
nology continues to evolve, the role of robots
is set to expand, heralding a future where our
domestic companions are not just helpers but
friends who share in our daily lives.
Robots Coexisting with Creative Applications
The field of robotics continues to demonstrate
how possibilities are endless for application.
In Switzerland, AI-powered robots are now
acting as security guards where their sur-
veillance capabilities and real-time response
mechanisms are improving safety without
the need for human intervention. South Korea
witnessed a remarkable blend of art and
robotics with EveR 6, a robot conductor that
fuses music and technology, broadening the
horizon of what robots can achieve in what
are typically viewed as human-centric creative
domains. In the US, robots designed to handle
hazardous materials were able to execute a
mission to safely dispose of chemical weap-
ons, reflecting a crucial application in high
risk environments. In education, hundreds of
kindergarten classes are now using a small
robot named KeeKo, which tells stories, poses
robotic arms and learn to perform different
tasks. RoboCat’s approach accelerates robot-
ics research by reducing the need for human
supervision, bringing us closer to versatile,
general purpose robots.
Robots in the Home
The landscape of domestic robotics is
transforming homes into hubs of automa-
tion, easing daily chores and offering com-
panionship. For instance, robotic vacuum
cleaners have become household staples,
tirelessly navigating living spaces to keep
them dirt-free. Similarly, robotic lawn mow-
ers now keep lawns well-trimmed without
the sweat, while robotic pool cleaners keep
swimming areas sparkling day or night. As
artificial intelligence has permeated these
robotic devices, their ability to recognize
voice commands, integrate with smart home
ecosystems, and operate with minimal
human intervention has made a significant
stride for at-home robotics and automa-
tion. More recently, however, robots at home
are becoming hubs for companionship. For
instance, ElliQ, an AI-powered co

Tech Trends Report 2024 Future Today Institute.pdf

More Related Content

Similar to Tech Trends Report 2024 Future Today Institute.pdf (20)

Recently uploaded (20)

Tech Trends Report 2024 Future Today Institute.pdf