AI: LAMDA, Singularity, Opportunity, Network Effects, Innovation and Responsibility

AI has really been hitting the headlines this quarter. From AI Generated Art going viral, The launch of Google's new AI Test Kitchen App, to a flurry of major funding rounds turning start up's into multi $bn enterprises, its easy to see the excitement gaining in the space.

Useful AI is starting to become more accessible as a service, and more worthwhile for common use-cases providing good return on investment, moving it from R&D pipe-dream into mainstream strategic focus. Today, however, while we are still in the very early stages of its development and becoming main-stream, I can see this quickly accelerating leveraging network effects and innovations in compute creating a world where first mover advantage quickly reaps the rewards of its investment. While holding great promise there is also great concern from some of the more taboo or accidental applications of AI in a world with a translucent approach to self-regulation and oversight.

This post discusses the near-stage potential for Commercial AI, how we are quickly accelerating towards singularity and why firms are investing now. It goes on to reflect on the horizontal integration opportunities for commercial AI, and a personal view on the need for oversight and regulation in certain areas.

Recent headlines: The path toward sentient AI (Singularity 2029)

In recent days Google released access to its new language generator - AI Test Kitchen, powered by LAMDA. The Conversational AI made famous recently when thought to have passed the Turing test. That essentially means when speaking with the AI it felt as if you were speaking with a human being.

Google propose that "As AI technologies continue to advance, they have the potential to unlock new experiences that support more natural human-computer interactions. We (Google) see a future where you can find the information you’re looking for in the same conversational way you speak to friends and family."

This concept stokes all kinds of conversations and opportunities. Just this month the WHO and partners launched the Worlds most extensive freely accessible AI health worker offering the ability for centrilised knowledge to be democratised globally, providing support to healthcare systems under-strain.

As AI becomes more immersive and human-like, and as VR becomes hyper-realistic does this present the opportunity to solve areas of Loneliness pandemic which is recently cited as being worse for aging than smoking and drinking. In fact 1.4 million older people in the UK often feel lonely, and in a survey in 2019, 22% of millenials stated that they had "no friends". The concept stokes mixed feelings, however could provide an element of a support ecosystem to those in need and I am excited to see the potential.

The future: Going beyond "Human-Computer Interaction" Into "Human-AI Augmentation"

Going further than just interacting with AI, Ray Kurzweil talks about AI Singularity, where "By 2029, computers will have human-level intelligence". Kurzweil describes the 'Singularity', as the point at which AI becomes so advanced that it is unrecognisable from a Human. He goes on to predict that by 2040 we may be able to upload a human brain into a computer in real time, requiring not only the Compute power but also bandwidth, roughly 1019 calculations per second and 1018bits of memory. Kurzweil believes that rather than an instantaneous scan and conversion to digital form, humans will most likely experience gradual conversion as portions of their brain are augmented with neural implants, increasing their proportion of non-biological intelligence slowly over time.

Within our lifetimes there could be the potential to be augmented by the power produced from AI to supercharge our existing brainpower and provide access to external data with search and compute to produce exponential power.

Now whether you would want to try that out or now, it presents an incredible future that is just 20 years away. As with our parents who together have lived in a non-digital and digital age, imagine a future where our children grow up in the augmented AI Age? This stokes all kind of questions and conversations and starts to create the case for investment now in the AI space.

Commercial AI Today: How The power of Network economics is driving huge start-up valuations

Rewind back to today and its evident to see the growing mainstream acceptance of AI within the VC and M&A Space. Just this week (17th October) we have seen some monumental investments and valuations;

• OpenAI, whose text- and image-generating artificial intelligence has become a mainstream hit is valued at nearly $20 Billion and is now in Advanced Talks with Microsoft For More Funding.
• AI content platform Jasper just raised $125M at a $1.5B valuation. A fantastic mass-market application of a specific use-case; Content Creation, and a brilliant marketing strategy has seen Jasper grow and grow. Its underlying technology and client base promises to pay dividends as it expands its product feature set.
• Owner of popular text-to-image platform Diffusion, Stability AI, has just announced $101 Million in Funding for Open-Source Artificial Intelligence, giving it a $1bn valuation. Stabiliy AI is building open AI tools that allow users and developers to access and build with their tools and ecosystem.

All three valuations share a common theme, the recognition for the power of AI to drive exponential future returns with an immediate commercial opportunity. Where as OpenAI and StabilityAI are making bets on providing the Ecosystem for inventors, designers and developers, Jasper proves the case for taking AI to market and a compelling product and future vision (essentially proving the use cast for OpenAI and StabilityAI).

When looking at the commercial potential of AI, its important to reflect on what has made of of the worlds most valuable firms today so successful. In the past 10 years we have seen the rise of platform businesses like Social Networks, e-Commerce conglomerates and Media Outlets use data to design and power products, use digital channels to provide global access and then use the network effects to scale and drive margin. (FB, Tiktok, Insta, Apple, Amazon, Netflix, Google...)

Bolt-on AI to those business models and the power of those network effects become exponentially powerful as we have to seen from some of the algorithms (good and bad) across social platforms and media platforms.

The power of network effects when adding in AI and scalable computing power becomes incredible.

As AI businesses like Jasper develop, not only do they have the data at the heart, a strong product/market fit, product feature set and roadmap... they also have an AI engine and limitless compute power to scale and innovate, providing an intelligence and audience platform for future pivots. Limitless leverage. It's not hard to see why AI is the new Data when it comes to market valuations and if there was ever a time to invest in a business case for AI-driven innovation within your firm, the time is certainly now.

Final thoughts on Governance and responsibility

While this story presents an aspirational dream and exciting future, for individuals, firms and investors, there are some major factors around AI which are less in focus however present real risks. While Governance around Bias and use-case for AI has been main stream, there is still major work to do around an approach to oversight and control which balances open-source innovation, security and stability.

Historically we have seen great pivots from businesses which explore new use cases for their existing firms resources to launch successful new businesses, for example Car firms launching financial products, and Facebook launching a marketplace.

Where Horizontal Integration comes to AI, the outcomes could be quite daunting and this is a key area I would like to see researched when it comes to oversight, responsible AI and governance.

For example;

AI + CRISPR - We can now give AI two topics and it can create pictures (Generative AI). In the future it would be possible to ask an AI to combine two viruses and with CRISPR, create a super virus. A scary prospect indeed.

AI + Dark Web - The world already has a number of Dark Webs and crime networks. Imagine a world where providers on the Dark Web use AI tools to decrypt personal information and clone any identity? And some banks still hand out leaflets to protect your pin...

AI + Deep Fake + Identity - As Deep-fake technology continues to become more realistic powered by AI, and were Zuck's dream to come true and we exist in the metaverse more often in the next 20 years, what's stopping the creation of an identical Joel Spence, doing all kind of things I knew nothing about? Including potential identity theft and fraud.

Let'ss not get started on the use of AI in the defense space...

Conclusion: Invest and experiment now, and support a vision for responsible and governed AI.

AI provides access the potentially unlimited leverage, powerful network effects and product features which were never thought to be so accessible and cheap to deliver. Invest now to be sure to take seriously the opportunity presented by experimenting with AI within your organisation. Explore how your firm may benefit from leveraging existing use cases, but invest in creative future thinking and strategic planning to explore where AI could monetise data sets and customer use cases within your domain. Successful application will be a major driver of market valuation as we go forward, but do not underestimate the focus AI will receive from regulatory bodies as it becomes more and more mainstream.

I hope you enjoyed reading this thought piece. If you'd like to connect on the subject or how to make AI a reality within your business please feel free to reach out.

Joel Spence

Like this content?

Check out my full blog at - https://wechoosethemoon.co.uk/

Views are my own and not that of my employer. Credits tagged in Links