HfS Webinar Slides - A (much-needed) reality-check on Enterprise Artificial Intelligence
2 likes2,296 views
The document discusses a webinar focusing on the impact of artificial intelligence (AI) on enterprise operations, emphasizing the integration of technology and service providers to solve business challenges. It outlines the shifting landscape of AI within organizations, the forthcoming transitions in job roles due to automation, and highlights the importance of a data-centric strategy. Key lessons include the dangers of underestimating the complexity of machine learning projects and the necessity for a coherent data strategy to support AI initiatives.
HfS Webinar Slides - A (much-needed) reality-check on Enterprise Artificial Intelligence
- 1. Proprietary│Page 1© 2018 HfS Research Ltd. HfS Webinar: A Reality-check on Enterprise Artificial Intelligence (AI) Tom Reuner Managing Partner tom.reuner@hfsresearch.com April 12, 2018 Phil Fersht CEO and Chief Analyst phil.fersht@hfsresearch.com
- 2. Proprietary│Page 2© 2018 HfS Research Ltd. Questions • Attendees can submit questions throughout the webinar by typing it in the ‘Question panel’ in the GoToWebinar control panel. • All questions are submitted to the organizer and panelists. We will try to answer as many as we can during the webinar Recording and slides • The webinar recording and slides will be made available on our website. If you have registered for the webinar, you will receive an email when they are available.
- 3. Proprietary│Page 3© 2018 HfS Research Ltd. Phil Fersht, CEO and Chief Analyst, HfS Research
- 4. Proprietary│Page 4© 2018 HfS Research Ltd. Our panellists Jesus Mantas Global Head of Strategy and Offerings IBM Global Business Services Mike Salvino Managing Director, Carrick Capital Partners and Executive Chairman Infinia ML Phil Fersht CEO and Chief Analyst HfS Research Tom Reuner Managing Partner Business Operations Strategy HfS Research
- 5. Proprietary│Page 5© 2018 HfS Research Ltd. HfS Research… separates the wheat from the chaff
- 6. Proprietary│Page 6© 2018 HfS Research Ltd. The Six Value Change Agents Driving the Digital Operations Industry Success in the future will be determined by how well clients, techniology and service providers are able to combine the power of multiple change agents into integrated solutions that solve crucial business problems Source: HfS Research, 2018
- 7. Proprietary│Page 7© 2018 HfS Research Ltd.
- 8. Proprietary│Page 8© 2018 HfS Research Ltd. Q. What do you see as the primary benefits of breaking down barriers between front, middle and back office moving toward an operating framework like the OneOffice? C-Suite’s Desires from OneOffice reorganization: Better Data and Alignment of Operations to Business Outcomes Source: HfS Research 2018 Sample: C-Level Enterprise Executives = 100 9% 22% 19% 9% 31% 4% 4% 14% 6% 11% 15% 11% 34% 6% 12% 10% 10% 18% 7% 19% 10% 35% 38% 40% 42% 49% 57% Increase competitiveness in the wake of digi disruption Increased operational simplicity Greater efficiency /reduced cost Improved workplace culture Improved quality and speed of execution Stronger alignment of business operations to business outcomes Better data to drive the business forward Rank 1 Rank 2 Rank 3
- 9. Proprietary│Page 9© 2018 HfS Research Ltd. When do you believe AI automation to be applicable for you within the following processes? All AI Techniques & Solutions Are Getting Evaluated 37% 42% 37% 41% 45% 19% 18% 21% 23% 21% 23% 21% 24% 19% 22% NEURAL NETWORKS NATURAL LANGUAGE PROCESSING (NLP) COMPUTER VISION VIRTUAL AGENTS MACHINE LEARNING (ML) Piloted / implemented Evaluating In next 2 years Source: “State of Automation 2017” Sample: Enterprise Buyers = 400
- 10. Proprietary│Page 10© 2018 HfS Research Ltd. Over Half of Enterprises Bracing for Major Changes in Internal Roles Q. In terms of the number of transactional internal roles within the following process areas, what proportion do you expect to be significantly impacted by automation in the next 2 years? ( Average Across Functions) 4% 8% 11% 26% 31% 21% N/A Prefer not to say Under 10% 11-20% 21-50% 50%+ % Employees Impacted by Automation Source: HfS Research in Conjunction with KPMG, "State of Operations and Outsourcing 2018, March, 2018 Sample: Interim Enterprise Buyers = 250
- 11. Proprietary│Page 11© 2018 HfS Research Ltd. A digital labor strategy: more emphasis on the LABOR!
- 12. Proprietary│Page 12© 2018 HfS Research Ltd.
- 13. Proprietary│Page 13© 2018 HfS Research Ltd. Tom Reuner, Managing Partner, HfS Research Shaken, definitely stirred tom.reuner@hfsresearch.com @tom_reuner Overview ▪ Tom Reuner is Managing Partner, Business Operations Strategy and M&A Advisory. Tom is responsible for driving strategic engagements in business operations, IT as well as M&A advisory. The responsibilities range from research over consulting to business development. This involves advising clients on the formulation of strategies, guiding them through methodologies and the analysis of research findings, as well as interactive liaison with the client throughout the course of projects from initial meeting to conclusion. ▪ Tom is driving thought-leadership and frameworks across business operations and in particular Intelligent Automation and Artificial Intelligence. Automation cuts across the whole gamut ranging from RPA to Autonomics to Cognitive Computing and AI. This includes increasingly the intersections of unstructured data, analytics, and Cognitive Automation while mobilizing the HfS analysts to research Intelligent Automation dynamics across specific industries and business functions. ▪ Previous Experience ▪ Tom’s deep understanding of the dynamics of this market comes from having held senior positions with Gartner, Ovum and KPMG Consulting in the UK and with IDC in Germany. He is frequently quoted in the leading business and national press, appeared on TV and is a regular presenter at conferences. Education ▪ Tom has a PhD in History from the University of Göttingen in Germany.
- 14. Proprietary│Page 14© 2018 HfS Research Ltd. Key lessons learned Hype around chatbots is distorting the marketing communications The Enterprise AI market has a duplexity of approaches: Industrialization and project-centric Enterprise AI is still at the periphery of the enterprise or applied as a bolt-on The Holy Grail of AI is at the intersection of iterative data inputs and minimal training of algorithms
- 15. Proprietary│Page 15© 2018 HfS Research Ltd. The frontier in service delivery is at the intersection of automation, analytics and AI The report is focused on Artificial Intelligence (AI). Our definition of AI includes cognitive solutions. Forthcoming reports on RPA and Smart Analytics Definition of AI for the purpose of this study: AI aims to automate intelligent activities that humans associated with other human minds through a combination of reasoning, knowledge, planning, learning, natural language processing (communication), and perception (aka cognitive).
- 16. Proprietary│Page 16© 2018 HfS Research Ltd. The Enterprise AI Market Has a Duplexity of Approaches: Industrialization and Project-Centric Project-centric approaches are highly domain-specific and the strategic logic is 1-to-1. Examples include the automation of a medical coding in a hospital to support diagnosis and better patient records. Industrialization Project-Centric Service orchestration Horizontal Out-of-the-box Alignment with Intelligent Automation Service delivery Mega ISVs Data lake RPA, autonomics, chatbots Narrow AI Sub-sector lens Design Thinking Expansions analytics Data Science Proprietary IP and open source Data silos Proprietary algorithms, Deep Learning Strong AI Specific requirements Industrialization is all about finding as many commonalities across delivery backbones as possible in order to scale and save costs at the same time. The strategic logic is 1-to-many. Examples would be monitoring of infrastructure or self-remediation technologies including IPsoft and Arago.
- 17. Proprietary│Page 17© 2018 HfS Research Ltd. The journey toward AI has disparate starting points
- 18. Proprietary│Page 18© 2018 HfS Research Ltd. Just like Intelligent Automation, AI should be seen as continuum AI Neural Networks Autonomics Virtual Agent Machine Learning Image Recognition Machine Reasoning Natural Language Processing Chatbot Deep Learning Computer Vision Speech Recognition Knowledge Represen- tation
- 19. Proprietary│Page 19© 2018 HfS Research Ltd. AI Technology Partner Landscape Virtual Agents Neural Networks Machine/ Deep Learning Autonomics Computer Vision NLP AI Building Blocks
- 20. Proprietary│Page 20© 2018 HfS Research Ltd. Move toward AI will bring mega-ISVs to the fore AI Data Algorithms Platform Compute Google Tensor Processing Unit NVIDIA Volta SAP Leonardo Salesforce Einstein Google Neural Machine Translations Google WaveNet Google TensorFlow Amazon Machine Learning Google Cloud Machine Learning Microsoft Cognitive Services Oracle Data Cloud Oracle Adaptive Apps IBM Watson Data Insights IBM Watson API Explorer HIRO Knowledge Core HIRO Engine Wipro Holmes TCS ignio IBM Watson Knowledge Studio Google DeepMind Celaton Instream Loop AI Cortana Intelligent Services Azure Machine Learning SAP Data Hub AWS Public Datasets Infor Coleman Amazon Rekognition Amazon Lex IBM Watson Virtual Agents Intel Movidius Amazon Connect Fujitsu DLU AMD/GloFlo Oracle Intelligent Bots Intel Loihi Adobe Sensei Infosys Nia
- 21. Proprietary│Page 21© 2018 HfS Research Ltd. Moving toward a data-centric mindset necessitates new requirements for talent Data Data Scientist: • Cleaning, organizing data • Custom algorithms • Statistical modelling • Feature engineering • Exploratory analysis Data Engineer: • Designing, testing, maintaining scalable data architectures • Evaluation, integration tools • Data ingestion • Deployment • Solution architecture Artificial Intelligence Technologies: • Ingestion of data • Pattern analysis • Knowledge representation • Integration of disparate approaches
- 22. Proprietary│Page 22© 2018 HfS Research Ltd. The Holy Grail of AI is at the intersection of iterative data inputs and minimal training of algorithms Limited (training data) Complex (expanding sources and formats) Unsupervised LearningSupervised Learning Reinforcement Learning problem definition data selection model selection model training model improvement model deployment input data complexity necessary training of algorithms
- 23. Proprietary│Page 23© 2018 HfS Research Ltd. Innovative AI use cases Virtual Assistant integrated with Hadoop cluster: Big 4 for global bank, Hadoop cluster to feed every customer channel; sentiment analysis and real-time analysis. Example for scale and service orchestration; goes beyond chatbot hype Setup of AI CoE and Intel Nervana AI Academy. TCS provides a platform to connect researchers, developers, and startups. Example for ecosystem enablement Integration of disparate sources for General Ledger: Global SI. Clients can drop disparate information for General Ledger requirements; Machine Learning and other technology building blocks allow for seamless processing Cross-fertilization from other sectors: Global SI has helped Australian company to automatically identify telephone posts using Google Tensorflow and Street View replacing manual inspection; broad replicability, think insurance scenarios Leverage of ML for medical coding: Global SI (not Watson) helped European hospital to apply medical coding at scale to allow for digital patient record and diagnosis. Example for complex Data Science approach at scale for critical processes Exploring quantum computing techniques on AI and Machine Learning algorithms in Block Chain applications: Global SI for leading bank in Australia. Example for integrated approach of next-gen technologies and approaches
- 24. Proprietary│Page 24© 2018 HfS Research Ltd. HfS Blueprint: Enterprise AI Services 2018 WHAT THIS BLUEPRINT REPORT COVERS The pace of change driven by the onset of AI is nothing short of astounding. Startups continuously change perceptions of what best-of-breed might mean. At the same time, we see PoCs progressing to projects literally within a few months. Consequently, many boards are paranoid about the emerging notions of cognitive and AI, but all too often fail to turn those fears into actionable items. Against this background this report takes stock where the enterprise market for AI really is at. How is AI enabling organizations journey toward the OneOffice? A crucial element of this report is to play back the lessons learned from the early deployments. WHO SHOULD READ THIS REPORT Executive leaders and business unit leaders, outsourcing and procurement managers, advisors, investors who have responsibilities for innovation, digital transformation and for building out service delivery capabilities. SERVICE PROVIDERS WE DISCUSS Accenture, Atos, Capgemini, Cognizant, Deloitte, DXC Technology, EY, Genpact, HCL, IBM, Infosys, KPMG, LTI, PwC, Syntel, TCS, TechMahindra, Wipro Access the document (freely accessible for HfS premium subscribers)
- 25. Proprietary│Page 25© 2018 HfS Research Ltd. Our panellists Jesus Mantas Global Head of Strategy and Offerings IBM Global Business Services Mike Salvino Managing Director, Carrick Capital Partners and Executive Chairman Infinia ML Phil Fersht CEO and Chief Analyst HfS Research Tom Reuner Managing Partner Business Operations Strategy HfS Research
- 26. IBM Services Enterprise-Grade AI HfS Webinar Jesus Mantas | Managing Partner, Cognitive Assets and GBS Ventures Global Head of Strategy & Offerings, IBM Services Webinar | April 12, 2018
- 27. 2727 Trusted, Transparent & Auditable • Transparency of training • Ability to explain decisions • Auditability of recommendations Integrates with work flows and talent flows • Business Purpose • Secure • Bidirectional human- system design Learns more from less data • Limited, not semantically consistent • Thousands, not billions • Industry-context specific Enterprise-Grade Artificial Intelligence 2
- 28. 28 Data is the natural resource required for ML or AI to distill any outcomes
- 29. 2929 Risk & ComplianceCognitive Care Knowledge Worker Enterprise-Grade Artificial Intelligence Use Cases at Scale HR / Talent Global Financial Institution
- 30. 30 ©2018 IBM Corporation 20 April 2018 IBM Services30 ©2018 IBM Corporation 20 April 2018 IBM Services30 Bridging the Gap Between Different Approaches to AI Leveraging AI in an agile DevOps lifecycle to reduce cycle time, automate tasks, and reduce trouble tickets to build horizontal solutions Industrialization Cognitive Garages creates a rapid, scalable, and cost effective way to design, develop, test and deploy transformative AI-powered solutions Cognitive Garages Project-Centric
- 31. The ML Epidemic Epidemic: a widespread occurrence of a particular undesirable phenomenon The undesirable phenomenon here is that all of a sudden everyone knows how to do ML and is an Expert. (The ML Epidemic) Don’t Fuel the ML Epidemic!
- 32. ML Epidemic–Point #1: I have a TEAM that knows ML! Your team that knows how to do Simple Neural Networks…An example is converting English words to French…These successful simple projects will inspire enterprises to want more out of ML and this is when your ML team will STRUGGLE. Building Deep Learning Neural Networks (Multi-Layered) is not easy.
- 33. ML Epidemic–Point #2: The ML projects my team are performing will scale to achieve true business IMPACT! The AlphaGo breakthrough was great PR (there is now a documentary on Netflix) but provided little business impact. Make sure your projects are answering these 3 questions. NO SCIENCE EXPERIMENTS! 1. Does the project solve a top 1, 2, or 3 question that a CEO or Executive wants answered? 2. Does the project help your company to reduce costs or increase revenue? 3. Does the project create a unique data set for your company?
- 34. ML Epidemic–Point #3: My Data is READY to go! There is NO ML without Data. Salvino prediction: “Companies will spend as much if not more money dealing with Data in the next 5 years as they have spent implementing mission critical systems like SAP, etc.” 1. Accessible – Can your ML team actually access your data? 2. Clean – Is your data clean or is there “junk” in fields? 3. Data Sets – Have you created data sets or have you created data “swamps”? 4. Maintain – Do you have a process and team to maintain the data (Data Science Culture)? 5. Utilize – Do you have a strategy to answer question that make business impact?
- 35. How To Get Started
- 36. Scarcity of resources is real and expensive. NIPS was a recruiting event this year instead of a research conference. Most luminaries are not teaching any longer. They are tied up doing work for companies so new resources are not being created to keep up with demand. Proficiency is developed by doing years of research and most companies don’t have access to labs. How many ML experts have you met this year that were Cloud or Security experts last year? How to Get Started–Point #1: EVALUATE your ML talent 1. Advanced degree in relevant quantitative field (statistics, computer science, applied mathematics, etc.) 2. 7+ years experience in machine learning, data science, data engineering, and/or computational software development 3. 3+ years development in Python, including libraries such as NumPy, SciPy, pandas, TensorFlow, etc. 4. Experience with deep learning models, including CNN and RNN architectures 5. Experience working with large datasets, including NoSQL and relational databases 6. Experience with cloud computing
- 37. Big data, IoT, Analytics, Digital, etc. All good buzz words but it really is not that hard. Create Data Sets and a Data Science Culture. It is not GLAMOROUS work but it matters! How to Get Started–Point #2: Create a coherent data strategy across your data warehouses, lakes, rivers, streams, puddles, swamps…
- 38. Magicians vs. Aliens. Magicians want to work with other Magicians not folks that view them as Aliens. This is not inspiring to them. People leave People – they don’t leave Companies so ORGANIZE for success! How to Get Started–Point #3: CENTRALIZE the ML Function
- 39. Proprietary│Page 39© 2018 HfS Research Ltd.
- 40. Proprietary│Page 40© 2018 HfS Research Ltd. Q. What are your greatest challenges preventing you from achieving the OneOffice Concept? IT Lacks Talent, Business Lacks Mindset 12% 10% 23% 29% 21% 0% 23% 25% 6% 35% 8% 8% 6% 13% 6% 6% We’re held hostage by legacy technology Lack of talent internally Legacy thinking / lack of a “digital mindset” from IT Legacy thinking / lack of a “digital mindset” from biz functions We’re held hostage by legacy technology Lack of talent internally Legacy thinking / lack of a “digital mindset” from IT Legacy thinking / lack of a “digital mindset” from biz functions Rank 1 Rank 2 IT C-Suite Business C-Suite Source: HfS Research 2018 Sample: C-Level Enterprise Executives, Major Enterprises = 100
- 41. Proprietary│Page 41© 2018 HfS Research Ltd. 3% 1% 4% 2% 14% 10% 10% 6% 32% 18% 3% 4% 8% 7% 9% 13% 13% 14% 9% 22% 6% 9% 10% 13% 5% 8% 11% 21% 5% 10% 12% 14% 22% 22% 28% 31% 34% 41% 46% 50% Understanding business processes and using automation and AI to improve business performance Understanding / using digital and cloud technology to improve business performance, drive change Analytical prowess to improve operations / productivity Improving end-to-end processes across external and internal delivery Defining business outcomes Influencing senior executives Vision and ability to drive change Commercial acumen (balance process, technology and innovation decisions with sustainable cost model) Exploring new ways of partnering across the services ecosystem Creative, entrepreneurial spirit, Curiosity for innovation; First Second Third Focus is on the Right-brain, not the Left! Q. What are the top three workforce requirements required now? Creative thinkers to solve problems and design solutions Less critical skills shift – creates more appetite to outsource Source: HfS Research, 2018 “Intelligent Operations Study” conducted in association with Accenture Sample: Enterprise Buyers = 460
- 42. Proprietary│Page 42© 2018 HfS Research Ltd. RPA, Cloud & IoT Lead Investment Focus 16% 19% 33% 33% 33% 37% 42% 44% 53% 32% 37% 48% 40% 30% 41% 35% 36% 28% 22% 20% 14% 23% 20% 18% 18% 14% 14% 31% 24% 5% 4% 18% 4% 5% 6% 5% Driverless Vehicles Drones AI/ML/Cognitive Blockchain Virtual and Augmented Reality Analytics Internet of Things (IOT) Cloud RPA Significant investment/focus Some investment/focus Limited / Modest investment / focus No investment /focus Q. How much investment/focus is your organization making in the following in the next year to help you achieve operational cost saving goals? Source: HfS Research in Conjunction with KPMG, "State of Operations and Outsourcing 2018, March, 2018 Sample: Interim Enterprise Buyers = 250
- 43. Proprietary│Page 43© 2018 HfS Research Ltd. The HfS Mission & Vision: Defining Future Business Operations • HfS defines and visualizes the future of business operations across key industries with its OneOfficeTM Framework. • The HfS mission is to provide visionary insight into the major innovations impacting business operations: Automation, Artificial Intelligence, Blockchain, Internet of Things, Digital Business Models and Smart Analytics. • HfS influences the strategies of enterprise customers to develop OneOffice backbones to be competitive and partner with capable services providers, technology suppliers, and third party advisors.