Solve for X with AI: a VC view of the Machine Learning & AI landscape
•
571 likes•374,903 views
What you'll get from this deck 1. The M&A race for AI: by the numbers 2. Watch out! hype ahead: definitions & disclaimers 3. Machine Learning drivers: why is Machine Learning a ‘thing’ now (vs before) 4. Venture Capital: forming an industry, the AI/ML landscape 5. The One Hundred (+13) AI startups to watch in the Enterprise 6. The great Enterprise pivot: applying Machine Learning at scale 7. - where to go next -
Report
Share
More Related Content
What's hot
What's hot (20)
Viewers also liked
Viewers also liked (14)
Similar to Solve for X with AI: a VC view of the Machine Learning & AI landscape
Similar to Solve for X with AI: a VC view of the Machine Learning & AI landscape (20)
More from Ed Fernandez
More from Ed Fernandez (13)
Recently uploaded
Recently uploaded (20)
Solve for X with AI: a VC view of the Machine Learning & AI landscape
- 1. Solve for X with AI: a VC view of the machine learning & AI landscape Ed Fernandez @efernandez
- 2. Mentor, Advisor at Singularity University & Berkeley’s Center for Entrepreneurship & Technology Early Stage & Start-Up VC at Naiss.io - VC boutique/Palo Alto Investor/board director @ BigML inc (MLaaS: Machine Learning as a Service) Former corporate EVP at BlackBerry & Nokia @efernandez ed@naiss.io +15614104388
- 3. What you’ll get from this deck 1. The M&A race for AI: by the numbers 2. Watch out! hype ahead 3. Machine Learning drivers: why is Machine Learning a ‘thing’ now (vs before) 4. Venture Capital: forming an industry, the AI/ML landscape 5. The One Hundred (+13) AI startups to watch in the Enterprise 6. The great Enterprise pivot: applying Machine Learning at scale 7. - where to go next -
- 4. Definitions & Disclaimer Machine Learning is NOT Deep Learning NOR AI or AGI ML is here AI: much of the data in these slides Deep Learning
- 5. By number of deals, quarterly www.cbinsights.comhttps://www.cbinsights.com/blog/top-acquirers-ai-startups-ma-timeline/ Google is the most active acquirer of AI startups, having acquired 11 startups since 2012. Apple, which has been ramping up its M&A efforts, ranked second with 7 acquisitions under its belt. Newer entrants in the race include Ford, which acquired Argo AI for $1B in Q1’17, cybersecurity company Sophos, and Amazon. 200+ Acquisitions since 2012 30+ M&A deals in Q1’17 11 Acquisitions by Google The M&A race for AI
- 6. latest update The M&A race for AI September 8th Update - CBinsights : There were 85 disclosed M&A deals targeting AI startups in 2017 year-to-date. This is more than the 75 we saw in 2016. Includes Facebook’s acquisition of Ozlo and Nasdaq acquisition of eVestment ($705M). John Deere acquired agricultural tech company Blue River Technology for $305M.
- 7. Entering the second wave of acquisitions www.cbinsights.com 11https://www.cbinsights.com/blog/top-acquirers-ai-startups-ma-timeline/ Google is the most active acquirer of AI startups, having acquired 11 startups since 2012. Apple, which has been ramping up its M&A efforts, ranked second with 7 acquisitions under its belt. Newer entrants in the race include Ford, which acquired Argo AI for $1B in Q1’17, cybersecurity company Sophos, and Amazon. 200+ Acquisitions since 2012 30+ M&A deals in Q1’17 11 Acquisitions by Google The M&A race for AI 1st Wave - Tech giants: Google, Facebook, Twitter, Apple, Intel, Microsoft, IBM, Yahoo, eBay Entering into the 2nd wave - now: John Deere, General Electric, Ford, Samsung, Uber, Oracle, Sophos, Meltwater
- 8. Machine Learning Emerging Technology hype cycle: Machine Learning Watch out! Hype ahead #GartnerSYM 16 CONFIDENTIAL AND PROPRIETARY | © 2015 Gartner, Inc. and/or its affiliates. All rights reserved. Gartner and ITxpo are registered trademarks of Gartner, Inc. or it's affiliates. Hype Cycle for Emerging Technologies, 2016 From "Hype Cycle for Emerging Technologies, 2016," 19 July 2016 (G00299893) Innovation Trigger Peak of Inflated Expectations Trough of Disillusionment Slope of Enlightenment Plateau of Productivity time expectations Years to mainstream adoption: less than 2 years 2 to 5 years 5 to 10 years more than 10 years obsolete before plateau As of July 2016 Smart Dust General-Purpose Machine Intelligence 802.11ax Context Brokering Neuromorphic Hardware Data Broker PaaS (dbrPaaS) Quantum Computing Human Augmentation Personal Analytics Smart Workspace Volumetric Displays Brain-Computer Interface Virtual Personal Assistants Smart Data Discovery Commercial UAVs (Drones) IoT Platform Affective Computing Gesture Control Devices Micro Data Centers Smart Robots Distributed Ledgers Connected Home Cognitive Expert Advisors Machine Learning Software-Defined Security Autonomous Vehicles Nanotube Electronics Software-Defined Anything (SDx) Natural-Language Question Answering Enterprise Taxonomy and Ontology Management Augmented Reality Virtual Reality We are here 2 years to mainstream
- 9. First movers unfair advantage
- 10. Content 1. The M&A race for AI: by the numbers 2. Watch out! hype ahead 3. Machine Learning drivers: why is ML changing everything 4. Venture Capital: and the AI/ML landscape 5. The One Hundred (& Thirteen) AI startups to watch in the Enterprise 6. The great Enterprise pivot: applying Machine Learning at scale 7. - where to go next -
- 11. Machine Learning why now The perfect storm Value Creation 01 02 Data Algorithms 03 04 Hardware Talent (humans) and tools (for humans)
- 13. Machine Learning drivers: Data: massive datasets, ‘dark’ data, crowd source and open source data 01 02 Data Algorithms 03 04 Data growth: From 8,5 EXAbytes in 2015 to 40K EXAbytes in 2020 = 40 trillion GB 15K EXAbytes in the cloud by 2020 = 37% Kryders law: storage density doubles every 18 months (driven by cloud) 5G access accelerates mobile data & video Unlocking ‘dark’ data & data silos in corporations
- 14. Machine Learning drivers Algorithms 01 02 Data Algorithms Widespread adoption of machine learning algorithms • ML as a Service • APIs • Tools and open source libraries & ML frameworks Faster hardware acceleration Better input & more data Neuroscience driving new algorithms
- 15. Content 1. The M&A race for AI: by the numbers 2. Watch out! hype ahead: definitions & disclaimers 3. Machine Learning drivers: why is this ML revolution happening 4. Venture Capital: and the AI/ML landscape 5. The One Hundred (& Thirteen) AI startups to watch in the Enterprise 6. The great Enterprise pivot: applying Machine Learning at scale 7. - where to go next -
- 16. 46% of AI acquired companies are VC backed Total # of funding rounds/ deals grew 4,6x from 150 in 2012 to 698 in 2016 245 funding deals in Q1 2017 for a total of $1,73 Bn Nearly 48% in seed/angel stage (new startups) Financing rounds Venture Capital - Machine Learning/AI Q1’17 MOST ACTIVE QUARTER FOR AI STARTUPS Before the close of Q1’17 (as of 3/23/17) AI startups received 245 deals and $1.7B in funding. Nearly 48% of the deals in Q1’17 were in the seed/angel stage, indicating newer companies are continuing to enter the space. www.cbinsights.com 28 ARTIFICIAL INTELLIGENCE: QUARTERLY FUNDING Q1’12-Q1’17 (as of 3/23/2017)
- 17. ML is driving efficiencies, productivity and ROI for the enterprise Savings, labor cost & automation improvement Financing rounds: deal distribution by category, heat map Venture Capital - Machine Learning/AI Fintech & Insurance Healthcare Horizontal platforms/apps Commerce/ad BI/analytics IoT
- 18. Content 1. The M&A race for AI: by the numbers 2. Watch out! hype ahead: definitions & disclaimers 3. Machine Learning drivers: why is this ML revolution happening 4. Venture Capital: and the AI/ML landscape 5. The One Hundred (+13) AI startups to watch in the Enterprise 6. The great Enterprise pivot: applying Machine Learning at scale 7. - where to go next -
- 19. Big players AMZ, Google, MS, IBM trying to drive cloud and infrastructure by offering ML in the cloud as part of wider portfolio. Lack of focus and customer orientation, ‘small’ market <$1Bn Wrong business models, charging by prediction, black box models, can’t be exported. Greenfield for Startups Enterprise AI start-ups to watch Business Intelligence Customer Management Finance & Operations Industrials & Manufacturing Consumer Marketing Digital Commerce B2B Sales & Marketing Productivity Engineering Security & Risk Data Science Enterprise AI Companies Presented by HR & Talent BigML DataRobot H2O Dataiku Google ML API Amazon ML Microsoft Azure ML IBM MLaaS vs Cloud Wars - MLaaS: Machine Learning as a Service http://www.topbots.com/essential-landscape-overview-enterprise-artificial-intelligence/
- 20. Technology stack: startups to watch Machine Learning technology stack
- 21. The Great Pivot - ML platform revolution Systems of Intelligence/ML drive efficiencies (1st), competitive advantages (2nd) & next defensible business models ultimately • Most large technology companies are reconfiguring themselves around ML. • Google was (arguably) the first company to move, followed by Microsoft, Facebook, Amazon, Apple and IBM. • 2nd tier corporations following suit: GE, Uber, even carriers as AT&T • Not only a US phenomena - Alibaba, Baidu chief Robin Li said in an internal memo that Baidu’s strategic future relies on AI • Ultimately all global players will need to re- tool their processes adopting a ML driven approach. h/t Jerry Chen - Greylock Partners https://news.greylock.com/the-new-moats-53f61aeac2d9
- 23. Fast-Forward to 2017: MWC - 4YFN MWC - 4YFN: Mobile World Congress - 4 Years From Now
- 24. Content 1. The M&A race for AI: by the numbers 2. Watch out! hype ahead: definitions & disclaimers 3. Machine Learning drivers: why is this ML revolution happening 4. Venture Capital: and the AI/ML landscape 5. The One Hundred (+13) AI startups to watch in the Enterprise 6. The Great Enterprise pivot: applying Machine Learning at scale 7. - where to go next -
- 25. Machine Learning Adoption Applying machine learning at scale We are here h/t BigML inc
- 26. Amazon Jeff Bezos’ letter to Amazon shareholders - May, 2017 “Machine learning and AI is a horizontal enabling layer. It will empower and improve every business, every government organization, every philanthropy — basically there’s no institution in the world that cannot be improved with machine learning” . Jeff Bezos
- 27. Google FBlearner Flow: Facebook’s ML platform for internal use - March, 2017 Google MLaaS was released in Beta to developers in 2016 Internal use since 2015
- 28. Facebook FBlearner Flow: Facebook’s ML platform for internal use - May, 2016 Facebook ML platform is used by more than 25% of its engineering team +1Mn ML models trained +6 Mn predictions/sec
- 29. The Great Pivot - ML platform revolution Systems of Intelligence/ML drive efficiencies (1st), competitive advantages (2nd) & next defensible business models ultimately h/t Jerry Chen - Greylock Partners https://news.greylock.com/the-new-moats-53f61aeac2d9
- 30. Content 1. The M&A race for AI: by the numbers 2. Watch out! hype ahead: definitions & disclaimers 3. Machine Learning drivers: why is this ML revolution happening 4. Venture Capital: and the AI/ML landscape 5. The One Hundred (+13) AI startups to watch in the Enterprise 6. The Great Enterprise pivot: applying Machine Learning at scale 7. - where to go next -
- 31. Where to go next http://www.pcmag.com/article/353293/7-tips-for-machine-learning-success A few tips for machine learning success • Focus on Features (vs Algorithms) • Same with Data (vs Algorithms): right data, clean data • Faster trial & error, rapid prototyping (vs Algorithms) • Use tools & ML platforms, cloud is friendly (Algorithms aren’t) • See next 5 min - DIY machine learning sales hack (and forget Algorithms)
- 32. √ Sales Hacking with Machine Learning DIY practical example: WHAT A practical sales hack using machine learning to identify & engage in real time your competitor’s unhappy users HOW • Twitter • Monkeylearn • Slack • Zapier REQUIREMENTS • a laptop with WiFi connectivity • 10 min of undivided attention time • Twitter, Monkeylearn, Slack & Zapier free accounts • Cup of coffee (to look cool while setting it up) Monitor mentions on competitors in social media Trigger: automatically analyze and classify mentions using machine learning and detect users complaining Alert & Action: notify sales team for real time action & engagement Automate process, set up rules, integrate services
- 33. Steps: 1. Create a Zap. 2. Select Twitter as Trigger App. 3. Select Search Mention as Trigger. 4. Input your competitor search query: trigger whenever someone mentions competitor. Type in: “[NameOfCompetitor] bad service filter:retweets” Filtering out tweets not referring to ‘bad service’ and retweets. Min 1: Monitor & Trigger
- 34. 4. Select MonkeyLearn as Action App. 5. Select Classify Text as Action. 6. Select a Sentiment Analysis Model. Classify your competitors mentions: Negative, Neutral or Positive tweets. You can use a pre-trained model or eventually train your own custom model. 7. Select text to classify (Tweet text) Min 3: Apply Machine Learning
- 35. Min 5: Filter & Trigger Alert & Action 8. Select Filter Action App. 9. Filter out Positive and Neutral tweets, only continue with negative 10. Select Slack Action App. A slack notification will arrive at the selected channel.
- 36. √ What the *heck* just happened Technical Debt - Legacy vs lean/API and cloud Corporate Startup IT infrastructure HW/SW provisioning $0 - cloud Integration 1 month, internal budget or 3rd party Zapier pro plan - $30/month Personnel Data Scientist - IT experts Part time Uber driver & developer Testing & deployment 1-3 months same day CAC inbound sales/ CRM Chatbot Monitor mentions on competitors in social media Trigger: automatically analyze and classify mentions using machine learning and detect users complaining Alert & Action: notify sales team for real time action & engagement Automate process, set up rules, integrate services