David Silver

David Silver is a British computer scientist, born in 1976, and a professor of computer science at University College London. He is the founder and chief executive officer of Ineffable Intelligence, the British AI research startup he launched in late 2025, and was previously vice president of reinforcement-learning research at Google DeepMind, where he led the team behind AlphaGo, AlphaZero, AlphaStar, and AlphaProof. As of May 2026, he runs Ineffable Intelligence following its April 27, 2026 emergence from stealth with a $1.1 billion seed round at a $5.1 billion post-money valuation, the largest seed financing in European technology history, and continues his UCL professorship in parallel.

At a glance

• Education: Bachelor of Arts in computer science, Christ's College, University of Cambridge (1997, Addison-Wesley award); PhD in computer science, University of Alberta (2009), supervised by Richard Sutton, with a thesis on reinforcement learning and simulation-based search in the game of Go.
• Current roles: Founder and Chief Executive Officer of Ineffable Intelligence since late 2025; Professor of Computer Science at University College London since 2013.
• Key contributions: lead researcher on AlphaGo (2015 to 2016), AlphaZero (December 2017), AlphaStar (2019), and AlphaProof (July 2024); co-founder of Elixir Studios (1998 to 2005); UCL reinforcement-learning lecture series (2015).
• Awards: Royal Society University Research Fellowship (2011); ACM Prize in Computing (2019, joint with Aja Huang and Demis Hassabis); Fellow of the Royal Society (2021); AAAI Fellow (2022).
• Wikipedia: David Silver

Origins

Silver was born in 1976 in the United Kingdom. He read computer science at Christ's College, Cambridge from 1994 to 1997, graduating with a BA and the Addison-Wesley award. At Cambridge he befriended Demis Hassabis, the future co-founder of DeepMind, beginning a working relationship that would carry through the British games industry into the founding-era of DeepMind.

In 1998 Silver co-founded Elixir Studios in London with Hassabis, serving as chief technology officer and lead programmer. Elixir shipped Republic: The Revolution in 2003 and Evil Genius in 2004, both built around AI-driven simulation systems, before closing in 2005.

In 2004 Silver moved to Canada to begin doctoral work at the University of Alberta under Richard Sutton, described in field histories as the father of reinforcement learning. Silver's PhD thesis, completed in 2009, examined reinforcement learning and simulation-based search in computer Go, and produced master-level 9 by 9 Go programs in collaboration with Sylvain Gelly on the MoGo system.

Career

Silver began consulting for Google DeepMind in 2010, two years after the lab's founding, and was among the earliest research scientists at the company. The Royal Society awarded him a University Research Fellowship in 2011, and UCL appointed him to a lecturer position the same year, creating the dual academic-and-industrial appointment that has persisted since. He moved to a full-time research role at DeepMind in 2013.

The AlphaGo project, which Silver led at DeepMind, was formed around 2014 to test whether deep reinforcement learning could compete at Go. AlphaGo's October 2015 match against European champion Fan Hui ended 5 to 0; it was the first time a computer Go program had defeated a professional human player on a full-sized board without handicap. The system's Nature paper was published on January 28, 2016, with Silver as first author. In March 2016 AlphaGo defeated Lee Sedol, the 9-dan Korean professional, 4 games to 1 in Seoul. The match was watched by an estimated 200 million people online and is consistently cited as the public moment at which deep reinforcement learning crossed into mainstream awareness.

AlphaZero followed in December 2017, with Silver as first author on the preprint and the December 2018 Science paper. The system generalized the AlphaGo approach to chess, shogi, and Go starting from random initialization, with no human game data. AlphaZero is the practical demonstration of the broader thesis that runs through Silver's research career: that an artificial agent learning through self-play and reward signal can reach superhuman capability on tasks where human-generated training data is either unavailable or insufficient.

AlphaStar, the StarCraft II reinforcement-learning system, reached grandmaster level in 2019. AlphaProof and AlphaGeometry 2, announced jointly by DeepMind on July 25, 2024, solved four of six International Mathematical Olympiad 2024 problems and scored 28 points out of 42, equivalent to a silver-medal performance in that year's competition. Silver was named alongside Quoc Le, Hassabis, and Pushmeet Kohli as overall coordinators. Through this period he held the title of Vice President of Reinforcement Learning Research at Google DeepMind.

Silver departed DeepMind in January 2026 after roughly fifteen years at the lab. Press coverage described his exit as part of a broader 2025 to 2026 pattern of senior researchers leaving frontier labs to found their own companies, with the strategic divergence between his reinforcement-learning thesis and DeepMind's increasingly large-language-model-centric direction cited as the principal motivation. He had founded Ineffable Intelligence in late 2025; the company emerged from stealth on April 27, 2026 with a $1.1 billion seed round at a $5.1 billion post-money valuation, co-led by Sequoia Capital and Lightspeed Venture Partners, with Nvidia, DST Global, Index Ventures, Google, the United Kingdom Sovereign AI Fund, and Madrona Venture Group also participating. The round is the largest seed financing in European technology history. Silver has committed via Founders Pledge to direct 100 percent of his personal Ineffable equity proceeds to charitable causes, characterized by the organization as the largest single pledge in its history.

Affiliations

• Elixir Studios: Co-founder, Chief Technology Officer and Lead Programmer, 1998 to 2005.
• University of Alberta: PhD candidate in computer science, 2004 to 2009 (advised by Richard Sutton).
• University College London: Royal Society University Research Fellow and Lecturer from 2011, Professor of Computer Science to present.
• Google DeepMind: Research consultant from 2010, then Research Scientist through Vice President of Reinforcement Learning Research, 2013 to 2026-01.
• Ineffable Intelligence: Founder and Chief Executive Officer, 2025 to present.

Notable contributions

Silver's published record runs from board-game reinforcement learning through self-play to general-purpose superhuman capability without human data.

• AlphaGo (2015 to 2016). First author on the January 2016 Nature paper "Mastering the game of Go with deep neural networks and tree search," with Aja Huang and others. AlphaGo defeated Fan Hui in October 2015 and Lee Sedol in March 2016 (4 to 1) in Seoul.
• AlphaGo Zero (October 2017). Successor that learned Go entirely from self-play with no human game data, surpassing the original AlphaGo in three days of training. Published in Nature with Silver as first author.
• AlphaZero (December 2017 preprint, December 2018 Science). Generalized the AlphaGo Zero method to chess and shogi alongside Go, demonstrating that the same self-play approach reaches superhuman performance from random initialization across the three games.
• AlphaStar (2019). The StarCraft II reinforcement-learning system, reaching grandmaster level in the open ladder.
• AlphaProof (July 2024). Mathematical-reasoning system using formal-proof reinforcement learning in Lean. Together with AlphaGeometry 2 it solved four of six IMO 2024 problems for 28 of 42 points, the equivalent of a silver-medal performance.
• Ineffable Intelligence (late 2025 to present). Founder and chief executive of the British research startup pursuing a "superlearner" trained without human data; holds the largest seed round in European technology history at $1.1 billion at a $5.1 billion valuation.
• UCL reinforcement-learning lecture series (2015). The ten-lecture course recorded at University College London, hosted on YouTube, is widely cited in graduate-level reinforcement-learning curricula.
• Awards. Royal Society University Research Fellowship (2011); ACM Prize in Computing jointly with Aja Huang and Demis Hassabis in 2019 for the AlphaGo work; Fellow of the Royal Society in 2021; AAAI Fellow in 2022. His Google Scholar profile shows roughly 300,000 citations and an h-index of 102.

Investments and boards

• Ineffable Intelligence (AI): Founder and Chief Executive Officer, 2025 to present. Privately held. $1.1 billion seed at a $5.1 billion post-money valuation closed April 2026, co-led by Sequoia Capital and Lightspeed Venture Partners.
• Founders Pledge (Philanthropy): pledged 100 percent of personal Ineffable equity proceeds, characterized by Founders Pledge as the largest single pledge in its history.

Silver's University College London professorship is an academic appointment held continuously alongside his industrial research roles since 2011, rather than a corporate-board seat. No public personal angel-investor activity in AI, semiconductors, datacenters, software, or energy is on record as of May 2026.

Network

Silver's longest-running professional relationship is with Demis Hassabis, whom he met at Christ's College, Cambridge in 1994. The two co-founded Elixir Studios in 1998 and reunited in 2010 when Silver began consulting for the then two-year-old DeepMind. Hassabis was Silver's joint co-recipient of the 2019 ACM Prize in Computing for the AlphaGo work, alongside Aja Huang, Silver's principal AlphaGo co-author.

His Alberta doctoral advisor was Richard Sutton, the lead author of the standard textbook Reinforcement Learning: An Introduction. The Sutton-Silver line at Alberta is one of the principal academic foundations of the modern reinforcement-learning research community.

At Google DeepMind, Silver's principal collaborators across the AlphaGo to AlphaProof arc include Aja Huang, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, and Karen Simonyan, the recurring co-authors on the Nature and Science papers. Shane Legg and Mustafa Suleyman are the other DeepMind co-founders, with whom Silver overlapped from 2010 onwards. Outside DeepMind, his frontier-lab contemporaries include Ilya Sutskever of Safe Superintelligence, Dario Amodei of Anthropic, Sam Altman of OpenAI, Mira Murati of Thinking Machines Lab, Yann LeCun of Meta, and Tim Rocktäschel of Recursive Superintelligence, the second UK-based reinforcement-learning founder of the 2025 to 2026 cohort.

Position in the field

As of May 2026, Silver is one of a small number of senior frontier-AI researchers who have moved from a chief-research role inside an established lab into the chief-executive role at an independent startup, alongside Ilya Sutskever at Safe Superintelligence, Mira Murati at Thinking Machines Lab, and Tim Rocktäschel at Recursive Superintelligence. The fundraise scale at Ineffable, $1.1 billion at a $5.1 billion valuation, places the company in the same capital tier as Sutskever's Safe Superintelligence.

The "no human data" or "superlearner" thesis underwriting Ineffable Intelligence is a deliberate contrast to the dominant frontier-lab approach. OpenAI, Anthropic, and Silver's former employer Google DeepMind build flagship systems pretrained on internet-scale text corpora and refined with reinforcement-learning-from-human-feedback. Silver's stated direction inverts the ratio: a smaller pretraining footprint, with the bulk of capability emerging from reinforcement-learning interaction with synthetic environments and verifiers. The thesis aligns with his published research at DeepMind, where AlphaZero (2017) demonstrated superhuman performance from random initialization and AlphaProof (2024) extended the same approach to formal mathematical reasoning. Industry coverage at the Ineffable launch characterized the positioning as a contrarian bet against the LLM-scaling consensus.

The 2019 ACM Prize in Computing, the 2021 Royal Society Fellowship, and the AlphaGo and AlphaZero authorship credit place Silver among contemporary AI researchers with both a defining research artifact and the formal recognition of the British scientific establishment. His sustained UCL professorship since 2011, held continuously alongside the DeepMind appointment and now the Ineffable founding role, is unusual for a senior industrial researcher.

Outlook

Open questions over the next 6 to 18 months:

• First Ineffable artifact. Whether the company ships any public-facing demonstration of capability before its next funding round, and what form such a demonstration takes.
• Talent acquisition trajectory. Which senior researchers Silver hires from DeepMind, OpenAI, and academic reinforcement-learning groups, and how the Ineffable team grows from its small founding cohort.
• Compute supply. Whether Nvidia's strategic investment translates into priority allocation of frontier-tier GPUs, and whether Ineffable supplements with additional supplier relationships.
• Series A timing and valuation. The timing, lead investor, and valuation of the next financing event, which will function as the principal external signal about research progress in the absence of public artifacts.
• Founders Pledge mechanics. How the 100-percent-of-equity-proceeds pledge is structured in practice as the company's valuation moves, and which charitable causes Silver directs proceeds toward.
• UCL appointment continuation. Whether the academic appointment continues as Ineffable scales.
• Research-direction validation. Whether closed-domain successes at AlphaZero and AlphaProof generalize to open-ended reasoning at frontier scale.

Sources

• David Silver (computer scientist). Wikipedia biographical entry; covers education, Elixir Studios, the DeepMind period, and the founding of Ineffable Intelligence.
• Mastering the game of Go with deep neural networks and tree search. The January 2016 Nature paper introducing AlphaGo, with Silver as first author.
• Mastering the game of Go without human knowledge. The October 2017 Nature paper on AlphaGo Zero, with Silver as first author.
• Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm. The December 2017 AlphaZero preprint.
• AI achieves silver-medal standard solving International Mathematical Olympiad problems. The July 25, 2024 DeepMind blog announcement of AlphaProof and AlphaGeometry 2 at IMO 2024, with Silver named as overall coordinator.
• TechCrunch: DeepMind's David Silver just raised $1.1B to build an AI that learns without human data. Launch coverage of Ineffable Intelligence with thesis description and investor list.
• CNBC: Ex-DeepMind David Silver raises $1.1 billion for AI startup Ineffable. Funding details and broader frontier-lab-exodus context.
• Tech.eu: Ineffable Intelligence launches with record-breaking $1.1B Seed round. European-market framing and round structure.
• David Silver: AlphaGo, AlphaZero, and Deep Reinforcement Learning | Lex Fridman Podcast #86. The April 2020 long-form interview covering AlphaGo, AlphaZero, self-play, and reward functions.
• ACM Prize in Computing 2019. Citation for the 2019 award to Silver, jointly with Aja Huang and Demis Hassabis, for breakthrough advances in computer game-playing.
• RL Course by David Silver: Lecture 1. The 2015 UCL reinforcement-learning lecture series, hosted on YouTube via the DeepMind channel.
• UCL News: UCL researchers lead two of Europe's largest-ever AI funding rounds. Institutional context for Silver's continued UCL professorship.
• Founders Pledge. The pledge organization to which Silver has directed 100 percent of his personal Ineffable equity proceeds.
• Photo: text-mode card generated via scripts/make_lab_card.py --text "David Silver". No CC-licensed portrait available on Wikipedia or Wikimedia Commons as of May 2026.