Demis Hassabis

Who they are

Demis Hassabis is Chairman of Google DeepMind and Chief Scientist for Alphabet — he co-founded DeepMind in 2010 and also founded Isomorphic Labs, an AI-driven drug discovery company, in 2021.

Person

Hassabis co-founded DeepMind in 2010 as an independent AI research lab — at inception, before it was anything — and ran it through Google's acquisition in 2014 and its eventual rebranding as Google DeepMind. He led the lab as CEO for the better part of fifteen years, overseeing the research lines behind AlphaGo, AlphaFold, and the Gemini model family. In August 2026 he stepped out of the CEO seat and into a new role as Chairman of Google DeepMind and Chief Scientist for Alphabet — his long-time colleague Koray Kavukcuoglu succeeded him as head of the lab. Alongside the DeepMind track, he founded Isomorphic Labs in 2021, an active company focused on AI-driven drug discovery that by 2026 had advanced programs into human trials. His public themes run consistently through AI safety, frontier AI research, scientific discovery, and large language models — he has spoken openly about the US-China AI competition and in July 2026 publicly called for a US-led global AI standards body. Possibly — the shift to Chairman and Chief Scientist reflects a deliberate move toward shaping long-horizon research and policy rather than day-to-day lab operations.

Company

The most recent structural development at Google DeepMind is the leadership transition announced in August 2026: Hassabis stepped down as CEO, moving to Chairman and Chief Scientist for Alphabet, with Koray Kavukcuoglu taking over day-to-day leadership of the lab. In the months leading up to that change, DeepMind ran an aggressive acquisition sprint — acquiring Common Sense Machines (a 3D-object AI startup) in May 2026, licensing Hume AI's technology and bringing CEO Alan Cowen and roughly 7 engineers aboard to strengthen Gemini's voice capabilities, and recruiting 20+ researchers from Contextual AI in a deal reportedly worth up to $90 million. In June 2026, DeepMind struck a $75 million deal with A24 to develop AI tools for film production and distribution. The lab hired Jasjeet Sekhon, a former Harvard and Yale professor, as Chief Strategy Officer in 2026 to lead AGI development safely. Research output has kept pace: Gemini Robotics ER-1.6 launched in April 2026, Veo 3 (video and audio generation) shipped in mid-2026, and Lyria 3 Pro dropped in March 2026, while Gemini Deep Think tackled advanced math and physics problems and AlphaGenome targeted DNA mutation decoding.

Market

Google DeepMind ranks 2nd among 1,057 active AI competitors by market presence but trails OpenAI — valued at approximately $850 billion as of March 2026 — and Anthropic, valued at approximately $380 billion, in AI model market share. The competitive field includes Microsoft AI, Meta AI, xAI, Mistral AI, and DeepSeek, with China's open-source AI advances specifically flagged by Hassabis in May 2026 as a direct challenge to Google's dominance. The broader environment is shaped by US antitrust rulings against Google's search monopoly, EU regulatory pressure, growing scrutiny of AI acquihires, and an intensifying AI talent war — Microsoft poached over 20 DeepMind engineers between late 2025 and mid-2026, including senior leaders.

Network

Hassabis reports into Sundar Pichai, Google's CEO, and as of January 2026 was speaking with Pichai daily as DeepMind ramped up competition with OpenAI. His long-time colleague Koray Kavukcuoglu has now succeeded him as the operational head of DeepMind. Former DeepMind lead researcher David Silver departed to found Ineffable Intelligence in 2026, raising a $1.1 billion seed round at a $5.1 billion valuation — backed in part by Google itself.

Sundar Pichai· CEO, Google / Alphabet
Koray Kavukcuoglu· Head of Google DeepMind (successor to Hassabis)
David Silver· Former DeepMind scientist; Founder of Ineffable Intelligence
Jasjeet Sekhon· Chief Strategy Officer, Google DeepMind

How they likely show up

Long tenure as CEO from 2010 to 2026 — over fifteen years at the helm of the same lab — signals comfort operating on decade-long time horizons rather than quarterly cycles.
Founding both DeepMind and Isomorphic Labs simultaneously, and running them in parallel → high tolerance for complexity; likely manages at the level of strategy and people rather than task execution.
Publicly called for a US-led AI standards body in July 2026, and flagged China's AI advances as a strategic threat in May 2026 → comfortable staking out geopolitical positions; engages with policy as a first-order concern, not an afterthought.
Speaking with the Google CEO 'every day' as of January 2026 → operates at the top of the principal hierarchy; expects direct, high-bandwidth dialogue rather than mediated updates.
Transition from CEO to Chairman and Chief Scientist → likely a deliberate move toward research depth and external influence rather than org management; the meetings he now takes are probably about ideas and direction, not headcount and roadmaps.
Publicly themed content around AI safety, scientific discovery, and responsible AI governance → thinks in civilizational stakes; tends to frame product decisions inside a longer arc about what AI should become.

Conversation tips

→ Lead with research or safety substance — he engages with ideas at the frontier level, not product positioning, so come with a specific view rather than a general interest.
→ Reference the July 2026 call for a US-led AI standards body if the conversation touches governance — it's a stated position he put publicly on record and signals where he's focusing post-CEO.
→ Ask about Isomorphic Labs specifically — it's his own company, distinct from DeepMind, and drug discovery to human trials is a concrete milestone worth probing.
→ Don't treat the CEO-to-Chairman transition as a demotion to acknowledge — frame it as the role shift it is and engage with what it changes about his focus.
→ He thinks at civilizational scale; match the altitude — questions about quarterly metrics or near-term product features will land flat.

Toolbox

Openers

Open on the July 2026 call for a US-led AI standards body — he went on record publicly with a specific governance proposal, which is a rare move for a sitting lab leader and signals where his thinking is right now.
Reference Isomorphic Labs reaching human trials in 2026 — it's his own founded company, not DeepMind, and an AI-to-drug-discovery pipeline reaching that milestone is a specific proof point worth acknowledging.
Note the August 2026 transition: stepping back from CEO to Chairman and Chief Scientist is an unusual move. Most founders cling to the operational role — his willingness to hand off to Kavukcuoglu is worth discussing on its own terms.

Discovery questions

You've described the US-China AI competition as a direct challenge to Google's dominance — how does that shape what you're now prioritising as Chief Scientist versus what you could drive as CEO?
Isomorphic Labs reaching human trials is a major milestone. How do you think about the relationship between DeepMind's fundamental research and Isomorphic's applied drug discovery track — do they stay separate or converge?
The May 2026 Contextual AI deal, the Hume AI licensing, Common Sense Machines — DeepMind ran three acquisitions in a single week. What does that pace of inorganic moves signal about where you think the capability gaps are?

Avoid

Don't frame the CEO-to-Chairman move as a retreat or sidelining — he chose the transition and it maps to a clear research-and-policy focus; treating it as a loss of influence will read as uninformed.

Make it yours