The Nobel laureate who built AlphaGo and AlphaFold recently handed day-to-day control of Google DeepMind to a trusted deputy. But his public message on what's coming next — and how soon — has only grown more urgent, and more ambitious, in the months surrounding that transition.

A Founder Repositioning, Not Retreating

In early August 2026, Hassabis moved from Chief Executive of Google DeepMind to Chair of Google DeepMind and Chief Scientist of Alphabet, handing daily operational leadership to long-time deputy Koray Kavukcuoglu. He continues to lead Isomorphic Labs and remains closely involved in strategic direction — a genuine repositioning of a founder who has run the same organisation since Google acquired it in 2014.

That transition hasn't softened his AGI predictions — if anything, his timeline has compressed further. Speaking at Stanford's Graduate School of Business: "I believe that we're only a few years away from that, maybe 2030, plus or minus a year — which is astounding to think, really." That's a tightening from his mid-2025 estimate of 2030-2035. Speaking to Axios, he connected the accelerated timeline directly to AI agents: "we can see agents really happening now and imagine what they will be in another year, and how useful they'll be."

An Explicitly Historic Framing

At the India AI Impact Summit 2026, Hassabis offered a striking formulation: "it's going to be something like ten times the impact of the Industrial Revolution, but happening at ten times the speed — probably unfolding in a decade rather than a century." He's separately compared the moment to the discovery of fire and the invention of electricity.

Notably, he's paired that ambition with technical uncertainty rather than pure hype — stating simply expanding data and compute may not be sufficient, since today's models still lack genuine understanding of the physical world, robust reasoning, and long-horizon planning. His estimate: one or two further breakthroughs on the scale of AlphaGo may be required.

Calling for International Coordination

In a joint conversation with Anthropic CEO Dario Amodei, Hassabis articulated a specific institutional vision: "my hope is for a 'CERN for AGI' type setup: an international research collaboration on the last few steps towards building the first AGIs" — envisioning AGI's endgame as shared international infrastructure rather than a race between competing companies and nations.

Asked about pursuing AGI amid a "massive geopolitical shock," Hassabis was blunt: "governments need to become more aware of what's at stake... people haven't understood that well enough yet."

The Optimistic Case He's Making Publicly

In an interview with Wired, Hassabis suggested that if development proceeds well, humanity could enter "an era of radical abundance, a kind of golden era" — extending, he suggested, to interstellar exploration. On employment, he's pushed back against framing AGI as primarily a job destroyer, instead emphasising which human skills he expects to become more valuable.

The Bottom Line

Demis Hassabis's 2026 has been a year of genuine institutional repositioning — stepping back from DeepMind's daily operations even as his own AGI predictions have become more specific and urgent. His compressed timeline, his "CERN for AGI" proposal, and his insistence that real scientific breakthroughs still stand between today's AI and genuine AGI together paint a picture of a scientist increasingly speaking as a statesman for the field. Whether his internationally coordinated vision gains traction against the fragmented reality of the current AI race remains uncertain — but his own transition this year suggests he intends to spend the coming years advocating for exactly that outcome.