DeepMind Loses AlphaFold Nobel Laureate and Two Co-Authors

A cluster of Google DeepMind departures across a single week in June points at something more specific than the usual talent-poaching story, because of who is walking out and where they are going. John Jumper, the 2024 Nobel laureate in chemistry for AlphaFold, is leaving Google DeepMind for Anthropic after nearly nine years. Within days, Jonas Adler and Alexander Pritzel, both original contributors to the AlphaFold research, followed him to Anthropic. Adler had been working on Google's AI coding efforts and Pritzel on pretraining. Gemini co-lead Noam Shazeer separately left the same week for OpenAI.

The detail that matters more than any single hire is what those people had been doing lately. Jumper, per Silicon Canals and other reporting, had reportedly been pulled toward Google's commercial coding tools in his final months, and other AlphaFold veterans had already been drifting toward Isomorphic Labs, the Alphabet spinout that turned the AlphaFold work into a drug-discovery company. So it is not simply that a Nobel laureate got a bigger offer. The team behind Alphabet's most celebrated scientific result had been steadily reassigned or spun out, and the moment a credible science-oriented rival opened its arms, several of them walked.

Anthropic is not accidentally on the receiving end. Through 2026 the company has, per reporting summarised alongside the exits at TechCrunch, opened wet labs and struck research partnerships with the Allen Institute and the Howard Hughes Medical Institute, which gives the incoming AlphaFold researchers a working environment for the kind of science they came from. Alphabet's stock reaction across the affected sessions was reportedly steep, with the aggregate market-cap loss estimated at roughly $270 billion.

The honest caveat is that most of what's public is a series of named departures, not a full count. The reporting doesn't tell you how many original AlphaFold authors still sit inside DeepMind versus at Isomorphic versus outside Alphabet entirely, doesn't quantify any retention offers, and doesn't say what any of the departing scientists are contractually free to build at Anthropic. Take the specifics as reported, not settled.

What is worth watching from here is less the next resignation and more whether Anthropic uses this cohort to open a real AI-for-science line of work, and whether Alphabet responds by giving Isomorphic more independence or by rebuilding a scientific track inside DeepMind itself. The scoreboard for the next twelve months is probably not model benchmarks; it is which lab publishes the next AlphaFold-scale scientific result.