Nobel-winning AlphaFold creator John Jumper leaves Google DeepMind for Anthropic

John Jumper’s move from Google DeepMind to Anthropic put one of AI’s most consequential scientists at the center of the industry’s latest talent fight. Jumper, best known as a co-creator of AlphaFold, said he was leaving the company on Friday to join the startup, a shift that underscores how top researchers have become strategic assets in their own right.

The departure carries unusual weight because Jumper shared the 2024 Nobel Prize in Chemistry with Demis Hassabis and David Baker for work on protein structure prediction. The Nobel committee recognized Hassabis and Jumper for solving a 50-year-old problem in biology, and DeepMind has said AlphaFold 2 solved the protein structure prediction challenge in 2020. Since then, the system has been used to predict more than 200 million protein structures and has drawn more than three million users from over 190 countries, making it one of the most visible examples of AI’s impact on science and medicine.

That scientific reach is what makes Jumper’s exit so important far beyond Silicon Valley. AlphaFold has helped speed drug discovery and other biomedical research, giving researchers a tool that can accelerate work with direct implications for public health. When a scientist tied to one of the field’s landmark breakthroughs walks out of a dominant lab, it signals that the next phase of AI competition is not only about launching stronger models, but also about which organization can attract the people most likely to shape the next breakthroughs.

Jumper’s move also lands in the middle of a broader reshuffling at the top of the sector. Noam Shazeer, another senior Google AI figure, left for OpenAI just days earlier, reinforcing the sense that the large language model race is still redrawing alliances among the biggest players. Google DeepMind thanked Jumper for his work and wished him well, a sign that the separation appears amicable even as the stakes rise.

Anthropic, meanwhile, is preparing to host a science event on June 30, 2026, placing Jumper inside a company that is trying to define itself as a serious destination for frontier research. The larger message is unmistakable: in today’s AI race, elite scientists are no longer just building products. They are the product, the signal, and increasingly the leverage.