Nobel Laureate John Jumper Leaves Google DeepMind for Rival Anthropic

John Jumper, a Nobel Prize-winning scientist and Vice President at Google DeepMind, has resigned from the company to join Anthropic PBC, an artificial intelligence startup. Jumper shared the 2024 Nobel Prize in Chemistry with DeepMind CEO Demis Hassabis for developing AlphaFold, an AI model that predicts protein structures.

Jumper announced his decision on X, formerly Twitter, stating, 'After nearly 9 years, I have decided to leave Google DeepMind and join Anthropic.' He added that Hassabis had taken a chance by allowing him to lead the AlphaFold team shortly after completing his PhD.

A Google DeepMind spokesperson confirmed the departure and expressed gratitude for Jumper's contributions. An Anthropic spokesperson also confirmed his upcoming role at the company.

Jumper's move highlights the intensifying competition among AI firms. Anthropic, co-founded by former OpenAI employees, is a direct rival to Google in developing advanced AI models. The departure follows that of Noam Shazeer, a prominent Google researcher who left for OpenAI earlier this year.

DeepMind employees have reportedly expressed concerns about the company's lack of a clear strategy for AI coding tools, a segment where Anthropic and OpenAI have gained traction. However, Jumper's decision appears to be driven by personal career choices rather than dissatisfaction.

Jumper earned a Marshall Scholarship to study at Cambridge University in 2007 and a PhD in theoretical chemistry from the University of Chicago. His work on AlphaFold has been hailed as a breakthrough in biology, enabling rapid prediction of protein shapes that are crucial for drug discovery.

Demis Hassabis acknowledged Jumper's contributions on social media, writing, 'Thanks John for an extraordinary partnership... What we achieved with AlphaFold changed the world.'

As of now, Google DeepMind continues to compete in the AI race, with investments in generative AI and coding tools. Anthropic, meanwhile, has secured billions in funding and is developing its own large language models.