Brief IA

DeepMind Restructures AlphaFold, Departures to Anthropic

🤖 Models & LLM·Tom Levy·

DeepMind Restructures AlphaFold, Departures to Anthropic

DeepMind Restructures AlphaFold, Departures to Anthropic
Key Takeaways
1DeepMind has restructured its AlphaFold team, redeploying the majority of researchers to other projects.
2Nearly a quarter of the AlphaFold team members have left Google DeepMind.
3This reorganization marks a significant turning point for DeepMind, which had bet on AlphaFold.
💡Why it mattersDeepMind's shift could influence the future of AI research and talent dynamics in the industry.
Le brief IA que lisent les pros

Le brief IA que les pros lisent chaque soir

Les 7 actus IA du jour, décryptées en 5 min. Gratuit.

Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.

Choisis ton rythme

Gratuit · Pas de spam · Désabonnement en 1 clic

📄
Full Analysis

DeepMind Restructures AlphaFold, Departures to Anthropic

The majority of researchers behind AlphaFold are now working on other projects, and nearly a quarter of them have left Google DeepMind. This reorganization marks a radical shift from the strategy that put the lab on the map.

According to the Financial Times, Google DeepMind has dissolved the team behind AlphaFold. Most of the original authors of the AlphaFold papers have been reassigned internally over the past year.

DeepMind has confirmed these movements. The researchers are now working on projects related to Google’s Gemini language model, as well as enzyme design, nuclear fusion, and genomics. Others have been transferred to Isomorphic Labs, Alphabet's subsidiary focused on drug discovery, which was separated from DeepMind in 2021 and has partnerships with Novartis and Eli Lilly. According to the FT, nearly a quarter of the full-time authors of the original AlphaFold papers have left the company.

AlphaFold, developed since 2018, predicts the three-dimensional structure of proteins. John Jumper, Vice President and Engineering Fellow, along with DeepMind CEO Demis Hassabis, received the 2024 Nobel Prize in Chemistry for this work. The project became the flagship example of DeepMind's strategy to pursue long-term scientific breakthroughs through dedicated research teams.

This approach is now being abandoned. "Our strategy over the past nine years has been to focus on major challenges... a concrete goal that each project is centered around," said Pushmeet Kohli, Vice President of Research at Google DeepMind, where he founded and leads the AI for Science team. "The strategy has evolved."

Rather than simply organizing researchers around unique scientific problems like protein folding, the lab is also building systems powered by Gemini that can assist scientists and, ultimately, automate certain parts of the scientific process. Meanwhile, DeepMind is competing with OpenAI and Anthropic to develop cutting-edge AI agents.

Key Talent Departures to Rivals

The departures are affecting the inner circle. Earlier this year, Jumper and his fellow AlphaFold researcher, Jonas Adler, joined Google’s internal "Code Strike" team, a group assembled to enhance the company's AI coding capabilities in its quest to catch up with Anthropic and OpenAI. Last month, Jumper announced his departure to Anthropic, where he will be joined by Adler and his AlphaFold colleague, Alexander Pritzel.

A DeepMind employee, who wished to remain anonymous, told the FT that the three had been "instrumental, important, essential" members of the company and that their departures had come as a surprise internally.

DeepMind stated that it is "extremely proud" of its scientific legacy and the global impact of AlphaFold, "which inspired the launch of Isomorphic Labs with the mission of reinventing drug discovery." Anthropic launched Claude Science for biology and drug discovery earlier this year.

Brief IA — L'actualité IA en français

L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.