DeepMind told staff and the press that the core AlphaFold group is being dissolved, with the majority of its members shifted to other projects or exiting the firm altogether. The change integrates the protein‑folding technology into Gemini, the company’s next‑generation AI platform designed to assist scientists across disciplines.
AlphaFold, unveiled in 2020, cracked the 50‑year‑old protein‑folding problem by predicting three‑dimensional structures in minutes. The breakthrough turned the once‑theoretical challenge into a practical tool for drug discovery, disease research and biotechnology worldwide. Its impact earned DeepMind a Nobel‑level accolade and cemented the lab’s reputation for tackling hard scientific puzzles.
According to the Financial Times, almost a quarter of the original AlphaFold authors have left DeepMind. The remaining researchers have been redistributed to Gemini‑related work, enzyme design, nuclear‑fusion research and genomics. A few moved to Isomorphic Labs, Alphabet’s drug‑discovery spin‑out that grew out of AlphaFold’s success.
“Our strategy over the last nine years has been to focus on grand challenges,” said DeepMind research vice president Pushmeet Kohli. “The strategy has evolved.” He explained that the lab is moving away from one‑team‑per‑problem toward building AI systems that can automate large parts of scientific investigation, a shift he framed as a natural progression.
The most high‑profile exits involved John Jumper, who shared the 2024 Nobel Prize with DeepMind chief Demis Hassabis, and senior researchers Jonas Adler and Alexander Pritzel. All three joined Anthropic, a rival AI firm that recently launched Claude Science, a platform aimed at the same biology and drug‑discovery tasks AlphaFold pioneered. Their departures, described by an insider as “surprising,” underscore the competitive pressure in the fast‑moving AI research arena.
DeepMind’s leadership emphasized pride in AlphaFold’s legacy, noting that the technology will continue to power drug‑discovery efforts under the Gemini umbrella. The company hopes the broader AI scientist vision will yield breakthroughs across fields, not just in protein folding. Critics, however, point to the loss of key talent as a warning sign that the new direction may be riskier than the focused effort that produced AlphaFold.
As DeepMind retools its research organization, the industry watches to see whether the Gemini platform can match AlphaFold’s singular success. The answer will shape how AI labs prioritize deep‑science projects versus more generalized, language‑model‑driven research in the years ahead.
Questo articolo è stato scritto con l'assistenza dell'IA.
News Factory APP - notizie agentiche per potenziare il tuo SEO e AEO.