Google just announced a major shakeup of its top AI leadership

Google on Wednesday reorganized its top AI leadership. CEO Sundar Pichai announced that Demis Hassabis, the longtime leader of Google DeepMind, will become chair of Google DeepMind and chief scientist at Alphabet. Hassabis will also keep leading Alphabet’s Isomorphic Labs, the unit that applies AI to drug development. The move is framed as a significant shift in how Alphabet assigns scientific authority and day-to-day AI leadership, not as a product launch or model release.

The technical substance of the change is organizational rather than model-side. Chair of Google DeepMind separates a scientific and strategic role from pure operational headship of the lab. Chief scientist at Alphabet puts Hassabis above a single product team and into a company-wide science post. Keeping Isomorphic Labs under him preserves a direct line from foundational research to a concrete AI application stack in biology and drug discovery, where models, data pipelines, and experimental loops are already treated as one system.

For engineers and builders inside and around Alphabet, dual roles of this kind usually rewire who sets research priorities, who owns long-horizon bets, and which org owns evaluation standards. Builders shipping on DeepMind research, Gemini-adjacent stacks, or Isomorphic-facing tooling should watch reporting lines and decision rights more than press language. When a lab chair and an Alphabet chief scientist are the same person, model direction, safety and eval culture, and which applied bets get compute and talent often move with that person.

In the broader AI market, Google’s competitors have also concentrated scientific leadership at the top of the org chart while splitting research brands from product brands. Alphabet’s structure already split Google DeepMind research from Isomorphic Labs applied drug work; elevating Hassabis at both DeepMind and Alphabet tightens that research-to-application link under one scientist while still leaving room for other operators to run day-to-day AI products. The incomplete public summary also points to further named changes (including a figure referred to as Koray), which matters because multi-person reshuffles usually rebalance research vs product power, not just one title.

What to watch next is who actually runs Google DeepMind day to day under a chair, how chief scientist authority at Alphabet is scoped across products and labs, and whether Isomorphic Labs keeps the same mandate and resources under Hassabis’s continued leadership. Any follow-on announcements that fill in the rest of the leadership slate—including the Koray piece of the story—will show whether this is mainly a scientific elevation for Hassabis or a wider redraw of AI reporting lines across Alphabet.