Google DeepMind announced a significant leadership restructuring this week with Demis Hassabis transitioning from chief executive to chair of the AI research unit. Hassabis will also assume the role of chief scientist at Alphabet, DeepMind’s parent company. The decision transfers day-to-day operational management to Koray Kavukcuoglu, a longtime colleague of Hassabis who will serve as senior vice-president of DeepMind in a non-CEO capacity.
Hassabis, who co-founded DeepMind and oversaw its acquisition by Google in 2014, has been instrumental in major AI breakthroughs including AlphaGo, which mastered the board game Go, and AlphaFold, which predicts protein structures. He received the 2024 Nobel prize in chemistry for the latter achievement. The leadership change marks a notable shift in the unit’s organizational structure, with observers noting it signals DeepMind being more directly integrated into Alphabet’s broader operations, moving away from its previous positioning as a more autonomous subsidiary.
The restructuring occurs amid internal workplace tensions at DeepMind, including employee concerns about the company’s defense partnerships and its involvement with military and homeland security agencies. Some staff members have raised objections to potential military applications of DeepMind’s AI technologies. The announcement coincided with the departure of Jeff Dean, a veteran Google engineer and DeepMind’s chief scientist, who is leaving to establish a new AI startup called Discovery Loop alongside three other senior researchers.
On the commercial front, Google’s flagship Gemini AI model series has reportedly lost competitive ground to systems developed by rivals Anthropic and OpenAI. Reports indicate that the anticipated Gemini 3.5 Pro model release has faced delays while developers work to enhance its code-writing capabilities. Google maintains that the leadership transition is unrelated to Gemini’s market performance and emphasizes Hassabis’s continued focus on scientific breakthroughs.
Hassabis has adopted an increasingly public-facing role, recently calling for the establishment of a US safety body to assess advanced AI models and develop global standards as the field approaches potential artificial general intelligence capabilities. Google notes that despite recent departures, its attrition rates for AI talent during the first half of this year remain lower than comparable periods the previous year.
Article Attribution | Read More at Article Source
Article summary produced by Claude AI