Taking over the day-to-day operations of Google DeepMind is Koray Kavukcuoglu, previously the lab's Chief Technology Officer and Google's Chief AI Architect. Appointed as Senior Vice President of Google DeepMind, he now reports directly to Google CEO Sundar Pichai and holds final decision-making authority over Gemini model development . Kavukcuoglu's mandate is clear: accelerate Gemini 4 pre-training, commercialize AI products at a faster pace, and close the gap with rivals . CNBC reported the appointment signaled Google was "prioritizing execution over deep research" .
While the formal organization chart was being rewritten, Sergey Brin—Google's co-founder, who holds no formal executive title at Alphabet—was working behind the scenes with increasing urgency. According to Reuters, Brin spent months personally pressing senior AI staff to concentrate their efforts on Gemini and close the distance with Anthropic and OpenAI .
In April 2026, Brin addressed hundreds of researchers at a town hall, urging them to "go all in" on Gemini . He created a "strike team" to accelerate coding improvements, pitching the effort to staff as the fastest route to self-improving AI systems . Multiple outlets confirmed Brin has taken an active, hands-on role in driving Gemini's commercialization and competitive positioning . His involvement reflects a board-level belief that DeepMind's research culture had grown too slow for the current AI race.
On the same day as the leadership reshuffle, Google announced a departure that sent shockwaves through the industry. Jeff Dean—Google's chief scientist, employee #30, and a 27-year veteran—announced he was leaving to co-found a startup called Discovery Loop . He was joined by three other senior researchers: Sanjay Ghemawat, a Google Senior Fellow and Dean's long-time collaborator on foundational distributed systems work; Quoc Le, a co-founder of Google Brain; and Oriol Vinyals, a vice president of research at DeepMind and a technical lead for Gemini .
The New York Times noted the departures exposed "the changing power structure inside Google" . Alphabet retained an investment stake and cloud ties to Discovery Loop, which is organized as a public-benefit corporation focused on automating scientific and engineering research .
Beyond the leadership changes, the structural relationship between DeepMind and the rest of Alphabet shifted significantly. Reuters reported that Alphabet moved some non-technical teams from DeepMind into corporate Google, further reducing the lab's independence . At an all-hands meeting the day after the reshuffle, staff learned that parts of the lab were being folded into the broader corporate structure—a move described by observers as Google "tightening its grip" on DeepMind as it trails in the AI race .
This centralization is a marked contrast to the hands-off approach Alphabet took after acquiring DeepMind in 2014, when the lab was allowed to operate with significant autonomy under the "Alphabet" umbrella. The shift signals that Google now considers DeepMind's AI models too strategically important to be developed at arm's length.
Collectively, all five changes—the leadership reshuffle, Brin's push, the talent exodus, the erosion of autonomy, and a new product-first mandate—point to a single strategic pivot: Google is shifting DeepMind from a research institution to an execution engine for Gemini. Sundar Pichai stated that Google had begun its "most ambitious pre-training run yet for Gemini 4" . Kavukcuoglu's immediate priority is to get that model into the market, with an analyst noting that observers should "expect faster releases" and "more experimentation" under his tenure .
The overhaul came at a moment of acute competitive pressure. Anthropic's Claude and OpenAI's GPT models have both made significant gains in reasoning and coding, areas where Gemini has been seen as catching up . By consolidating leadership under a single operational chief, injecting Brin's urgency from the board level, and moving support teams into the corporate structure, Google is betting that speed and execution will reclaim the lead it once held in AI.
Taking over the day-to-day operations of Google DeepMind is Koray Kavukcuoglu, previously the lab's Chief Technology Officer and Google's Chief AI Architect. Appointed as Senior Vice President of Google DeepMind, he now reports directly to Google CEO Sundar Pichai and holds final decision-making authority over Gemini model development . Kavukcuoglu's mandate is clear: accelerate Gemini 4 pre-training, commercialize AI products at a faster pace, and close the gap with rivals . CNBC reported the appointment signaled Google was "prioritizing execution over deep research" .
While the formal organization chart was being rewritten, Sergey Brin—Google's co-founder, who holds no formal executive title at Alphabet—was working behind the scenes with increasing urgency. According to Reuters, Brin spent months personally pressing senior AI staff to concentrate their efforts on Gemini and close the distance with Anthropic and OpenAI .
In April 2026, Brin addressed hundreds of researchers at a town hall, urging them to "go all in" on Gemini . He created a "strike team" to accelerate coding improvements, pitching the effort to staff as the fastest route to self-improving AI systems . Multiple outlets confirmed Brin has taken an active, hands-on role in driving Gemini's commercialization and competitive positioning . His involvement reflects a board-level belief that DeepMind's research culture had grown too slow for the current AI race.
On the same day as the leadership reshuffle, Google announced a departure that sent shockwaves through the industry. Jeff Dean—Google's chief scientist, employee #30, and a 27-year veteran—announced he was leaving to co-found a startup called Discovery Loop . He was joined by three other senior researchers: Sanjay Ghemawat, a Google Senior Fellow and Dean's long-time collaborator on foundational distributed systems work; Quoc Le, a co-founder of Google Brain; and Oriol Vinyals, a vice president of research at DeepMind and a technical lead for Gemini .
The New York Times noted the departures exposed "the changing power structure inside Google" . Alphabet retained an investment stake and cloud ties to Discovery Loop, which is organized as a public-benefit corporation focused on automating scientific and engineering research .
Beyond the leadership changes, the structural relationship between DeepMind and the rest of Alphabet shifted significantly. Reuters reported that Alphabet moved some non-technical teams from DeepMind into corporate Google, further reducing the lab's independence . At an all-hands meeting the day after the reshuffle, staff learned that parts of the lab were being folded into the broader corporate structure—a move described by observers as Google "tightening its grip" on DeepMind as it trails in the AI race .
This centralization is a marked contrast to the hands-off approach Alphabet took after acquiring DeepMind in 2014, when the lab was allowed to operate with significant autonomy under the "Alphabet" umbrella. The shift signals that Google now considers DeepMind's AI models too strategically important to be developed at arm's length.
Collectively, all five changes—the leadership reshuffle, Brin's push, the talent exodus, the erosion of autonomy, and a new product-first mandate—point to a single strategic pivot: Google is shifting DeepMind from a research institution to an execution engine for Gemini. Sundar Pichai stated that Google had begun its "most ambitious pre-training run yet for Gemini 4" . Kavukcuoglu's immediate priority is to get that model into the market, with an analyst noting that observers should "expect faster releases" and "more experimentation" under his tenure .
The overhaul came at a moment of acute competitive pressure. Anthropic's Claude and OpenAI's GPT models have both made significant gains in reasoning and coding, areas where Gemini has been seen as catching up . By consolidating leadership under a single operational chief, injecting Brin's urgency from the board level, and moving support teams into the corporate structure, Google is betting that speed and execution will reclaim the lead it once held in AI.