A period of significant transition has enveloped Google DeepMind, with its cofounder Demis Hassabis stepping back from the CEO role and a series of high-profile departures raising questions about the organization’s trajectory. Hassabis, a figure synonymous with DeepMind’s scientific identity, has transitioned to a new chair position, ceding day-to-day control to Koray Kavukcuoglu, the company’s chief technology officer. Kavukcuoglu will now report directly to Sundar Pichai, rather than holding a standalone DeepMind chief title, a structural change that has not gone unnoticed within the company. Adding to the executive reshuffle, chief scientist Jeff Dean also announced his departure shortly after Hassabis’s move.
These leadership changes unfold against a backdrop of increasing competitive pressures in the artificial intelligence landscape. Google’s Gemini 3.5 Pro, a key offering, has reportedly missed three release dates since its May I/O unveiling. Further, independent benchmarking firm Artificial Analysis informed *Fortune* that Google’s currently shipped model, Gemini 3.6 Flash, trails competitors like Anthropic, OpenAI, xAI, Meta, and at least one Chinese laboratory in raw intelligence. This marks a notable shift from the previous year, when Google briefly held the top position on the AI leaderboard. The performance dip and executive changes have fueled internal concerns, particularly among engineers who perceive a gradual shift of power from DeepMind’s founding location in London to Google’s Mountain View headquarters. While a Google spokesperson maintains London’s centrality and DeepMind’s research autonomy, the sentiment among some staff suggests a breaking down of the historical firewall between the subsidiary and its parent company.
The talent drain at DeepMind also underscores the challenges it faces. In a single week in June, the company saw Gemini co-lead Noam Shazeer move to OpenAI, and AlphaFold co-inventor John Jumper departed for Anthropic. Jumper was subsequently followed by two other AlphaFold veterans, Jonas Adler and Alexander Pritzel. Engineers attribute this exodus to several factors: aggressive recruitment by well-funded rivals, perceived frustrations over falling behind in coding benchmarks, and the allure of pre-IPO equity opportunities at competitor firms nearing public offerings. The competitive landscape for top AI talent is fierce, and these departures highlight the industry’s dynamic nature.
Despite these headwinds, Google maintains substantial structural advantages that its AI-native rivals cannot easily replicate. The company possesses extensive computing chips, a robust cloud infrastructure, and unparalleled distribution channels. These assets could potentially offset any current weaknesses at the model layer. However, the current period represents another critical juncture for Google’s AI ambitions, reminiscent of an “AI code red” scenario encountered four years prior. The imperative remains for DeepMind to demonstrate its ability to innovate and execute with sufficient speed to keep pace with an accelerating industry. This time, however, it must do so without Hassabis and Dean, two figures who have been central to its scientific identity and past successes.
Meanwhile, the broader AI ecosystem continues to attract substantial investment. Recent funding rounds illustrate the ongoing financial interest in the sector, with Thrive Holdings, an AI holding company, securing $2 billion, and River AI, a developer of personal AI tools, raising $1.1 billion. Other notable investments include $60 million for Heaviside Industries in defense tech, $45 million for payments software company Yuno, and $38 million for Gravity, an ad network for AI products. These capital injections across various AI applications underscore the continued belief in the technology’s transformative potential, even as established players like Google DeepMind navigate their own complex transitions.
