Google co-founder Sergey Brin has launched an emergency push to accelerate the company's AI development after Anthropic released Claude Mythos, described as the most powerful AI model at the time. The move reflects growing anxiety within Google as it struggles to keep pace with Anthropic and OpenAI in the high-stakes race for frontier artificial intelligence supremacy.

Why Is Google Falling Behind in the AI Race?

Google's position in the AI market has weakened significantly despite commanding a massive user base. While the Gemini app has surpassed 1 billion users globally, the company remains substantially behind competitors when it comes to enterprise customers, the segment that generates billions of dollars in annual revenue for AI leaders like OpenAI and Anthropic.

The competitive pressure intensified in April 2026 when Anthropic released Claude Mythos, which became the most powerful AI model available at that time. This release prompted Brin to address hundreds of Google DeepMind employees in a town hall meeting, urging them to accelerate work on Gemini models and move faster to close the gap with rivals.

Google has also delayed the release of Gemini 3.5 Pro, its next frontier model, because the system failed to meet internal goals in coding capabilities. This setback underscores the technical challenges Google faces in competing with models from Anthropic and OpenAI that have made significant advances in recent months.

What Leadership Changes Are Happening at Google DeepMind?

Beyond Brin's intervention, Google is undergoing major organizational restructuring at DeepMind, its London-based frontier AI laboratory. DeepMind CEO Demis Hassabis announced he was stepping down from day-to-day management responsibilities, with Koray Kavukcuoglu, a senior vice president at DeepMind, taking over his responsibilities.

The reorganization extends beyond leadership changes. Several non-technical teams are being moved out of DeepMind and into Google's corporate reporting structure, marking a further reduction in the lab's autonomy since Google acquired it in 2014. Despite assurances from leadership that day-to-day work would remain unchanged, the restructuring has raised concerns among employees that DeepMind is being absorbed more fully into Google's corporate hierarchy.

The shift has also moved influence away from London, where DeepMind had been based. Some London-based leaders felt their influence over Gemini's technical direction diminished after Hassabis reduced his involvement in the model and Kavukcuoglu relocated to Google's headquarters in California.

How Is Google's Talent Exodus Affecting Its AI Ambitions?

Google's competitive challenges are compounded by significant talent losses. Four key scientists, including Jeff Dean and Sanjay Ghemawat, recently left the company to create their own AI startup, depriving Google of experienced researchers at a critical moment in the AI competition.

This departure mirrors the urgency Brin felt when he held a similar all-hands meeting with Google employees after OpenAI launched ChatGPT in November 2022. That earlier moment marked the beginning of the current AI race, and Brin's renewed intervention suggests Google leadership views the current competitive gap as equally existential.

How to Understand Google's Current Position in the AI Landscape

• Gemini's User Reach: The Gemini app has achieved over 1 billion users globally, demonstrating strong consumer adoption and brand recognition in the general market.
• Enterprise Revenue Gap: Despite consumer success, Gemini lags significantly behind OpenAI and Anthropic in enterprise customer adoption, the segment that drives billions in annual revenue for AI companies.
• Competitive Model Landscape: Google's Gemini 3 caught up to the frontier in November 2025, but newer models like Anthropic's Fable 5 and OpenAI's GPT-5.6 Sol have since surpassed it, along with advances from Meta, SpaceXAI, and Chinese competitors like Moonshot's Kimi K3.

The competitive landscape has become increasingly crowded and technically demanding. Google had caught up to the frontier in November 2025 with the Gemini 3 release, but the company remains behind newer models such as Anthropic's Fable 5 and OpenAI's GPT-5.6 Sol. Meanwhile, Meta and Elon Musk's SpaceXAI have also made advancements with their respective models, and Chinese players like Moonshot's Kimi K3 have demonstrated significant progress.

Brin's intervention and the organizational restructuring at DeepMind signal that Google recognizes the urgency of its situation. While the company maintains enormous resources and technical talent, the loss of key researchers and the delay of critical model releases suggest that organizational challenges and competitive pressure are mounting. Whether these changes will be sufficient to restore Google's position as a leader in frontier AI remains an open question as the competition intensifies throughout 2026.