Google’s AI Juggernaut Hits a Speed Bump: What’s Behind DeepMind’s Struggle?

Google’s DeepMind has long been the poster child for artificial intelligence innovation. However, beneath its glossy surface lies a story of stagnation and power struggles. The recent shakeup, which saw co-founder Demis Hassabis step down as CEO and several prominent researchers depart, is just the latest symptom of a deeper problem: Google’s AI juggernaut is losing momentum.

At its peak, DeepMind was synonymous with AI excellence, producing groundbreaking models like AlphaFold that left rivals in its dust. However, over the past year, delays in releasing new models have piled up, and top talent has begun to flee for greener pastures – most notably Anthropic, a rival lab founded by former DeepMind engineers. The Gemini 3.5 Pro model, once touted as a game-changer, has missed three release deadlines, with insiders blaming a lack of prioritization on AI coding abilities.

The increasing pressure to deliver commercially viable results in an increasingly crowded field is one factor contributing to this slide. Google may have initially tolerated DeepMind’s eccentricities, but as rivals like OpenAI and Anthropic gain ground, it’s clear that Mountain View is no longer willing to let its London-based lab coast on past glories.

The recent changes also reflect a broader shift in power within the Google empire. Hassabis’ departure from the CEO role may be seen as a demotion by some, but in reality, it represents a shift towards Mountain View and away from London. The new arrangement, where Koray Kavukcuoglu takes over day-to-day operations as senior vice president, reporting directly to Sundar Pichai, is a clear indication of this trend.

Google has faced similar challenges before. Remember the “code-red” declaration in late 2022? That was triggered by OpenAI’s ChatGPT launch, which sent shockwaves through Google’s core search business. Since then, DeepMind has managed to claw back some ground, but its recent struggles suggest that it may not be able to sustain this momentum.

The future of Google’s AI ambitions is uncertain. Will new leadership under Kavukcuoglu restore momentum, or will internal tensions worsen at a critical moment in the AI race? One thing is certain: if DeepMind doesn’t get its act together soon, it risks losing its status as an innovation leader and ceding ground to rivals.

Behind the scenes, employees are bracing themselves for more changes – and some are already speaking out about the impact of this reshuffle on morale. “A lot of people are quite upset by it,” one DeepMind engineer told Fortune. “Both Demis exiting, and losing some of the separation from Alphabet and identity as DeepMind.” This is a warning sign that Google may be sleepwalking into a crisis of its own making.

Google’s treatment of DeepMind raises questions about the role of innovation labs in the corporate world. Should these incubators be given more autonomy to pursue cutting-edge research, or should they be tightly controlled by headquarters? The answer lies somewhere in between – but Google’s recent moves suggest that it may be leaning towards a more centralized approach.

Google is not alone in its struggles to manage AI innovation. IBM, Microsoft, and even OpenAI itself have all faced their own challenges in balancing commercial pressures with research ambitions. What can we learn from these stories? That the AI revolution is as much about people as it is about technology – and that leadership matters.

As the dust settles on this latest shakeup, one thing is clear: Google’s AI juggernaut needs a shot in the arm. Will new leadership under Kavukcuoglu be enough to restore momentum, or will internal tensions worsen at a critical moment in the AI race? Only time will tell – but if DeepMind doesn’t get its act together soon, it risks losing its status as an innovation leader and ceding ground to rivals.