AI Science Dreams Collide with $205 Billion Bill! Why Google Rushed to Adjust Hassabis' Role

Demis Hassabis, co-founder of Google DeepMind, is one of the most iconic figures in the AI era. Google (GOOGL-US) recognized his potential just four years after he founded DeepMind, acquiring the small London-based startup. Today, DeepMind has become a core force for Google in its race for leadership against OpenAI, Anthropic, and Chinese AI players.

However, Google announced a major executive reshuffle this Wednesday: Hassabis stepped down as CEO of Google DeepMind to become Chairman, while Alphabet's Chief Scientist Jeff Dean departed to launch a startup. The news triggered investor concerns, dragging down Alphabet's stock price.

A CNBC column noted that while Hassabis' role change appears sudden, it is actually a natural evolution. What's truly significant is that this move reveals a long-standing internal conflict within Google's AI strategy: Hassabis wants to use AI to transform scientific research, but Google must accelerate technology commercialization, scale up operations, and generate revenue to justify its annual investment of tens of billions of dollars.

Hassabis Pursues Scientific Breakthroughs, Viewing AI as the Ultimate Research Tool

Hassabis has long been clear about his goals in developing AI. In a December interview with CNBC's tech podcast "The Tech Download," he said he dedicated his entire career to AI because he believes it will ultimately become the "ultimate tool for science."

Under Hassabis' leadership, DeepMind has achieved several major breakthroughs, including AlphaGo—the Go-playing system that defeated world-class players—and AlphaFold, which can predict the 3D structure of proteins. The latter helps scientists understand diseases and accelerate drug discovery, enabling AI to expand from computer science research into biomedical fields.

Hassabis admitted that these achievements would have been impossible without Google's vast computing power and custom-designed chips. He continues to push forward on artificial general intelligence (AGI)—AI capable of human-level performance across most cognitive tasks. In his article announcing the chairmanship transition, Hassabis stated he believes AGI is "within reach."

By stepping away from day-to-day operations, Hassabis will now dedicate more time to AGI strategy, global issues, and advising Koray Kavukcuoglu, the Senior Vice President taking over primary responsibilities at Google DeepMind. He will also focus on Isomorphic Labs, the AI-driven drug discovery company he co-founded.

Sebastian Mallaby, author of "The Infinity Machine" and a writer on DeepMind, believes Hassabis' decision isn't unusual. He pointed out that even when Facebook offered a higher acquisition price, Hassabis chose Google due to his emphasis on AI safety. He later negotiated with Alphabet for three years to secure external oversight of DeepMind's AI deployment. This role shift can be seen as Hassabis returning to the side of AI safety and long-term research.

Spending $205 Billion Annually, Google Can’t Afford to Chase Only Scientific Ideals

Although DeepMind’s scientific research continues, today’s AI competitive landscape is vastly different from the past. Alphabet plans to spend up to $205 billion this year. Amid heightened investor scrutiny of big tech capital expenditures, Google also faces fierce competition from OpenAI, Anthropic, cloud computing giants, and Chinese AI firms.

Hassabis described the current environment last December as extremely "brutal." The CNBC column argues this reflects that while Google still supports DeepMind’s scientific research, intense competition is forcing the company to prioritize AI commercialization, revenue generation, and faster product rollouts—while ensuring technical capabilities don’t fall behind rivals.

This shift had already begun. Hassabis previously revealed he discusses strategy almost daily with Google CEO Sundar Pichai, striving to maintain AGI’s long-term goals while speeding up product delivery. Google believes it once lagged competitors in turning research into products and must now shorten the lab-to-market gap.

The new organizational structure may preserve the separation between research and operations but further integrate DeepMind with Google’s product divisions. Hassabis will focus on AGI, scientific research, and long-term safety, while Kavukcuoglu takes on greater responsibility for daily management and executing commercial strategies.

For Alphabet, this arrangement helps retain Hassabis’ research vision while building an operational system better equipped to rapidly launch products, grow users, and generate revenue. The challenge lies in the fact that scientific research requires time, caution, and long-term investment, whereas capital markets demand that Google quickly prove its massive AI spending translates into growth and profit.

Scientific Exploration Meets Global Race—Google Must Move Pragmatically Forward

In his CNBC interview, when asked whether AGI development should slow down, Hassabis acknowledged the current geopolitical and corporate competition system is highly complex—not only are many companies racing to build AGI, but governments are also entering the race.

Ideally, he said, AGI should be a purely scientific exploration, with every step carefully considered. But the real world isn’t like that; DeepMind must pragmatically face the competitive environment and advance more advanced AI as responsibly as possible.

These words perfectly capture the core of Google’s recent executive changes. Hassabis wants to focus on scientific breakthroughs, long-term AGI strategy, and safety governance. Meanwhile, Alphabet is locked in a global AI race where falling behind is not an option—it must launch products faster, capture market share, and respond to shareholder demands for return on investment.

Therefore, Hassabis’ move to Chairman is not merely a personal career adjustment; it symbolizes Google redefining the boundary between scientific ideals and commercial reality. Google needs Hassabis to keep exploring AI’s long-term possibilities, but it also needs stronger operational leadership to rapidly turn research into products and revenue. How well it balances these two forces will determine whether Alphabet, after investing $205 billion, can truly emerge as the winner in the global AI race.