Google DeepMind CEO Warns of AI Power Shift: “Humans Could Become Subordinate to Machines Within 4 Years”

Artificial Intelligence is advancing at an unprecedented pace, and now one of the world’s leading AI experts has issued a serious warning about humanity’s future. Demis Hassabis, CEO of Google DeepMind and a prominent figure under Alphabet, has stated that the world may be just a few years away from reaching a point where AI systems surpass human intelligence.

Speaking at a recent company event, Hassabis said humanity is approaching the era of “Singularity” — a stage where machines become more intelligent and capable than humans. According to him, the transition toward Artificial General Intelligence (AGI) could happen much sooner than most people expect.

“The Next Year Will Be Just a Trailer”

Hassabis explained that the AI tools and agents arriving over the next year should only be viewed as an early glimpse of what is coming next. He described the upcoming wave of AI systems as merely a “trailer” for far more powerful technologies that could fundamentally transform industries, economies, and daily life.

The DeepMind chief believes the AI revolution is accelerating rapidly, with future systems expected to perform complex reasoning, decision-making, and creative tasks at levels beyond human capability.

Why Demis Hassabis Wants Governments to Act Now

Hassabis emphasized that he intentionally uses strong language to push governments, economists, regulators, and the public to take AI development seriously. He warned that the world needs clear global safety standards before advanced AI models become widely accessible.

According to him, mandatory safety testing and rigorous oversight should be implemented before companies release powerful AI systems to the public. He also revealed that discussions are underway between leading global AI firms to establish common AI safety frameworks.

Can AI Improve Itself?

One of the biggest concerns surrounding AI is whether machines can eventually make themselves smarter without human intervention.

Hassabis noted that while AI has not fully reached that stage yet, early signs are already visible. Engineers using AI-assisted coding tools are now able to write software faster and more efficiently than before. This growing capability is speeding up AI research itself, creating a cycle where AI helps improve future AI systems.

Experts believe this self-improvement loop could significantly accelerate the arrival of AGI.

DeepMind CEO Criticizes AI-Driven Layoffs

Hassabis also addressed the ongoing wave of layoffs across the global technology sector. Reports indicate that more than 142,000 tech workers have lost their jobs in 2026 as companies increasingly focus on AI automation.

However, Hassabis strongly criticized firms using AI as a reason to cut engineering staff.

He argued that if AI allows developers to become three or four times more productive, companies should expand their ambitions rather than reduce their workforce.

According to Hassabis, businesses should use AI to build more products, conduct more research, and solve larger problems instead of eliminating jobs.

Open Invitation to Laid-Off Engineers

The Google DeepMind CEO also extended an open invitation to talented engineers affected by layoffs. He said he would gladly hire skilled professionals to work on advanced AI projects involving healthcare, scientific discovery, gaming, and other emerging technologies.

Hassabis stated that he has “hundreds of thousands of ideas” ranging from drug discovery to game design that could benefit from the expertise of experienced engineers.

Why This AI Warning Matters

The comments from the head of Google DeepMind come at a time when AI adoption is rapidly increasing across industries worldwide. From coding assistants and chatbots to autonomous systems and advanced robotics, AI technologies are already reshaping the global economy.

While many experts believe AI can improve productivity and innovation, others warn that unchecked development could create major ethical, economic, and security risks.

As the race toward AGI intensifies, Hassabis’s warning highlights a growing global debate: Will AI become humanity’s greatest tool — or its greatest challenge?