Demis Hassabis on the Unseen Potential of AI and Its Transformative Journey

Demis Hassabis, co-founder and CEO of Google DeepMind and a leading pioneer in artificial intelligence, has long been aware of the transformative potential of AI. He is astonished that it has taken the world so long to recognize its implications.

In a virtual interview with Axios from London, where DeepMind is headquartered, Hassabis, 48, shared his perspective: "I've been considering this for decades. It was evident to me that this was the most significant development," he stated. "I didn't anticipate that it would happen in my lifetime. Even when we launched DeepMind 15 years ago, very few people were seriously exploring this field."

The significance: AI reached a landmark moment recently when Hassabis and DeepMind colleague John Jumper were recognized with a shared Nobel Prize in chemistry. Moreover, the Nobel Prize in physics was awarded to Geoffrey Hinton, known as the "godfather of AI," and machine-learning pioneer John Hopfield.

Hassabis reflected on this progress, saying, "Perhaps it's a pivotal moment for AI, as it is now mature and advanced enough to contribute meaningfully to scientific discovery." He emphasized that there’s no need to wait for artificial general intelligence (AGI) — systems exceeding human intelligence, the ultimate ambition for AI developers. AI can already "transform drug discovery," he noted.

Broad context: Hassabis acknowledged that AI might be "overhyped in the short term" due to the success of OpenAI's ChatGPT, which has sparked investor interest. He echoed the sentiments of many veteran researchers who have dedicated years to methodical development behind the scenes. "I would have preferred it to remain at a scientific level," he admitted, "but it's become too mainstream for that."

Hassabis believes AI is "still largely underrated in the long term," adding, "People don’t fully grasp what I have contemplated and experienced for 30 years."

Behind the scenes: Axios' Ina Fried reported that Hassabis has taken a leading role in overseeing Google’s entire AI endeavors, consolidating efforts from different teams under DeepMind.

DeepMind's influence extends beyond Google, with co-founders overseeing AI at both Google and Microsoft. Mustafa Suleyman, another DeepMind co-founder, took on the role of CEO of Microsoft AI in March, directing Copilot and consumer AI.

Background: Hassabis began playing chess at age 4, quickly surpassing his father and uncle in skill and later leading English youth chess teams when England was second only to Russia. At age 11, he wrote his first AI program to enhance his performance in the strategy game Othello (Reversi). He led the Cambridge chess team while achieving top honors in computer science, followed by a Ph.D. in cognitive neuroscience at University College London.

An avid reader of science fiction, Hassabis was always fascinated by "the big questions," which typically draw people toward physics. However, even then, he sensed a larger pursuit awaited.

"Physics was my favorite subject," he said. "If you seek to understand the nature of reality, time, or the universe, you study physics. Yet I felt that, even with the insights from physics greats, we lacked tools to tackle these monumental questions."

Throughout his journey, even when his professors doubted his focus on AI, particularly his choice to pursue it privately rather than academically, Hassabis remained undeterred.

"I never worried about that," he recalled. "I was committed to this path, regardless."

Conclusion: Hassabis remains in awe when he observes Earth from a 747 or converses via Zoom across 3,000 miles—both are testaments to human ingenuity.

"If we could artificially create that intelligence and make it abundant, leading to super-intelligence… that would revolutionize the world," he concluded. "So it's somewhat surprising that more people didn’t recognize this sooner."