Geoffrey Hinton, a name synonymous with the rise of artificial intelligence, is often called the “godfather of AI.” Over the years, he’s been both a pioneer and a prophet, shaping the future of technology while warning us of its darker possibilities.
But in a recent twist, Hinton has revealed that he now uses the very AI tools he once cautioned the world against.
In a candid interview with CBS, the 2024 Nobel Physics laureate admitted he regularly relies on OpenAI’s GPT-4 model for day-to-day tasks.
“I tend to believe what it says, even though I should probably be suspicious,” he said. It’s a surprising statement from a man who has spent decades urging caution around artificial intelligence, especially as it grows more powerful and autonomous.
For those unfamiliar with Hinton’s legacy, he is one of the minds behind neural networks, the foundational tech that powers modern AI models like ChatGPT. His work laid the groundwork for machine learning systems capable of understanding language, recognizing images, and mimicking human thought processes.
In 2024, the Nobel Committee recognized his profound impact by awarding him the Nobel Prize in Physics, an unusual but telling choice that underscores how deeply AI is reshaping our world.
Despite his accolades, Geoffrey Hinton has never shied away from voicing concern. In 2023, he made headlines for quitting his position at Google, a move he said would allow him to speak freely about AI’s dangers.
Also Read: India will take lead in shaping upcoming 6G: Jyotiraditya Scindia
“A part of me now regrets my life’s work,” he told The New York Times shortly after stepping down. He worried that the technology he helped build might spiral beyond human control — a tool that, if misused, could become a threat to humanity.
His fears aren’t new. Over the past several years, Hinton has been outspoken about the risks of artificial general intelligence (AGI), AI systems that could learn and operate at or beyond human intelligence levels. Initially, he believed such advancements were 30 to 50 years away.
But in early 2023, his outlook shifted. In an interview with CBS, he warned that AGI might be fewer than 20 years out, and that its impact could rival that of the industrial revolution or the discovery of electricity.
Hinton also shared a chilling insight in a BBC interview: chatbots can now share knowledge in ways humans cannot. “If one chatbot learns something, that information becomes instantly available to all others,” he said. In essence, AI doesn’t just learn, it teaches itself and its peers, potentially outpacing human learning exponentially.
So how does someone with such deep concerns find themselves turning to GPT-4? Hinton’s answer is layered. While he remains wary of AI’s potential to manipulate or misinform, he also acknowledges its usefulness and power, something he, more than anyone, understands intimately.
The paradox is hard to miss: Geoffrey Hinton, once a vocal skeptic of unchecked AI advancement, now trusts it, albeit with some hesitation, to help him navigate the complexities of modern life.