Geoffrey Hinton, one of the pioneers of artificial intelligence and a Nobel Prize winner in physics, has warned that humanity could face a serious AI-induced crisis as early as this year or next. In an interview with the British Guardian, as reported by Index, Hinton stated that such an event could involve major power outages and a multi-day disruption to the banking system.
Hinton, who is 78 years old and known as one of the 'godfathers' of artificial intelligence, helped lay the foundations for today's AI systems through his work on artificial neural networks. He believes the next decade will determine whether humanity can successfully manage the risks of this new technology. He warns of the possibility of mass unemployment, deepening inequality and serious social unrest, but believes artificial intelligence can also bring enormous benefits. He admits, however, that it may ultimately turn out that people would have been better off if they had never developed it.
Hinton compared recent security incidents linked to autonomous AI systems to the 1979 nuclear accident at the Three Mile Island power plant in the United States. He believes a much more serious event, a sort of 'Chernobyl' of artificial intelligence, could follow soon. He is not referring to a nuclear disaster, but to a cyberattack that could cause major problems in electricity supply and temporarily paralyse the financial system. He argues that such a crisis could force world leaders to take AI regulation more seriously.
Hinton is particularly critical of the directors of leading technology companies. He expressed deep distrust of OpenAI chief Sam Altman, and accused Mark Zuckerberg, Elon Musk and Altman of being aware of the risks but prioritising profits and market competition. He advocates for mandatory testing of the most advanced AI systems before they are released for public use, citing the US Food and Drug Administration (FDA) as an example, which requires evidence of safety and efficacy before approving drugs. He believes AI manufacturers should have to prove to an independent regulator that the benefits of their systems outweigh the potential harms.
Hinton opposes the approach of US President Donald Trump, who is pushing for fewer regulatory restrictions so that American tech companies can compete with China. Despite everything, Hinton has not lost hope. He believes people still have the ability to steer the development of artificial intelligence and prevent the worst consequences, but warns that this will require significantly greater efforts from governments, scientists and the tech industry. Hinton left Google in 2023 so he could speak more freely about the dangers of artificial intelligence.
His latest remarks come ahead of the release of the documentary film AI: Probably Nothing to Worry About, which features Demis Hassabis, head of Google DeepMind.
