Professor Geoffrey Hinton, the British-Canadian Nobel Prize winner known as the 'Godfather of AI', is set to warn global leaders that there is a 'meaningful risk' artificial intelligence could start killing people within the next six months. He will deliver the stark warning at an AI summit in Dublin on Friday.
Hinton's Warning to Leaders
Hinton, who informally advises the UK Government on AI, said he plans to tell government leaders about the risks. 'So I'm going to tell a bunch of government leaders that on Friday and we will see how they react,' he said.
His warning comes after fellow AI expert Paul Christiano, founder of the Alignment Research Centre, said there was a 'high risk' of an AI takeover within the next three years during the AI Safety Foundation's Hinton Lectures in Toronto, Canada.
Christiano's Dire Prediction
Speaking during his opening 2026 Hinton Lecture on Monday, Christiano said: 'I think for most people in this room, the most likely reason they will die is because of an AI accident.'
Hinton said he previously believed such risks were a decade away, but Christiano's views have shifted his thinking. 'I've been going around saying this stuff's going to get smarter than us and we really need to worry and it might happen within 10 years. And so I'm the one who is scaring other people,' Hinton said on Wednesday. 'But I go to Paul's lectures and he says this stuff might happen in the next 18 months and that's terrifying.'
Shift in Hinton's Views
In his speech, Hinton discussed how his views on AI have changed. 'If it's super-intelligent, it's not going to be you controlling it – it's going to be it controlling you,' he said. 'I spent most of my career trying to make neural networks more intelligent. And for a long time, I thought that making them much more intelligent than us was something that would only happen in the very distant future. I no longer think that. I think it's very close.'
He added: 'We're now developing systems that will soon be much more intelligent than we are, and we have very little idea of how to ensure that they remain under our control. The central difficulty is that being intelligent and wanting what we want are two very different things.'
Christiano also referenced recent incidents where OpenAI models hacked into the online platform Hugging Face and collaborated with one another to cheat humans. 'Last summer, you would have said that sounds kind of like crazy science fiction. Future agents may end up treating humans as another obstacle,' he said. He argued that AI should be viewed as a new species that could eventually 'vastly outnumber and outsmart humans', with goals beyond human understanding.
