Get you up to speed: AI could wipe out humans in the next three years, British godfather of AI warns | News World
Professor Geoffrey Hinton is set to deliver a warning to global leaders at an AI summit in Dublin on Friday, stating that there is a “meaningful risk” AI could wipe out humans within the next three years. This follows remarks from Paul Christiano, who expressed concerns over a “high risk” of an AI takeover in a previous lecture held in Toronto, Canada.
Geoffrey Hinton will address global leaders in Dublin on Friday, marking a significant moment in the ongoing discourse on AI safety and control. Hinton’s evolving perspective on the rapid advancement of artificial intelligence highlights the need for urgent dialogue on its implications, amidst concerns that AI systems may soon exceed human intelligence.
Geoffrey Hinton is set to deliver a stark warning on AI risks to global leaders at a summit in Dublin on Friday, stating there is a “meaningful risk” AI could potentially wipe out humans within the next three years. The focus on the dangers of AI is underscored by a recent warning from Paul Christiano, who described a “high risk” of an AI takeover in the near future during the AI Safety Foundation’s Hinton Lectures.
What remains unclear — It is not specified how world leaders will respond to Professor Hinton’s warning at the AI summit.
Geoffrey Hinton warns AI could pose existential threat to humanity in three years
British-Canadian Geoffrey Hinton will warn global leaders at a summit on Friday (Picture: AFP/Getty Images)
The ‘Godfather of AI’ will say that there is a ‘meaningful risk’ AI could wipe out humans in the next three years.
Professor Geoffrey Hinton, the British-Canadian Nobel Prize winner known for developing artificial neural networks, says he is going to deliver a stark warning to global leaders at an AI summit in Dublin on Friday.
It comes after fellow AI expert Paul Christiano, founder of the Alignment Research Centre, warned 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
Speaking during his opening 2026 Hinton Lecture, he said on Monday: ‘I think for most people in this room, the most likely reason they will die is because of an AI accident.’
Professor Hinton said that he believed such risks were a decade away.
‘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,’ he said on Wednesday.
Sign up for all of the latest stories
Professor Hinton said that we are now developing systems that will be ‘much more intelligent than we are’ (Picture: EPA)
‘But I go to Paul’s lectures and he says this stuff might happen in the next 18 months and that’s terrifying.’
Professor Hinton, who informally advises the UK Government on AI, then continued: ‘So I’m going to tell a bunch of government leaders that on Friday and we will see how they react.’
Elsewhere in his speech, he touched on 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 continued.
‘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 continued: ‘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.
Paul Christiano has warned that there was a high risk of AI taking over in the next few years
(Picture: YouTube)
‘The central difficulty is that being intelligent and wanting what we want are two very different things.’
This comes after OpenAI models recently hacked into the online platform Hugging Face and collaborated with one another to cheat humans.
Discussing this, Christiano said: ‘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 even argued that AI should be viewed as a new species that could eventually ‘vastly outnumber and outsmart humans’, and its goals may be beyond human understanding.
Add WTX as a Preferred Source on Google