The man known as the ‘Godfather of AI’ has spoken about the possibility that artificial intelligence could ultimately replace humanity. In 1986, Professor Geoffrey Hinton proposed a method to predict the next word in a sequence, work which laid the foundation for large language model AI. Last year, he was awarded the Nobel Prize in Physics alongside John Hopfield ‘for foundational discoveries and inventions that enable machine learning with artificial neural networks’. Given his expertise, Hinton’s perspective on technology and AI carries considerable weight.
++ Pioneering gene therapies offer hope to those with rare genetic mutations
Hinton believes AI will have a transformative effect on society but has expressed serious concerns about its potential future. In a recent interview with CBS, he likened AI to raising a tiger cub, cautioning: “Unless you can be very sure that it’s not going to want to kill you when it’s grown up, you should worry.” The fear that AI could develop to a point where it no longer needs or listens to humanity is a concern not confined to science fiction, but increasingly voiced by experts within the field.
The eminent professor estimated the risk of AI replacing humanity to be between 10 and 20 per cent — odds he suggested were worrying enough not to ignore. Hinton has warned that major companies are lobbying for reduced AI regulation at a time when oversight is already minimal. Reflecting on the unprecedented nature of the challenge, he noted humanity has “never had to deal with things more intelligent than ourselves before,” and questioned whether it was realistic to believe a more intelligent entity could be controlled by a less intelligent one.
++ Britain’s starling population plummets to all-time low
Hinton further observed that AI development is progressing “much faster” than he had anticipated, with many experts now predicting that AI could surpass human intelligence within the next two decades — a prospect he describes as “very scary.” He advocates for stronger government regulation to slow the pace of development and ensure adequate research is conducted. In the meantime, a light-hearted suggestion is offered: continuing to use ‘please’ and ‘thank you’ when interacting with AI systems — though, as he notes, the cost of such politeness is rapidly adding up.
