The British-Canadian computer scientist called the ‘Godfather of AI‘ has outlined the risks of Artificial Intelligence (AI) destroying humanity over the next 30 years, warning that the technology could one day ‘master it’.
Professor Geoffrey Hinton said we should be ‘extremely cautious’ and ‘very serious’ about AI development which he says is ‘very dangerous’.
He has previously stated that there is a 10 percent chance of technology causing the end of the human race – but now predicts that number to be ’10 percent to 20 percent’, due to the rapid pace at which AI is developing.
Still talking BBC Radio 4’s Today programme, Professor Hinton said: ‘You see, we’ve never had to deal with smarter things than we did before.’
He continued: ‘How many examples do you know of a very intelligent thing being controlled by an irrational thing? There are very few examples.
‘There is a mother and a child. The theory of evolution made a big deal about allowing the child to control the mother, but that’s the only example I know of.’
Professor Hinton, who won this year’s Nobel Prize in Physics, warned AI was changing ‘much faster’ than expected and there was not enough time to complete what he believed to be important research.
When his work has laid the foundation machine learninga technology that allows computers to mimic human intelligence, his recent efforts have focused on safe AI.
Professor Geoffrey Hinton has downplayed the potential for artificial intelligence (AI) to destroy humanity in the next 30 years, warning that the technology could one day ‘take control’.
Last year he made headlines after resigning from Google, saying that ‘bad actors’ could use technology to harm others.
Reflecting on where he thinks AI development would have been when he started the project, he said: ‘I never thought it would be where we are now. I thought that sometime in the future we would get here.
Because where we are now is that many experts in the field think that at some point, within the next 20 years, we will create AIs that are smarter than humans.
‘And that’s a very scary thought.’
He added: ‘I like to think of it as, likening yourself to a three-year-old – we’ll be three-year-olds, and they’ll be adults.’
Prof Hinton said he thought AI’s impact on the world would be similar to the industrial revolution.
‘In the industrial revolution, human power was no longer necessary because machines were powerful, and if you wanted to dig a hole, you dug it with a machine.
‘What we have now is what is replacing human intelligence, and ordinary human intelligence will no longer be at its peak. It will be a machine,” he said.
Hinton has previously stated that there is a 10 per cent chance of technology causing the extinction of the human race – but now he predicts that the number will be ’10 per cent to 20 per cent’.
Asked what he thought life would be like 10 or 20 years from now, Prof Hinton said: ‘It will depend a lot on what our politicians do with this technology.
So my biggest concern right now is that we’re in a situation where we have to be very careful and thoughtful about developing technology that could be very dangerous.
‘It will have many positive effects in the medical field, and in almost every industry it will make things better, but we have to be very careful about its development.
‘We need laws to prevent people from using them for bad things, and it seems we don’t have those kinds of politicians at the moment.’
Prof Hinton said he feared AI would be bad for society if more people lost their jobs and all the profits went to the rich.
‘If you have a big gap between the rich and the poor, it’s very bad for people,’ he said.
‘These things are smarter than us. So there was no chance in the industrial revolution that machines would take over humans because they were powerful. We were still in control because we had wisdom.
‘Now there is a risk that these things will become dominant. So that’s a big difference,” he added.
Hinton is considered one of the three pioneering ‘Godfathers’ of AI, along with Yann LeCun and Yoshua Bengio, the trio winning the Turing Award for their work in the field.
