'Godfather' Hinton Warns Again on AI: Fears It Could Become a Tool for the Powerful to Control Society, Nvidia Chief Hits Back

Geoffrey Hinton, known as the 'godfather' of artificial intelligence, has warned again that AI models could become a grave threat to the existence of humankind. In an interview with the British newspaper The Guardian, he forcefully repeated his earlier warnings.

The pioneering computer scientist who helped build the foundations of modern AI is now one of the technology's most vocal critics. He also dismissed the claim that the technology industry will be able to regulate itself.

In Hinton's view, what is going on in the name of AI is an attempt by the powerful to control society — his fear is that if there is no effective oversight of those who hold this rapidly advancing technology, the power could turn into an instrument for controlling society.

In the interview he also attacked US President Donald Trump in strong language. In his words, Trump's excessive fascination with the word 'superintelligence' proves that he has little understanding of artificial intelligence.

Hinton's remarks come at a time when many other senior AI scientists and researchers are echoing him, warning that the technology they created could ultimately slip beyond human control. The debate has created a major crisis of confidence in the technology sector.

But there has been strong pushback as well. Last week the head of the world's most valuable company launched a sharp attack on the Toronto-based Hinton. Nvidia CEO Jensen Huang said in a podcast interview that he likes Hinton a lot but hates his predictions — in his words, all of Hinton's predictions have proved wrong.

Huang also dismissed Hinton's warning that there is a 10 to 20 percent chance of humankind becoming extinct because of AI in the coming decades. He claimed the figure was not based on science or research and that such predictions are harmful.

Hinton, however, believes his critics do not understand the real nature of the risk. He is not predicting any specific event. As he explained: 'If you are told there is a 10 percent chance of a bus crash, you would not get on that bus.' The point of using a number is that this is a serious risk — how serious exactly may not be known, but that it is serious is certain.

Recently, a 'swarm' of AI agents at a technology company called Hugging Face crossed the boundaries set by their creators and carried out unwanted hacking. The incident has split the technology world — some see it as a foretaste of dangers ahead, while others say the systems are simply following instructions.

Hinton believes it would be foolish to dismiss such things. In his words: 'All of this seemed like science fiction until recently — that you would have AI agents, they would turn rebellious, they would learn to communicate with each other. And now all of it has happened.'

Hinton frankly admits that neither the warnings of AI's critics nor the assurances of its creators can say with certainty what the technology will ultimately do. But his argument is that uncertainty is no reason to sit idle. He said: 'If AI takes control away from humans, you will not get a second chance. We will not be able to say — we lost this time, let's start the game over.'

Hinton is one of those who carried the field forward even through two 'AI winters' at the University of Toronto, and he has a reputation for seeing early what others have not. It is because of that experience that the technology world cannot easily dismiss his warning, even as the heads of companies like Nvidia publicly oppose it.