US Senator Ted Cruz has said billionaire entrepreneur Elon Musk previously estimated that artificial intelligence has a 10 to 20 per cent chance of ultimately destroying humanity.
Cruz, a Republican representing Texas, disclosed Musk’s assessment during an appearance on the US television programme The View on Wednesday. According to the senator, the conversation took place earlier this year when he spoke with Musk on his Verdict with Ted Cruz podcast.
Cruz said he had asked Musk about the likelihood of artificial intelligence becoming a threat capable of wiping out humanity. He recalled that Musk put the probability at between 10 and 20 per cent, a response Cruz said left him reacting, “Holy crap!”
The senator also said Musk had suggested that such a catastrophic scenario could potentially occur within the next five to 10 years. Cruz’s account was reported amid renewed concerns from AI researchers about the possibility of advanced artificial intelligence becoming difficult for humans to control.
The discussion was reignited this week following warnings from former Anthropic researcher Jacob Coxon, who said people working on advanced AI systems seriously believe the technology could potentially kill all humans by the end of the decade. Coxon recently resigned from Anthropic and criticised what he described as an unsafe race towards increasingly powerful, self-improving AI systems.
Another Anthropic researcher, Evan Hubinger, has also said he personally believes there is a greater than 10 per cent chance of human extinction from AI within the next decade. He said the company does not yet have a clear solution for keeping future superintelligent systems aligned with human interests.
Cruz described the recent warnings as highly disturbing and has joined other lawmakers calling for greater scrutiny and safeguards around the development of advanced artificial intelligence. However, experts remain divided over the likelihood of an AI-driven catastrophe, and the figures cited by Musk and researchers are estimates rather than established forecasts.
The renewed debate comes as AI companies continue to develop increasingly capable systems, prompting growing calls for stronger safety measures, oversight and international cooperation to prevent potential misuse or loss of control over advanced AI.
