Nature examines AI extinction risk after Coxon resignation

Anthropic researcher Jacob Coxon resigned on September 8, telling the Wall Street Journal he feared the systems his employer was building could spiral out of control and destroy humanity. His warning drew more than 90 million views in less than 24 hours, and inside Anthropic it was not immediately dismissed: Evan Hubinger, who leads alignment science at the company, put the chance of AI killing all humans within the next decade at greater than 10 percent.

Nature's science-desk assessment, forwarded within a day of publication by four names on our Who's Who tracker, sets those numbers against researchers who consider the framing overblown. Michael Vermeer, a science-and-technology policy researcher at RAND, authored a 2025 study, "On the Extinction Risk from Artificial Intelligence," concluding that a complete nuclear-driven extinction was infeasible while not ruling out biotechnology or atmospheric-modification pathways. Vermeer told Nature that extinction predictions "involve so many untestable claims that you just end up with a conversation that is really more like faith than something scientific or empirical."

Heidy Khlaaf, chief AI scientist at the AI Now Institute in New York, is blunter. She calls the extinction framing "fear-mongering" and points to AI's low reliability and accuracy rates in critical environments with life-or-death consequences as the more concrete harm to legislate against.