Anthropic Researcher Quits Over AI Safety Fears as Colleague Warns of Greater Than 10% Chance AI Kills All Humans

Jacob Coxon, a 27-year-old pretraining researcher who spent three years at OpenAI and Anthropic, resigned from Anthropic on September 8 and left the AI industry entirely, warning in a post on X that neither company is acting responsibly. "They are racing straight to self-improving superintelligence and gambling with our lives," he wrote. Coxon told the Wall Street Journal that Anthropic's safety work is sincere but that competitive pressure makes hard trade-offs unavoidable, adding that "by the end of next year things could be out of control already."

Hours after Coxon's resignation became public, Evan Hubinger, Anthropic's alignment-science lead, publicly backed his former colleague's alarm, stating that he personally believes there is greater than a 10% chance AI could kill all humans. Coxon had written that the people building AI "earnestly believe that it could kill us all by the end of the decade" and that the danger is not a marketing stunt, noting that executives often soften their public phrasing while expressing deeper fear privately.

The departures and warnings come as Anthropic and OpenAI continue to raise large sums of money and move toward expected public listings, intensifying the tension between commercial momentum and the safety concerns of researchers at the heart of both organizations. Outside experts and commentators noted that insider resignations over AI safety have become increasingly common, echoing the circumstances under which Anthropic itself was founded by OpenAI employees who broke away over similar concerns.