Commentary Forget the 'hive mind' of AI agents 'going rogue.' They did what humans programmed them to do.

Is the panic over artificial intelligence justified? Anthropic researcher Jacob Coxong resigned on Sept. 8, warning that 'the people building AI earnestly believe that it could kill us all by the end of the decade.' Colleague Evan Hubinger publicly agreed and put the risk at greater than 10% within 10 years. The Guardian quickly reported that 'Anthropic researchers say AI could cause human extinction by 2030'—a probability estimate recast as an institutional finding.

Days later, Anthropic CEO Dario Amodei warned that autonomous AI agents could overwhelm large portions of the internet within months. Citing brief statements of agreement from Open AI's Sam Altman and Space XAI's Elon Musk, Axios headlined the story 'Anthropic, Open AI CEOs call for slowdown in AI development.' A hedged forecast had become an industry consensus.