Anthropic Staff Warn of AI Extinction Risk; Musk Calls It a 'Psyop'

Several current and former Anthropic researchers publicly warned that the AI systems they build could pose an extinction-level risk within a decade, with one alignment lead stating the company "earnestly believe[s] AI could kill all humans." Elon Musk and allied commentators dismissed the warnings as a coordinated "psyop," while Anthropic separately disclosed it had disrupted an attempt to use its models to help build a biological weapon. The dispute reveals deepening division among AI's own builders over whether the technology is racing past meaningful human control.