Anthropic researcher Jacob Coxon resigns citing fears of out-of-control AI

Anthropic pretraining researcher Jacob Coxon resigned, stating that frontier labs including Anthropic and OpenAI are racing toward self-improving systems without adequate safeguards and that builders privately believe AI could kill all humans by the end of the decade. Anthropic Alignment Science Lead Evan Hubinger publicly agreed the extinction risk exceeds 10 percent within a decade and that the company lacks a clear plan to solve alignment for superintelligence.