1 article on this topic.
A former Anthropic employee's viral claim that AI could 'kill us all by the end of the decade' drew a pointed public rebuttal — and raises a real question for anyone building AI risk assessments: what actually counts as evidence?