I think it gets the chronology wrong? Some of these AI safety orgs were around before Anthropic. AI safety concerns are not new.