AI bosses crying doom to boost their own profits, UK officials warn as Anthropic whistleblower sparks public fear

From the Newshound archive · first published 24 September 2026.

UK officials have accused artificial intelligence companies of deliberately stoking apocalyptic fear to drive profits and justify their work. The row erupted after Jacob Coxon, 27, a British former Anthropic researcher, went viral warning that AI could trigger a catastrophe — including a bioweapon or novel virus — within this decade, with colleagues quickly echoing his alarm. Anthropic's own senior safety researcher Evan Hubinger estimated more than a 10 per cent chance AI could kill all humans within ten years.

Coxon told Andrew Marr the biggest near-term risk is a biological weapon, not killer robots.

Anthropic boss Dario Amodei and OpenAI's Sam Altman both called for their own industry to slow down — while racing each other to build it faster.

Critics say the doomsday warnings are 'sci-fi' that help frontier labs recruit staff and cement their political importance.