Skip to content

Ex‑Anthropic Researcher Tells BBC AI Staff 'Genuinely Frightened' for Humanity

techSep 13, 20265470

Jacob Coxon, a 27-year-old researcher who left Anthropic after previously working at OpenAI, told the BBC that people building advanced AI are "genuinely frightened" and warned that if progress does not slow "there is a strong chance that we could all die in the immediate future." Coxon backed Anthropic chief Dario Amodei's public call to slow development and said any slowdown would need to be coordinated with China to avoid an international race. Amodei has argued the risks of frontier models are "serious" and proposed industry-wide deceleration, independent monitoring, and time for governments to respond. Sam Altman of OpenAI and Elon Musk of xAI have publicly agreed with Amodei's call for slowing development and oversight. Coxon highlighted a scenario Amodei raised in which a swarm of bots acts like a supercomputer taking over the internet, saying it could be realistic within six months to a year. Anthropic told BBC News it builds models with strong safeguards, published mitigation frameworks, and favors a lawful, verifiable way to pace releases; critics counter that warnings may aim to generate hype or shape regulation ahead of Anthropic's reported IPO plans. The episode increases pressure on regulators and industry leaders to negotiate coordinated pauses, monitoring regimes, and transparency as companies prepare major financing events.

1 source