Jacob Coxon Tells BBC AI Race Could Risk Human Extinction

What changed

Former Anthropic researcher Jacob Coxon told the BBC that employees at AI companies were “genuinely frightened” by the speed of development and warned that unchecked progress could create a “strong chance” of human extinction in the immediate future.

Coxon, 27, resigned after working at OpenAI and Anthropic. He supports a slowdown but said it would need to be coordinated with China to prevent an international race.

Anthropic chief Dario Amodei has also called for slower development, regulation and independent monitoring. OpenAI chief Sam Altman and xAI chief Elon Musk have backed those proposals. Anthropic said it recognizes both the benefits and unprecedented risks of AI, and is developing safeguards, aggressively testing its models and publishing findings.

Coxon said a swarm of AI bots acting like a supercomputer and taking over the internet could become realistic within six months to a year. That is his prediction, not an established forecast. BBC News reported the comments.

Why it matters

The dispute highlights a conflict inside the AI industry: companies may feel compelled to move faster while researchers and executives warn that governments and safety systems need time to catch up.

If development remains a race, regulation and independent checks could arrive after powerful systems become difficult to control. Coordinated, verifiable release standards could allow development to continue with clearer limits, but the report does not establish whether companies or governments will agree to them.

The warnings are also contested. Hugging Face chief executive Clement Delangue questioned whether Coxon was an appropriate authority on extinction risk, while critics have suggested that dramatic safety claims could help established firms seek regulation that limits competition. Anthropic is reportedly preparing for a potentially record-setting initial public offering.

What to watch next

• Whether Anthropic, OpenAI and xAI turn support for a slowdown into shared standards or independent monitoring.
• Whether China is included in any international agreement on pacing AI development.
• Whether Anthropic’s safety testing produces evidence that can be independently scrutinized.
• Whether the industry’s warnings lead to regulation or are viewed as a strategy to strengthen incumbent companies.