Jacob Coxon leaves Anthropic over AI extinction risk
An Anthropic researcher quit this week. He claims leading AI scientists genuinely believe AI could kill humanity within a decade.
Jacob Coxon resigned from Anthropic this week. He said leading AI companies are behaving badly in their rush to advance their models.
He mainly meant OpenAI and Anthropic. His main claim was this: the people building these systems believe AI could really end humanity by 2036.
He compared the leading labs to runaway trains. Experts have long worried that the newest AI models are developing faster than safety guardrails.
Even their creators doubt humans will fully control them.
- Anthropic
- Resigned from
- Within a decade (by 2036)
- Timeline claimed
- OpenAI and Anthropic
- Companies mentioned
- Safety guardrails lag model development
- Issue
Why it mattersIf AI researchers themselves believe in extinction risks, pressure will grow. That pressure is for better safety oversight and rules about AI work.
AustraliaAustralia will likely align its AI regulation with global safety standards shaped by these concerns.
✓ Claims checked against the source. checked 22 h ago
