Anthropic Researcher Jacob Coxon Resigns Over AI Extinction Fears

Anthropic researcher Jacob Coxon resigned from the company in early September, citing growing concerns about the existential risks posed by advanced artificial intelligence. Coxon, who had worked on safety and alignment projects at Anthropic for several years, said that the pace at which the organization and the broader industry were developing increasingly powerful models outstripped the available safety protocols and oversight mechanisms.

In a statement released by the company, Coxon urged the AI community to adopt “pacing agreements” that would coordinate the release schedules of new technologies across laboratories. He argued that such agreements could help prevent a race to the bottom in safety standards and provide a framework for shared risk assessment. The call follows a series of high‑profile warnings from AI researchers and ethicists about the potential for rapid, uncontrolled deployment of systems that could surpass human control.

The resignation has prompted discussions within the AI safety community about formalizing collaborative safety frameworks. While Anthropic has not yet announced a replacement for Coxon, it has pledged to review its internal safety protocols and to engage with other research institutions on the proposed pacing agreements. The move underscores the increasing pressure on AI developers to balance innovation with robust risk mitigation.