Jacob Coxon announced his resignation from Anthropic on September 9, 2026, after three years of pretraining research at both OpenAI and Anthropic. He criticized both companies for what he described as irresponsible behavior, stating they are rushing toward self-improving superintelligence and risking human safety, according to his post on X (formerly Twitter).
Coxon detailed his concerns in a series of posts, emphasizing that neither OpenAI nor Anthropic is managing the development of advanced AI systems with sufficient caution. He highlighted the rapid pace at which these companies are advancing AI capabilities without adequate safeguards, framing it as a gamble with global consequences. His resignation marks a public break from two leading AI research organizations.
The resignation underscores ongoing debates within the AI community about the risks of accelerating AI development without robust ethical frameworks. Anthropic and OpenAI have been at the forefront of AI innovation, but Coxon's departure draws attention to internal dissent over the direction and safety of AI research. This event adds to broader scrutiny of AI firms racing to develop autonomous, self-improving systems.
Coxon's announcement received significant attention, with his post reaching over 18 million views on X. His departure may influence discussions on AI governance and the responsibilities of leading AI developers in managing emerging technologies.