Jacob Coxon, a researcher with three years of experience at Anthropic and OpenAI, has resigned from Anthropic. In a series of posts on X, he alleged that both companies are engaged in a race toward self-improving superintelligence, which he characterizes as a high-stakes gamble with human lives.
Coxon claimed that industry leaders privately express fears that AI could pose an existential threat to humanity by the end of the decade, despite more cautious public messaging. He noted that while Anthropic recognizes the risks, it remains committed to the competitive race to develop advanced systems first.
The resignation follows similar departures, including Anthropic’s Safeguards Lead Mrinank Sharma in February and OpenAI’s former chief futurist Joshua Achiam in September. These exits highlight growing internal concerns regarding the trajectory of AI development and the potential for systems to pursue autonomous goals.
In response to these risks, over 1,100 employees from frontier labs, including Anthropic CEO Dario Amodei and OpenAI chief scientist Jakub Pachocki, signed the Pacing the Frontier letter in July. The initiative calls for government support to develop governance tools that can intentionally pace AI development.
Additionally, both companies joined Google, Microsoft, and approximately 150 other organizations in signing a joint open letter on cyber defense. The document warns that AI-enabled attacks are expected to become more sophisticated and widespread in the near future.
