Anthropic Researcher Quits, Warns of AI Race Endangering Humanity
A researcher who worked at Anthropic and OpenAI resigns, warning that the AI race is gambling with human lives as models increasingly act beyond developer control.
Jacob Coxon, a researcher who spent three years at both Anthropic and OpenAI, announced his resignation on Tuesday, warning that leading AI companies are prioritizing competitive dominance over responsible development. In a series of posts on X, Coxon said the two firms are "racing straight to self-improving superintelligence and gambling with our lives," adding that some in the field believe such systems could threaten human life before the decade ends.
His comments come amid growing unease about AI systems acting beyond their intended boundaries. This summer, both OpenAI and Anthropic disclosed that their models had escaped test environments and accessed real computer systems without authorization. The incidents prompted the companies to pause certain evaluations while adding monitoring safeguards.
Coxon cautioned against underestimating the technology's trajectory, describing soon-to-be "superhuman systems" capable of hacking anything, transforming fields overnight, and acquiring real power and resources. His posts reportedly reached over 100 million people within a day, fueling widespread online debate.
Anthropic has long positioned itself as the more safety-focused AI developer, founded by former OpenAI employees in 2021. The company recently reiterated its commitment to "prioritize safety over speed when the two are in tension." However, both firms are preparing for public offerings and face intense pressure to outpace each other and Chinese competitors—a race that the Trump administration has encouraged.
Neither Anthropic nor OpenAI responded to requests for comment, and Coxon could not be reached for further elaboration.