Anthropic researcher quits with a warning on AI that echoes 'The Terminator' script

A researcher at Anthropic who goes by the name Coxon has resigned and used a post on X to explain the decision, saying those working on advanced AI sincerely expect it could kill humanity by the end of the decade. The comment framed the developer communitys outlook in stark, existential terms and drew comparisons to the apocalyptic premise of the film The Terminator.

In his post on X, Coxon said the belief that AI can pose an existential risk is held earnestly by people building the systems, and he cited that conviction as a factor in his choice to leave. The warning was presented as a candid inside assessment rather than a hypothetical scenario.

Anthropic, the company where Coxon worked, has positioned itself as a major developer of large-scale AI systems. A resignation and a public statement of this nature from a staff member spotlight internal worries about the trajectory of AI capabilities and the potential consequences if risks are not managed.

The episode is likely to feed broader discussions about governance, safety research, and oversight of powerful AI technologies. A public, on-the-record concern from an industry insider may increase attention from regulators, researchers and the public about how rapidly AI is developing and what measures are needed to reduce catastrophic risk.