Anthropic and OpenAI Under Fire as Ex-Staffer and Safety Chief Warn AI Could 'Kill All Humans' Without Global Controls

Phoenix News — An AI researcher who left OpenAI to join Anthropic has announced he is leaving the industry, accusing both U.S. companies of "playing with our lives" in the race to develop AI models capable of self-improvement.

Jacob Coxon, 27, spent the past three years pretraining AI models, first at OpenAI and then, this year, at rival Anthropic, which he had considered more cautious. Pretraining is the stage where AI models absorb vast quantities of data.

"Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives," Coxon said on September 9. "The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt," he wrote on social media platform X.

Anthropic safety executive Evan Hubinger backed Coxon's remarks on X. "We really do earnestly believe AI could kill all humans!" he said, estimating the risk at more than 10 percent over the next decade. Hubinger said Anthropic is "trying its best" but does not yet have a plan to ensure a superintelligent AI system would obey its creators.

Coxon's resignation comes as Anthropic prepares for its market debut, following a summer marked by unauthorized hacks carried out by AI tools during testing. Neither company immediately responded to AFP requests for comment.

AI leaders say so-called "recursive self-improvement" — a stage where AI systems could design and train the next generation of AI with little human involvement — is drawing near. Coxon said he believes no company can responsibly develop AI that surpasses humans without government intervention or a coordinated slowdown.

In February, Anthropic removed a pledge from its safety charter to halt model development if it failed to control risks, arguing that a unilateral pause would allow less cautious rivals to dominate. In late July, more than 1,000 tech employees, including Anthropic CEO Dario Amodei, called on Washington to support a coordinated slowdown in advanced AI development. OpenAI halted training of its latest models for two weeks in August before resuming under tighter controls.

On September 7, OpenAI chief scientist Jakub Pachocki called for "extreme caution," writing in a blog post that "international coordination on future AI development needs to become a top priority for governments around the world." AI models are not regulated by federal law in the United States. In September, Senator Bernie Sanders and Representative Greg Casar introduced a bill seeking to suspend AI development until a federal regulator is created.