Anthropic researcher quits, warning AI 'could kill us all'

A former OpenAI and Anthropic researcher has quit his job with a stark warning: the companies building some of the world’s most powerful artificial intelligence (AI) systems are racing towards technology they themselves fear could threaten humanity.

Jacob Coxon announced his resignation from Anthropic in a post on X that has been viewed more than 140 million times.

“I spent the last three years doing pretraining research at both OpenAI and Anthropic,” Coxon wrote. “Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives.”

Coxon warned that future systems could become capable of outperforming humans across fields, conducting sophisticated cyberattacks and eventually improving their own capabilities.

“The people building AI earnestly believe that it could kill us all by the end of the decade,” he wrote.

Other AI researchers back warning

Coxon’s claims quickly received support from people currently or previously working inside leading AI laboratories.

Anthropic researcher Evan Hubinger said Coxon was “correct”, adding that he personally estimates the probability of AI killing all humans at more than 10 per cent within the next decade.

“I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence,” Hubinger wrote.

Anna Wang, who previously worked at Google DeepMind and now works at Anthropic, described the concern as common among her peers and said there was not yet a viable scientific plan for managing risks from recursively self-improving AI.

Former Google DeepMind researcher Alex Turner also backed Coxon, saying many researchers believe they are building technology that could eventually kill everyone on the planet.

Warning reaches Washington

The extraordinary public exchange has also drawn attention from U.S. lawmakers. Senator Bernie Sanders backed Coxon’s warning and said he would introduce legislation seeking to ban artificial superintelligence and pause AI development.

Other lawmakers have called for stronger safeguards, including Republican Senator Ted Cruz, who described the potential consequences of AI as a “catastrophic risk”.

Anthropic said it has “always been transparent” that AI could bring both enormous benefits and unprecedented risks, adding that it continues to develop safeguards and supports greater coordination over the release of increasingly powerful models.

Coxon, however, argues that voluntary restraint is unlikely to be enough while companies remain locked in competition.

“Should you put your head down because ‘it’s happening anyway’,” he asked fellow researchers, “or take this moment to call for different conditions?”

Read more about AI:

• AI pioneer warns against race for superintelligence
• Daily Brief - 10 September 2026
• Chinese humanoid robot beats Usain Bolt's 100m world record

Tags

News AI AI danger Jacob Coxon OpenAI Anthropic Anthropic AI humanity exctinction