Srinagar, September 9: A researcher at Anthropic and OpenAI has issued a stark warning about the potential dangers of advanced artificial intelligence, claiming that some people developing cutting-edge AI systems privately believe the technology could pose an existential threat to humanity within the decade.
Jacob Coxon, who said he spent the past three years conducting pretraining research at both companies, announced his resignation from Anthropic on Wednesday. In a series of posts on X, Coxon alleged that leading AI firms are engaged in an increasingly risky race to develop self-improving artificial intelligence.
“The people building AI earnestly believe that it could kill us all by the end of the decade,” Coxon wrote, stressing that he did not consider the warning a marketing stunt.
Coxon further claimed that executives and senior researchers at major AI companies may express stronger concerns about AI risks privately than they do in public. According to him, some industry leaders use more cautious language when speaking to the media while privately expressing fears about the consequences of increasingly powerful systems.
Explaining his decision to leave Anthropic, Coxon accused both Anthropic and OpenAI of failing to act responsibly in the race toward advanced AI.
“I resigned from Anthropic today,” he wrote, arguing that the companies were “racing straight to self-improving superintelligence and gambling with our lives.”
His comments add to growing debate over AI safety, regulation and the potential consequences of developing highly autonomous and increasingly capable AI systems.
