Hugging face CEO criticizes AI extinction warnings as misplaced concern

On September 12, 2026, Hugging Face CEO Clement Delangue criticized the recent debate surrounding Jacob Cockayne, a former Anthropic researcher. Delangue, who posted on X on September 11, compared Cockayne’s warnings about AI extinction risk to a refrigerator repairman discussing climate change. Cockayne resigned this week, citing concerns that Anthropic and OpenAI were racing toward self-improving superintelligence, risking human lives. He claimed AI developers genuinely believe AI could end humanity by the 2030s. Delangue clarified that his analogy was not meant to discredit Cockayne but to emphasize the need for broader, more balanced discussions on AI risks. He noted Cockayne’s work focused on pre-training models, not risk research. The debate highlights growing tensions within the AI community over the pace and implications of technological advancement.