Jacob Coxon

Jacob Coxon's resignation from Anthropic became public when he announced it on X, saying he stepped down on September 8. His stated reason was that OpenAI and Anthropic are racing toward self-improving superintelligence. The announcement placed him inside a wider internal safety argument that had, by then, moved into the open.

The response from inside Anthropic was immediate. Evan Hubinger, who leads alignment research at the company, replied that Coxon is correct and gave his own estimate of over 10 percent within the next decade. That exchange linked Coxon's departure to the broader push among researchers who want the pace of capability work slowed, including Sam Altman of OpenAI and Dario Amodei of Anthropic, both of whom the coverage notes had already been discussing slowdowns.

Days later, the argument reached legislation. Senator Bernie Sanders and Representative Greg Casar announced on September 3 a plan to submit a bill banning superintelligence. On September 16, Bilal Chughtai, a researcher who worked on AGI safety and AI alignment at Google DeepMind, announced his own resignation and warned that AI could cause human extinction, saying capabilities are improving far faster than alignment understanding.

Coxon's coverage and Chughtai's coverage do not reference each other directly. What connects them is the same claim: researchers inside the major labs saying the alignment side is not keeping up. Coxon's dossier rests on the September 10 story and on his own public statement that he resigned on September 8. The stories do not give a current role, employer, or next step for him.