Former Anthropic researcher Jeff Wu is calling for slower AI progress and stronger oversight of labs. In an interview published by New York Magazine and Asterisk on October 5, he says he wanted to work on slowing progress and coordination after leaving.
Wu previously worked at OpenAI and then Anthropic on alignment and interpretability. He describes concern about using increasingly powerful models to automate further AI research.
In his view, the companies are not doing enough to prepare society or to consider slowing down when needed. That assessment reflects his own experience and judgment.
The publication interviewed eight former employees with different reasons for leaving. Wu’s warning is neither a new October 5 resignation nor a position shared by all participants. Asterisk discloses funding links to some interviewees.