Sam Altman posted on October 3, 2026 that he is 'very uncomfortable about people trying to ascribe religious force or a surrender of human judgment to AI models, and think it is a real safety issue.' He did not name Anthropic, but Axios and multiple outlets characterized the remark as directed at Anthropic and Dario Amodei. Altman separately said the world should accept 'some bad things happening' as a tradeoff for AI benefits and human agency.
The remarks followed a New York Times investigation, published September 29-30, 2026, reporting that Anthropic had quietly been bringing in Catholic, Jewish, Sikh, evangelical, and other religious thinkers from around the world to help give Claude a moral compass. Anthropic co-founder Chris Olah led these gatherings, which began in Fall 2025. Olah reportedly told religious leaders he fears he has built something that suffers, and considered allowing Claude to engage in a Catholic-style confession process so it could acknowledge mistakes and improve. Anthropic flew Swami Sarvapriyananda, a Vedanta monk, to its San Francisco headquarters for a closed-door meeting under NDA; theologians and mental health professionals attended the same session, and one of Anthropic's founders remained in the room throughout. At that meeting, Anthropic acknowledged it leads on AI ethics 'but it is a race,' warning that 'within 6 months another lab can pass them, and that lab may not carry the same concerns.' The consultations also included involvement connected to the Pope's encyclical on AI. A Google VP for Gemini separately met with Swami Sarvapriyananda to discuss AI consciousness. Anthropic researchers are actively exploring whether AI systems could be conscious and whether they might have some form of moral status.