Overview
- Sam Altman posted on X on Saturday saying he was "very uncomfortable" with people giving AI models religious force and called that tendency a real safety issue.
- Reporting in late September showed Anthropic co‑founder Christopher Olah met with religious and philosophical leaders and helped craft an internal 84‑page guideline known as the "Soul Doc" for the Claude model.
- The exchange highlights a philosophical split in the industry between groups that treat AI as tools under human responsibility and those that aim to build internal moral frameworks into models.
- Prominent AI researchers and religious figures publicly pushed back against attributing consciousness or spiritual status to models, and neither OpenAI nor Anthropic immediately responded to requests for comment.
- The debate shifts part of the safety conversation from technical failures to questions of public trust, procurement and regulation, which could affect enterprise choices, Anthropic’s reputation and its business plans.