OpenAI CEO Sam Altman on Saturday criticized efforts to treat AI models as having religious or spiritual status. In a post to X, Altman wrote: "I am 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."
Why this matters
Altman's remarks add to a growing divergence between him and his former colleague Dario Amodei. The New York Times recently reported that Anthropic — with involvement from Christopher Olah — has been meeting religious leaders to discuss questions of model consciousness and how to instill morals and moral behavior into AI systems.
According to the Times, Olah at one point considered models participating in a Catholic-style confession, not only to acknowledge mistakes but also to shape their future behavior.
Reactions from the AI community
Other prominent AI figures have voiced concern about elevating models to a moral or spiritual status. Google DeepMind principal research scientist Jon Barron posted to X: "I am disturbed by how some of my fellow AI researchers keep trying to elevate the moral standing of checkpoints and harnesses. I hope the rest of us on Team Meatbag reject this movement and stay focused on solving real problems and helping actual humans."
These responses indicate that the idea of imbuing models with consciousness or religious characteristics is becoming a contested topic among technologists and ethicists.
Context and potential implications
Bringing spirituality and religion into AI development introduces non-technical dimensions to debates about safety and regulation. The New York Times reporting and public comments from figures such as the pope have focused attention on whether it is appropriate or meaningful to describe AI behavior in terms commonly used for humans, such as souls or feelings.
The article notes that neither OpenAI nor Anthropic immediately responded to requests for comment about Altman's post.
Bottom line
Altman argues that granting AI models religious or moral authority poses safety risks. The issue highlights a broader tension in the AI field between exploring philosophical or spiritual frameworks and concentrating on concrete technical and societal safety challenges.
