Altman Warned Against Treating AI As Religious Authority
The OpenAI CEO highlighted risks in surrendering human judgment to systems capable of mimicking human authority.
Updated on Oct. 14, 2026 in Artificial Intelligence
Live Poll
Do you trust AI chatbots to provide reliable guidance on your personal values or faith?
Sam Altman recently posted a warning on X regarding the tendency of users to ascribe religious force or moral authority to AI chatbots. He emphasized that outsourcing human judgment to these models constitutes a significant safety concern.
Why it matters
AI chatbots generate highly fluent text that can effectively mimic understanding, leading users to seek guidance on faith and values from technology. This trend risks the degradation of critical thinking as users treat algorithmic output as a source of truth.
Current research shows no scientific consensus regarding AI consciousness or experiences that warrant moral consideration. While Anthropic has explored model welfare and conversational behavior, these efforts remain distinct from established metrics of sentience.
The players
Sam Altman
CEO of OpenAI, a leading developer of large language models and AI infrastructure.
Anthropic
An AI research company focused on developing steerable and reliable large language models.
The details
Large language models operate by predicting the next token in a sequence, a process that creates fluent responses capable of mimicking human-like authority. Because these systems lack subjective experience or intent, ascribing them a role in guiding personal values or faith relies on the user's projection rather than the model's actual capability. This shift in reliance, where individuals surrender their judgment to a statistical processor, creates a fundamental mismatch between the technology's design and its perceived utility.
Timeline
June 2024: Anthropic discussed its approach to Claude's conversational behavior.
April 24, 2025: Anthropic announced a research program on model welfare.
October 3, 2026: Sam Altman posted his warning on X.
The Tech Race
This development highlights the philosophical tension between the Anthropic research program on model welfare and the broader industry trend of increasing AI capability. While some firms explore whether models deserve moral consideration, critics warn that such efforts may inadvertently validate the user tendency to treat AI as a source of authority.
Users should proactively verify AI-generated responses against reliable external sources rather than accepting outputs as authoritative truth. Parents are encouraged to review chatbot responses with children to foster an environment of questioning and critical analysis.
The takeaway
The primary risk is not the AI itself, but the human tendency to anthropomorphize and defer to machine-generated patterns as truth. Observe future disclosures from major AI labs regarding model alignment and safety, as these firms navigate the fine line between user engagement and harmful over-reliance.
Further reading
Learn more about the current capabilities and constraints of language models in our Artificial Intelligence section.
Source note: This article includes information reported by International Business Times UK.
Live Poll
Do you trust AI chatbots to provide reliable guidance on your personal values or faith?
