Tech Companies Turn to Religion for AI Ethics, But Experts Warn Against Oversimplification

Tech Companies Seek Religious Guidance for Ethical AI Development

As the race to build powerful artificial intelligence intensifies, tech companies turn to religion for answers on teaching AI systems right from wrong. Anthropic co-founder Chris Olah articulated this approach at the Vatican earlier this year, stating that questions about AI ethics are not ones a lab can answer but rather issues that faith traditions have carried for millennia1

1

. This turn toward religion in AI reflects a broader recognition that developing morally good AI models requires input beyond technical expertise. Companies like Anthropic, OpenAI, Meta, Google, Amazon, and Microsoft are now actively engaging with philosophers and faith leaders to navigate the ethical challenges posed by increasingly powerful AI systems2

2

.

The Vision: Aggregating Humanity's Moral Views

The executives' philosophy behind this religious outreach centers on using diverse moral guidance for AI to create systems that reflect broad human values. OpenAI CEO Sam Altman articulated this vision last year, suggesting that ChatGPT should reflect a weighted average of humanity's moral view1

1

. Because these companies are not bound to a particular religion, they aim to aggregate moral ideas from multiple traditions to inform their models. Anthropic reported in a May blog post that they had consulted scholars, clergy, philosophers and ethicists from more than 15 religious and cross-cultural groups, with plans to broaden this pool further2

2

. This approach to AI and philosophy represents an attempt to ground technical development in millennia-old wisdom traditions.

Scholars Warn Against Treating Religion as a Rulebook

Despite these efforts at ethical AI development, experts caution that the approach risks fundamental oversimplification. Webb Keane, an anthropologist at the University of Michigan, argues that there's a conventional way of thinking of religion as a rulebook, which fails to account for the messy, practical, ad hoc ways adherents actually live their faith3

3

. The core issue is that religion cannot be easily separated from culture and politics, even though many assume it can be treated as a distinct set of extractable moral principles. Keane warns that using religion to train AI risks prioritizing certain traditions and distorting others, particularly faiths that do not see themselves primarily as systems of ethics. This conventional view is tied to scriptural religion, and scriptural religions ultimately turn out to be the Abrahamic religions, potentially marginalizing other faith traditions1

1

.

The Challenge of Contradictory Worldviews

Experts and adherents alike warn that the idea of aggregating religious moral guidance fails to reckon with how different traditions and worldviews often contradict one another. The notion that religion in AI can be simplified into compatible moral principles ignores the reality that faith traditions carry fundamentally different assumptions about existential questions ranging from whether humans have souls to practical matters like whether third graders should access chatbots for literacy education2

2

. Paolo Carozza, a member of the independent Meta Oversight Board, warned that combining traditions in AI could homogenize them rather than allow them to change organically through interaction3

3

. This risk of homogenization represents a significant concern as companies attempt to navigate cultural complexities in developing ethical frameworks for AI systems.

Inside Anthropic's Religious Engagement

Meghan Sullivan, a philosopher of religion who leads an initiative on AI ethics at the University of Notre Dame, offers an insider perspective on how these consultations actually work. Sullivan attended an Anthropic meeting earlier this year that included Christians from different denominations and was impressed by the rigorous engagement with religion, noting that Anthropic is widely seen as the AI company taking ethics and safety most seriously3

3

. The meetings are split into two parts: first, the company explains their models in layman's terms to attendees, then faith leaders respond with feedback and discussions. Sullivan described the culture inside Anthropic as similar to that on a good college campus, with intellectually curious people engaging in nerdy discussions about every kind of moral question the technology raises1

1

. This thoughtful approach suggests that at least some companies are making genuine efforts to engage with the depth and complexity of religious traditions.

What This Means for AI's Future

The turn toward religious consultation for AI ethics signals recognition that technical solutions alone cannot address the profound moral questions raised by increasingly powerful AI systems. Watch for how this engagement evolves: will companies move beyond treating religion as extractable moral principles toward deeper engagement with doctrine, ritual, and relationships? The short-term implication is that AI systems may begin reflecting more diverse ethical inputs, but the long-term concern is whether this approach will genuinely honor the cultural complexities of different faith traditions or simply create a homogenized ethical framework that distorts the very wisdom it seeks to capture. As AI capabilities expand, the question of whose morality gets encoded into these systems becomes increasingly urgent for users, developers, and society at large.