Religious scholars and Anthropic are in talks

The question of who influences the values embedded in powerful systems has a surprisingly long history. Before artificial intelligence, this question arose around textbooks, television, news media, and search engines. Each time a technology became consequential enough to shape how large populations think, the question surfaced: whose values will be baked in? Religious institutions have a particular stake in the answer because they claim expertise in questions of meaning, ethics, and how humans should treat one another. When a technology begins to advise people on moral questions, faith leaders reasonably ask whether their traditions are represented in that counsel. This matters because AI language models are not neutral. They are trained on human text, which means they absorb the values, assumptions, and priorities of whatever human-created material they learned from. They are also engineered by humans who make choices about what outputs are acceptable, what the system should refuse, and what tone it should adopt. Those engineering choices reflect decisions about values. The further an AI system's reach extends into people's everyday thinking, the more visible those embedded values become. Religious scholars study not just theology but also ethics, philosophy, anthropology, and the foundations of how communities decide what matters. That expertise becomes relevant when an AI system might eventually answer questions about how to live, whether a choice is moral, or what a person should do in a dilemma. A rabbi, imam, monk, or theologian brings centuries of tradition and rigorous ethical reasoning to those conversations. They also bring awareness of how different faith communities approach the same question differently, and why that diversity of perspective might matter for a tool that serves global populations. The broader conversation sits at the intersection of artificial intelligence development and what scholars call value alignment. This is the challenge of ensuring that AI systems behave in ways that reflect human values, particularly the values of the people most affected by the system. In practice, this is genuinely difficult. Different cultures and traditions disagree about what matters. A choice that honors one value might contradict another. There is no formula that resolves this automatically. Engagement between technology makers and traditional knowledge holders is not new, but it is growing more structured. As AI systems become more capable and visible, companies developing them are increasingly convening conversations with experts in ethics, philosophy, theology, and social science. These conversations are not designed to turn engineers into theologians or vice versa. Rather, they surface different perspectives on hard questions: What should this system refuse to do? How should it explain itself to someone who asks for moral guidance? What assumptions does it carry, and are those assumptions universal or culturally specific? The