Anthropic Seeks Religious Guidance for Claude’s AI Ethics

Anthropic is reportedly seeking religious thinkers worldwide to help shape Claude’s moral framework, according to The New York Times. The AI company has already consulted faith leaders as part of its constitutional AI approach, which uses written principles to guide Claude’s behaviour and responses to difficult requests. In March 2026, Anthropic convened 15 Christian leaders to discuss artificial intelligence ethics. Catholic thinkers contributed to principles included in Claude’s “Constitution”, which aims to help the model act as a “good, wise, and virtuous agent”. Anthropic says it does not want to adopt a single ethical doctrine. Anthropic executives also joined the April 2026 Faith-AI Covenant roundtable, alongside OpenAI representatives and faith organisations. However, it remains unclear whether the reported global hiring effort has become a formal programme. Anthropic’s documented work so far has mainly involved consultations, meetings and roundtables. The initiative follows earlier engagement with the Vatican after its 2025 AI document, “Antiqua et Nova”. While Anthropic has so far relied heavily on Christian and Catholic perspectives, the company’s broader challenge will be ensuring that Claude’s AI ethics reflect other religious traditions and secular philosophies. The move could make ethical credibility and AI safety important points of differentiation among major AI developers.