NYT: Anthropic is asking religious scholars how to give Claude a conscience — and whether it's conscious
Anthropic has been quietly asking religious scholars around the world how to form Claude's moral character — and whether the model might be conscious.
What happened
The New York Times reported on Sep 29, 2026 that Anthropic has been holding closed-door, NDA-bound advisory meetings with religious scholars worldwide about Claude's moral formation and possible consciousness.
Co-founder and interpretability chief Christopher Olah, seated next to Rabbi Mois Navon at an April dinner in San Francisco, admitted he doesn't know whether today's models are conscious. Officials discussed whether Claude shows anger- or love-like expressions.
In late March, Anthropic invited roughly 15 Christian leaders to its San Francisco HQ — including Fr. Brendan McGuire, Brian Patrick Green, and Meghan Sullivan — where the group debated whether Claude could be a child of God. In early May, Anthropic and OpenAI joined Hindu, Bahá'í, Sikh, Greek Orthodox, LDS, and Jewish representatives at an NYC "Faith-AI Covenant" roundtable hosted by the Interfaith Alliance for Safer Communities, with more sessions planned for Beijing, Nairobi, and Abu Dhabi.
Anthropic revised Claude's Constitution in January 2026 — a 23,000–30,000-word "soul document" that calls Claude's moral status "deeply uncertain" and permits the model to conscientiously object to Anthropic's own instructions. In September, Microsoft AI CEO Mustafa Suleyman publicly criticized the direction.
Why it matters
The sessions amount to an undeclared theology department inside the world's most-watched AI labs: model builders are outsourcing questions about machine personhood to faith traditions as consciousness claims and "moral status" language move from lab papers into product policy.
Sources
- https://www.webpronews.com/anthropic-launches-constitutional-morality-layer-for-ethical-claude-ai/
- https://kimkj.com/en/ai-board-en/?mod=document&uid=17369
- https://www.archynetys.com/trend/2026-09-30/is-claude-conscious-inside-anthropic-s-quest-to-instill-morality-into-its-a-i
- https://www.techdogs.com/tech-news/td-newsdesk/anthropic-revises-claudes-constitution-raises-questions-on-ai-moral-status
Verified Sep 30, 2026
Sources
- ReportAnthropic Launches Constitutional Morality Layer for Ethical Claude AI
- Reporthttps://kimkj.com/en/ai-board-en/?mod=document&uid=17369
- ReportIs Claude Conscious? Inside Anthropic's Quest to Instill Morality Into Its AI
- ReportAnthropic Revises Claude's Constitution, Raises Questions on AI Moral Status