Religious leaders meet OpenAI and Anthropic to discuss a 'Faith‑AI Covenant' on ethics

Artificial intelligence companies are starting to court a different kind of power center — not just regulators, investors, or academics, but clergy. That is the real news here. On April 30, representatives from OpenAI and Anthropic sat down in New York with leaders from multiple faith traditions for the first “Faith‑AI Covenant” roundtable, a new initiative meant to shape ethical norms for AI before the technology gets even more embedded in daily life. (natlawreview.com) ### What actually happened? (natlawreview.com) The meeting was convened by the Geneva-based Interfaith Alliance for Safer Communities, working with Precognition and its CEO, Baroness Joanna Shields. The stated goal was simple but ambitious — bring AI companies, faith institutions, civil society, and academics into the same room to talk about the values that should guide AI development and deployment. (fastcompany.com) OpenAI and Anthropic were there alongside groups including the Baha’i International Community, the Sikh Coalition, the Archdiocese of Newark, the Greek Orthodox Archdiocese of America, the New York Board of Rabbis, and representatives tied to the Latter-day Saints. ### Why are AI labs doing this now? Because the old AI-governance script is not enough anymore. Safety papers, model cards, and Washington hearings help, but they do not answer the broader question of who gets to define acceptable behavior for systems that increasingly shape knowledge, work, and judgment. (natlawreview.com) Shields put the problem bluntly — regulation cannot keep up. So the labs are widening the circle, looking for institutions that already have public credibility on moral questions. ### Why faith leaders? Faith groups bring something tech companies do not have — longstanding moral authority, civic networks, and direct relationships with billions of people. That does not mean they suddenly become AI engineers. It means they can pressure the debate toward human dignity, labor, bias, manipulation, and the boundaries of acceptable machine influence. (natlawreview.com) Basically, the companies are discovering that “can we build this?” is a narrower question than “should people trust this?” ### Is this about religion controlling AI? Not really. The stated aim is not sectarian doctrine baked into software. It is a cross-faith process to develop norms or principles that companies might voluntarily follow. That matters. A covenant is softer than regulation but broader than a corporate ethics memo. (fastcompany.com) It is trying to create social license — a sense that AI is being shaped in conversation with institutions outside Silicon Valley. ### What makes this more than a one-off photo op? The organizers say New York is only the first stop. More roundtables are planned through 2026 in Beijing, Bengaluru, Nairobi, Paris, Singapore, and Abu Dhabi. That turns the event from a symbolic meeting into an attempt at a global process. The catch is that global moral consensus is much harder than hosting one roomful of dignitaries. (natlawreview.com) Different faiths disagree with each other, and governments and companies will have their own agendas too. ### What are the hard questions underneath this? One is whether moral guidance can stay meaningful once it gets translated into product decisions. Another is whether companies want genuine constraint or just broader legitimacy. And a third is who gets left out — because if faith leaders are in the room, labor groups, teachers, artists, and communities most affected by automation will also want a say. (fastcompany.com) None of that kills the idea, but it does show the limits of a covenant without enforcement. ### So what is the bottom line? This meeting matters less for any immediate policy change than for what it signals. OpenAI and Anthropic are acting like AI governance is now a legitimacy problem, not just a technical one. Once companies start asking religious leaders to help define the guardrails, they are admitting something important — the fight over AI is no longer only about capability. (natlawreview.com) It is about who gets to say what kind of future is morally acceptable.