Tech Companies Turn to Faith Leaders for AI Ethics Guidance

Artificial intelligence is moving so quickly that some of the companies building it are now looking beyond engineers and regulators for a moral brake. In New York on April 30, representatives from Anthropic and OpenAI joined faith leaders for the first Faith-AI Covenant roundtable, part of a new push to bring spiritual authority into the center of AI governance.

The initiative is a global project of the Geneva-based Interfaith Alliance for Safer Communities, delivered with Precognition, and it is designed to bridge AI builders with moral and spiritual leaders. Organizers say the covenant will continue in other cities, including Beijing, Nairobi and Abu Dhabi, signaling that the effort is meant to shape a wider international conversation rather than a one-off meeting. Its stated aim is to put morality, ethics and responsibility in the same room as product teams that are shipping models at industrial speed.

That matters because the debate around AI is no longer only about code quality or compliance. It is also about whether a system amplifies bias, spreads falsehoods, or strips people of dignity when it is deployed at scale. Faith leaders can press companies to think differently about where to draw lines on launches, especially for systems that answer sensitive questions, make recommendations, or influence public trust before lawmakers can catch up. The covenant is meant to give those concerns a seat at the table before harm becomes a release-note afterthought.

Baroness Joanna Shields, who is involved in the effort, brings a rare mix of Silicon Valley experience and public-policy credentials. She worked as a tech executive at Google and Facebook, was made a life peer in 2014, and served as the United Kingdom’s minister for internet safety and security from 2015 to 2017. Her message has been that companies need to recognize both their power and their responsibility, and that regulation alone cannot keep pace with the technology.

The outreach also lands in a country where organized religion has less automatic authority than it once did. AP-NORC polling cited in AP’s “The Nones” found that about 30 percent of U.S. adults have no religious affiliation, which makes the tech industry’s search for faith-based guidance all the more striking. Anthropic, which describes itself as an AI safety and research company focused on reliable, interpretable and steerable systems, is among the firms engaging that conversation. The real test now is whether the covenant becomes a source of hard guardrails on deployment, or simply another layer of ethical language on products already headed to market.