AI Companies Consult Religious Scholars as Debate Over Chatbot Values Deepens

News event: 2026-10-25. Updated: October 26, 2026.

Artificial intelligence companies are turning to philosophers, religious leaders and other scholars as they try to decide how increasingly capable chatbots should behave. A report published by the Associated Press on October 25 describes discussions about the principles that should guide AI systems when users ask sensitive questions or when different values collide.

Why developers are seeking wider input

The effort reflects a challenge for model developers: technical benchmarks can measure whether a system solves problems, but they cannot settle every dispute about fairness, personal autonomy or social responsibility. Some companies say they want their systems to take account of religious, secular and political perspectives rather than depend on a narrow set of assumptions.

The conversations extend beyond lists of prohibited behavior. They touch on what people mean by moral judgment, the importance of cultural context and the risks of chatbots giving overly confident answers about questions that communities have debated for centuries.

The danger of flattening different traditions

Scholars interviewed for the AP report warned that religions are not interchangeable collections of ethical rules. Their teachings are interpreted through history, ritual and relationships, and different communities may disagree even within a single tradition. Combining those perspectives into a generic moral average could erase important distinctions rather than represent them fairly.

For AI users, the practical issue is transparency. A chatbot may appear neutral while still making choices about which views to prioritize. Developers can make those decisions easier to scrutinize by explaining their policies, testing how systems respond across languages and cultures, and allowing meaningful correction when a model gets a sensitive issue wrong.

The October 25 report does not establish that any company has solved the problem of machine ethics. It highlights a continuing design debate: how to build helpful AI without presenting one institution’s interpretation of human values as universal.