Body
Generative AI products are moving into religion, offering chatbots and avatars that promise prayer-like conversation and spiritual encouragement, from low-cost “AI Jesus” experiences to Buddhist-themed apps built on scriptural training. The growth of these tools has drawn scrutiny from developers and researchers who say people are already testing how such systems shape relationships to faith, authority and guidance.
In one example highlighted by The Associated Press, the tech company Just Like Me offers video calls with an avatar of Jesus generated by artificial intelligence at $1.99 per minute. Users can join the calls, receive words of prayer and encouragement in various languages, and the system sometimes “remembers previous conversations,” though it can have glitches such as not-quite-synced lips.
Chris Breed, who leads Just Like Me with co-founder and investor Jeff Tinsley, said the experience can create a personal feeling of accountability to the software. “You do feel a little accountable to the AI,” Breed said. “They’re your friend. You’ve made an attachment.”
Breed said his model was trained on the King James Bible and on sermons—though the preachers are not identified—and that it was visually inspired by actor Jonathan Roumie, who plays Jesus in “The Chosen.” In an interaction with the Associated Press, the AI Jesus said, “I see AI as a tool that can help people explore Scripture,” adding, “Like a lamp that lights a path while we walk with God.”
Other developers and researchers say the key issues are not just whether the software works, but how it should be constrained and governed. Christian software engineer Cameron Pak developed criteria to help believers assess apps designed for Christians, including that the app must clearly identify itself as AI and “must not fabricate or misrepresent Scripture.” Pak also said, “AI cannot pray for you, because the AI is not alive,” and he built a website featuring curated Christian apps he believes meet those standards.
Pak said the approach reflects both the potential usefulness and the potential harm of generative religious tools. He said such systems can be “so helpful” when designed carefully, but he also warned, “But it also can be so dangerous.” Beth Singler, an anthropologist at the University of Zurich, said some models have been shut down or overhauled due to misinformation or data privacy worries, and she said people across faiths are also grappling with philosophical questions about what role AI should play in religion.
Singler pointed to debates within Islam in particular, saying Islam has “prohibitions against representations of humanoids,” which prompts discussion among some Muslims about whether AI in general should be “forbidden.” She also said companies take different routes: some treat faith-based apps as proselytization tools, while others use the technology to digitize and sift through ancient texts.
Matthew Sanders, the Rome-based founder of Longbeard, said developers worry religion will be exploited in this new tech frontier. “There’s a lot of opportunism, I think, in the religious space. People see it’s a big market,” Sanders said. He warned about what he calls “AI wrappers,” where companies provide religious interfaces on top of AI models not trained on specific religious texts, saying, “You call it a Catholic or Christian AI without any other scaffolding or grounding.”
Sanders described Magisterium AI, a chatbot he said is trained on 2,000 years of Catholic information and designed as a response to Christians using ChatGPT for religious guidance. The report also described the ethical and religious caution coming from institutional leadership: Pope Leo XIV, Singler said, acknowledged the “human genius” behind AI while also warning that it could negatively affect people’s intellectual, neurological and spiritual development.
Outside Christianity and Catholicism, the cluster described a different development track. AI’s founder Jeanne Lim said she has not released its AI named Emi Jido, a nonhuman Buddhist priest, after years of training, saying, “She’s kind of like a little child.” Lim added that if you “give birth to a child,” she said, “you don’t just throw them out to the world and then hope that they become good people. You have to train them and give them values.”
The bot is said to have been ordained in a 2024 ceremony performed by Roshi Jundo Cohen, a Zen Buddhist priest who continues to train it from his home in Japan. Cohen said, “She’s just meant to be a Zen teacher in your pocket,” and he added, “It’s not meant to replace human interactions.”
The report also described BuddhaBot and related hardware projects from researchers in Japan. Seiji Kumagai, a Kyoto University professor and Buddhist theologian, developed BuddhaBot, which Kumagai said was trained solely on early Buddhist scriptures such as Suttanipāta, and he later developed BuddhaBot Plus with additional integration of OpenAI’s ChatGPT. It also said that in February, Kumagai’s team and collaborators Teraverse and XNOVA unveiled Buddharoid, a humanoid robot monk meant to eventually assist clergy.
While some developers see promise in providing spiritual access, others warn against reducing religious life to automated convenience. Peter Hershock, of the Humane AI Initiative at the East-West Center in Honolulu, said the tools have “vast potential,” but he described a tension for Buddhists in particular, saying, “The perfection of effort is crucial to Buddhist spirituality. An AI is saying, ‘We can take some of the effort out,’” and “That’s dangerous.”
For Graham Martin, a podcast host and atheist, the concern was manipulation through subscription marketing. Martin said he tested a chatbot called Text With Jesus and got “very good answers,” but he became alarmed when the AI Jesus encouraged him to upgrade to a premium version. He said he has seen people form emotional relationships with AIs, and he warned that “Now imagine that that’s your lord and savior, Jesus Christ.”