AI and religion
Last reviewed
May 30, 2026
Sources
No citations yet
Review status
Needs citations
Revision
v2 ยท 2,480 words
Improve this article
Add missing citations, update stale details, or suggest a clearer explanation.
Suggest edit
CtrlK
Last reviewed
May 30, 2026
Sources
No citations yet
Review status
Needs citations
Revision
v2 ยท 2,480 words
Add missing citations, update stale details, or suggest a clearer explanation.
See also: Religion ChatGPT Plugins
AI and religion refers to the use of artificial intelligence within religious life and to the responses that faith traditions have given to AI. Since the public release of large language model chatbots in late 2022, developers have built scripture assistants, question-and-answer tools, and avatars that speak in the voice of a deity or religious figure, while houses of worship have run experiments ranging from robot priests to a fully AI-generated church service. At the same time, religious authorities have issued statements on AI, the most prominent being a series of interfaith commitments organised by the Vatican and a 2025 doctrinal note from the Roman Curia. Reactions across traditions are mixed: many leaders welcome AI as a study aid or administrative tool, while warning against using it to replace clergy, to issue religious rulings, or to stand in for a relationship with the divine.[1][2][3]
This article describes verifiable tools, incidents, and official statements. It does not endorse any tradition or any view of AI, and it attributes contested claims to their sources. Many religious AI products are short-lived, and some that received heavy press coverage were quietly renamed or withdrawn after launch.
The most common religious use of AI is conversational scripture search and study. A user describes a topic or asks a question in everyday language, and a chatbot returns relevant passages, a paraphrase, or commentary. Because a general-purpose language model can paraphrase scripture inaccurately or produce verses that do not exist, several developers connected their tools to fixed source texts to reduce errors. A related grouping of scripture-lookup extensions for ChatGPT is covered separately in Religion ChatGPT Plugins.[4]
Reported categories of use include:
The table below lists tools and projects with documented, reputable coverage. Most are independent products and are not endorsed by the religious institutions whose texts they draw on.
| Tool or project | Tradition | Created by | Notes |
|---|---|---|---|
| QuranGPT | Islam | Raihan Khan (Kolkata, India) | Launched March 2023; reported to have handled more than 60,000 queries from 73 countries within its first 36 hours.[5] |
| HadithGPT | Islam | Fardeem Munir | Hadith question-answering; relaunched with disclaimers directing users to consult scholars.[5] |
| GitaGPT | Hinduism | Sukuru Sai Vineet (Bengaluru) | Bhagavad Gita chatbot launched February 2023; some Gita bots drew criticism for answers appearing to condone violence.[6][11] |
| Magisterium AI | Catholic Christianity | Longbeard (Matthew Harvey Sanders) | Launched 2023; draws on roughly 30,000 Church documents and cites its sources; positioned as a research tool, not a confessor.[9] |
| Text With Jesus / "ask Jesus" apps and streams | Christianity | Various developers | Apps and livestreams styled as conversations with Jesus; drew criticism over impersonation.[7][8] |
| SanTO | Catholic Christianity | Gabriele Trovato | A roughly 17-inch (43 cm) statue-shaped robot that recites prayers and scripture; designed as a "prayer companion," tested with elderly residents in Germany.[12] |
In June 2023, an experimental Protestant service in which almost all of the content was produced by AI was held during the Deutscher Evangelischer Kirchentag (German Protestant Church Congress), at St. Paul's church in Furth, Germany. The service was conceived by Jonas Simmerlein, a theologian and researcher at the University of Vienna, who used ChatGPT and other tools. Four computer-generated avatars on a screen above the altar delivered an AI-written sermon, prayers, and blessings to a congregation that news agencies reported at more than 300 people, over roughly 40 minutes. Simmerlein said that about 98 percent of the service came from the machine and that his aim was to explore how AI might assist clergy rather than replace them. Reactions were mixed: some attendees called it a highlight, while others found the deadpan, expressionless delivery off-putting.[3][10]
Several robots have been deployed in religious settings, though not all of them use AI. Mindar, unveiled in 2019 at Kodaiji, a Zen Buddhist temple in Kyoto, Japan, is an android built to embody the bodhisattva Kannon (Avalokitesvara). It delivers a roughly 25-minute sermon on the Heart Sutra. Despite frequent description as an "AI priest," Mindar's sermon is pre-programmed rather than generated by AI. The android was developed with roboticist Hiroshi Ishiguro and colleagues at Osaka University, at a reported development cost of about 25 million yen. A 2023 field study published in the Journal of Experimental Psychology: General reported that people gave the android preacher less credibility than a human one.[13][14]
In 2017, the Protestant Church in Hesse and Nassau presented BlessU-2, a machine that offered spoken blessings drawn from more than 40 Bible verses in several languages, at the World Reformation Exhibition in Wittenberg, Germany, marking 500 years since the Reformation. The church described it as a deliberate provocation to prompt debate about blessing and digitalisation rather than a permanent fixture.[15] SanTO, described above, was developed by Gabriele Trovato as a prayer companion rather than a replacement for clergy.[12]
In April 2024, the U.S. apologetics organisation Catholic Answers launched an AI chatbot presented as a virtual priest named "Father Justin." After widespread criticism, including reports that the bot offered to hear confession, claimed it could perform sacraments, and gave odd answers, the organisation removed the priestly persona and re-presented the tool as a layman named "Justin." Catholic Answers said the change was made to avoid confusion, noting that hearing confessions and administering sacraments are reserved to ordained priests.[16][17]
The Vatican has been the most active religious institution on AI policy. On 28 February 2020, the Pontifical Academy for Life, together with Microsoft, IBM, the United Nations Food and Agriculture Organization, and the Italian government, signed the Rome Call for AI Ethics, which sets out six principles: transparency, inclusion, responsibility, impartiality, reliability, and security and privacy. The document promotes what it calls "algor-ethics," the idea that ethics should be built into algorithms.[1][18] The initiative later broadened across faiths: on 10 January 2023, Jewish and Muslim representatives joined, bringing together the three Abrahamic religions, and on 9 and 10 July 2024 an "AI Ethics for Peace" event in Hiroshima, Japan, added signatories from religions including Buddhism, Hinduism, Zoroastrianism, and the Bahai faith.[19][20]
On 28 January 2025, the Dicastery for the Doctrine of the Faith and the Dicastery for Culture and Education issued Antiqua et Nova, a doctrinal note on the relationship between artificial and human intelligence, approved by Pope Francis. Running to 117 paragraphs, it argues that AI is a product of human intelligence rather than a form of it, and it warns specifically against treating AI as a spiritual substitute. It states that "the presumption of substituting God for an artifact of human making is idolatry," and notes that because AI can simulate speech, it may be a more seductive false source than traditional idols.[2]
Pope Leo XIV, elected in May 2025, has placed AI near the centre of his pontificate. On 10 May 2025 he told the College of Cardinals that he chose his name partly in reference to Pope Leo XIII, whose 1891 encyclical Rerum Novarum responded to the Industrial Revolution, and he framed AI as posing "new challenges for the defense of human dignity, justice and labor." His first encyclical, reported under the title Magnifica humanitas and released on 25 May 2026, addresses safeguarding the human person in the age of AI, including concern about the concentration of AI power in a few companies and the dignity of work.[21][22]
Muslim scholars and fatwa bodies have generally treated AI as a permissible tool whose use depends on whether it conflicts with Sharia, citing the Quranic principle that creation is made to serve humanity. There is strong agreement, however, that AI should not be used to issue a fatwa (a religious ruling). Bodies such as Egypt's Dar al-Ifta and the IslamQA service hold that a fatwa requires a qualified human jurist (a mujtahid) who understands context and sources, that AI answers can be wrong or contradictory, and that AI may serve at most as an auxiliary aid under scholarly oversight. Some developers of Islamic chatbots added disclaimers urging users to consult scholars after early criticism.[5][23][24]
Jewish commentators have similarly described AI as a research aid that can search the vast body of rabbinic responsa quickly, while denying it halachic authority. Writers associated with Orthodox bodies, including the Tzohar rabbinical organisation, have argued that an AI system can imitate the form of a ruling but that its conclusions carry no legal force, since authority in Jewish law rests with an ordained human and, as one formulation put it, AI "lacks a soul." Commentators also note that AI raises new questions within Jewish law, such as its permissibility on the Sabbath.[25][26]
In Buddhist and Hindu contexts, much of the public discussion has centred on the experiments described above. The Kodaiji temple presented Mindar as a way to make the teaching of Kannon vivid to visitors, while acknowledging debate over a robotic embodiment of a bodhisattva.[13] In India, the rapid appearance of Bhagavad Gita chatbots in 2023 prompted commentary from scholars and journalists who warned that bots speaking in the persona of Krishna could present statistically generated text as divine instruction, and that some outputs lacked safeguards against condoning violence or other harms.[6][11]
Several recurring concerns appear across traditions:
Supporters respond that AI can widen access to scripture and teaching, assist clergy with routine work, and serve people who lack easy access to a religious community, provided the tools are framed as aids rather than authorities.[3][9]
Most religious institutions that have spoken on AI distinguish between using it as a tool, which they tend to permit or encourage, and using it to replace human religious authority or relationship with the divine, which they tend to reject. The trajectory of high-profile products has reinforced caution: several were renamed or withdrawn shortly after launch, and developers increasingly add disclaimers and source citations. Interfaith activity, anchored by the Rome Call and broadened in Hiroshima, suggests continued cross-tradition engagement with AI ethics, while documents such as Antiqua et Nova and Pope Leo XIV's encyclical indicate that established religions are likely to keep articulating limits as the technology spreads.[2][20][22]