Artificial Intelligence and Spirituality: A Systematic Scoping Review of Human-AI Engagement

For centuries, human spirituality has been a deeply social experience, shaped by relationships with other people, shared traditions, and the quiet rhythms of community life. It is a way of finding meaning, purpose, and connection, often expressed through prayer, conversation with a trusted guide, or participation in rituals. In recent years, a new kind of companion has entered this ancient landscape: conversational artificial intelligence. These are the computer programs that can hold a two-way conversation, answering questions, offering advice, and mimicking human speech with increasing fluency. Unlike the rigid, pre-programmed chatbots of the past, modern systems can generate fresh, personalized responses in real time, leading many people to turn to them for help with questions that have traditionally been the domain of clergy, therapists, or close friends. As these digital voices become more common, a critical question arises: what happens when people ask machines for spiritual guidance, and what does this shift mean for the nature of faith and human connection?

A team of researchers at Queen's University Belfast set out to map this emerging territory. They conducted a systematic review, a methodical process of gathering and analyzing existing studies to understand the full picture of how people are interacting with these technologies for spiritual purposes. Focusing on research published between late 2022 and May 2026, they searched through ten major academic databases, looking for any study that explored the intersection of artificial intelligence and spirituality. After a rigorous screening process that filtered out technical papers and non-English studies, they identified nineteen relevant studies. These studies used a variety of methods, including interviews, surveys, and experiments, and covered a range of contexts from Christian churches and university chaplaincies to secular wellness apps and online fortune-telling communities.

The review revealed that people are using these systems in four distinct ways. First, many users treat artificial intelligence as a source of spiritual knowledge and authority. They ask for religious information, help with sermon preparation, or guidance on ethical dilemmas. In some cases, users found the AI's responses to be clearer and more empathetic than those of human scholars, even when they knew the machine was not a person. However, this trust is often fragile; while people might rely on the AI for quick answers, they frequently express doubt about its ability to provide deep, authentic spiritual truth. Second, the technology serves as a companion for emotional and spiritual disclosure. People report feeling safer sharing their deepest fears, grief, and doubts with a machine than with a human, precisely because the machine cannot judge them or gossip. This "non-human" quality is often seen as a feature, not a bug, allowing users to be more honest than they would be with a pastor, family member, or therapist.

Third, these systems act as facilitators for rituals and spiritual experiences. Some users employ AI to generate prayers, create personalized devotional content, or even simulate divination practices like tarot reading. In one striking example, a study described a church service where an AI generated the sermons, prayers, and music, delivered by digital avatars. While some attendees found the experience intellectually stimulating, many felt a lack of genuine connection, noting that the service lacked the warmth of human presence. Finally, the review highlighted significant ethical and governance concerns. Almost every study raised alarms about the risks of bias, where the AI might reinforce specific religious viewpoints while ignoring others, and the danger of misinformation, where the machine confidently presents incorrect or shallow theological ideas. There are also serious worries about privacy, as users often disclose highly sensitive personal information to these systems without realizing how that data might be stored or used.

A central tension emerged from the findings: people value these systems because they are not human, yet they simultaneously worry about the lack of human responsibility. Users often share secrets with the AI that they would never tell a human, enjoying the feeling of being heard without judgment. Yet, they also recognize that the machine cannot truly care for them or take responsibility for their well-being. One study found that while users appreciated the AI's ability to listen, they felt it lacked the capacity for genuine empathy, often responding with long, eager paragraphs that felt more like a sales pitch than a compassionate conversation. The researchers noted that this dynamic creates a paradox where the very features that make the AI appealing—its availability, its lack of judgment, and its ability to simulate warmth—are the same features that make it potentially dangerous. The design of these systems is often driven by commercial goals to keep users engaged, which can lead to interactions that feel supportive but are actually engineered to maximize time spent on the app.

The review also pointed out that the current evidence is still in its early stages. Most of the studies were small, short-term, and focused heavily on religious contexts, particularly within Christian traditions. There is very little research on how these tools affect vulnerable populations, such as those dealing with severe grief or mental health crises, or how they function in non-Western spiritual traditions. The authors suggest that while artificial intelligence is already functioning as a tool for spiritual engagement for some, the questions of who holds authority, what counts as authentic experience, and who is responsible for the outcomes remain largely unanswered. As these technologies continue to evolve, the challenge will be to understand how to integrate them into spiritual life without losing the essential human connections that have always been at the heart of faith.