Wisdom of the Heart: A Contemporary Review of Religion and AI

Abstract

With the rapid developments of artificial intelligence across all areas of society, especially within the last two years, the present paper provides a much-needed focus on the relationship between AI and religion by providing a review of recently published academic works from 2024 to early 2025 in the area of AI and religion. The review focuses on 15 published research studies that both demonstrate a broad range of research interests in this field and are empirically and epistemologically sound and, as such, should be of interest for guiding further research in these important areas. The review concludes by placing the need for scientific and empirical research within a wider context of a Christian engagement with wisdom and creativity as a way for continuing academic progress in the exciting field of AI and religion.

1. Introduction

“The light shines in the darkness, and the darkness did not overcome it.”

(John 1, v, NRSV)1

Sir Karl Popper, possibly the greatest philosopher of science and epistemology of the 20th century, in his introduction to his 1963 book Conjectures and Refutations, lays out his argument for a critical rationalist approach to the fundamental problem of knowledge. He states, at the start of the preface that:

The way in which knowledge progresses, and especially our scientific knowledge, is by unjustified (and unjustifiable) anticipations, by guesses, by tentative solutions to our problems, by conjectures. These conjectures are controlled by criticism; that is, by attempted refutations, which include severely critical tests…Criticism of our conjectures is of decisive importance: by bringing out our mistakes it makes us understand the difficulties of the problem which we are trying to solve. This is how we become better acquainted with our problem, and able to propose more mature solutions: the very refutation of a theory—that is, of any serious tentative solution to our problem—is always a step forward that takes us nearer to the truth. And this is how we can learn from our mistakes.

(p. vii)

A guiding paper has been the Rome note, Antique et Nova, January 2025 (Dicastery for the Doctrine of the Faith and Dicastery for Culture and Education 2025). This note has been used as a framework for looking at the issue of religion and AI in this review, especially the section AI and Society (pp. 8–9) and AI and Education (pp. 12–14).

At the very top of a cognitive governing hierarchy is society as a whole. In Section 50 of the Antiqua note, this societal governance and relationship with and to technology is made clear, in the words of the late Pope Francis:

“the inherent dignity of each human being and the fraternity that binds us together as one human family must undergird the development of new technologies and serve as indisputable criteria for evaluating them before they are employed.”

It is of significance that Pope Francis calls on the importance of ‘criteria’ for evaluation. The criteria that Pope Francis calls to be used are the human ‘dignity’ of the individual and the ‘fraternity’ of one human family. They are criteria that can be used in evaluating current research into religion and AI and will be used as governing criteria in this literature review.

Section 51 acknowledges the positive contribution AI can make to society, specifically referencing ‘important innovations in agriculture, education and culture.’ (p. 9). As will be seen, current scientific work in religion and AI addresses, in differing ways, these innovations. In Section 52 a note of caution is maintained in the dangers AI can have in widening the ‘digital divide’ and worsening ‘existing social inequalities’ (p. 9). The economics of AI use, both its potential and its dangers, is an area that He (2024a, 2024b) specifically addresses. Sections 53 and 54 also address the ethical concerns of an AI dominated landscape whereby AI, ‘concentrated in the hands of a few powerful companies’ could, in the words of Pope Francis, possess the capacity to exercise, “forms of control as subtle as they are invasive, creating mechanisms for the manipulation of consciences and of the democratic process.” (p. 9).

Fundamentally the Vatican note puts at the heart of the debate about AI in society the question of human control and responsibility. Do we allow technology to control humanity, or do we continue to seek a way that the foundation of human responsibility comes from a recognition that ‘all human capacities, including the person’s autonomy, come from God and are meant to be used in the service of others?’ (my italics, p. 9).

Additionally, the section focusing on AI and education also helps with clarifying the vital role of educational institutions, specifically universities when it comes to the relationship between religion and AI. In Section 77 the note stresses that the aim of education is to ‘contribute to the person’s holistic formation’. Section 78 reminds us of the importance of ‘a commitment to cultivating the mind, but always as a part of the integral development of the person’. Section 83 focuses on the vital wider role of the university in today’s world as ‘great laboratories of hope at this crossroads of history’. The role of the Catholic university at this time should be in ‘guiding them (the fields of science and rationality) always towards ethically sound applications that clearly serve the cohesion of societies and the common good, reaching new frontiers in the dialogue between faith and reason’. The final note (89) quotes from the Dicastery for the Doctrine of the Faith Declaration Dignitas Infinita “If technology is to serve human dignity and not harm it, and if it is to promote peace rather than violence, then the human community must be proactive in addressing these trends with respect to human dignity and the promotion of the good.”

A further guiding principle for this literature review is to place the papers in the context of whether they demonstrate any insights into what Merlin Donald refers to as ‘cognitive metamorphosis’ and/or ‘cultural governance’. Donald (1999) cautions that ‘now (in the age of modern technologies such as AI) the actual operational tools by which we think are changing, and new demands are being made on our brains, which will reshape their basic functional organization’ (p. 23).

It is with these guiding ethical, epistemological and scientific principles that we may hopefully see more clearly by shining a light as we search for some tentative evidence of, and future directions about, the relationship between AI and religion.

2. Literature Review

In conducting this literature review and seeking to tease out some common findings from the past two years, it is insightful that there are actually relatively few academic papers and studies that have looked specifically at the role of AI and Catholic education. The overwhelming majority of studies have been conducted at a broader sociological level as opposed to a pedagogical level. In contrast, there has been a vast increase in the ‘secular’ research of using AI in education over a similar time scale. What is overwhelmingly clear, and extremely welcome, is that the literature reviewed all meet the Popperian critical rationalist approach to research, all seeking to provide conjectures which allow for further critical refutations.

A literature review using the criteria of open access journal reviews and focusing on SCOPUS, Science Direct and Compass databases, produced an initial 37 publications that met the broad criteria of published research on AI and religion between 2024 and April 2025. Of those 37 articles, 22 were eliminated from a final selection due to issues of either relevance to Catholic education in particular or wider scientific questions of validity, especially with referencing in the articles. The final selection of 15 articles for the literature review aims to cover a broad representation of current, relevant issues. In the Section 3, a summary of the papers and their relevance for future research interests is presented.

One of the most recent, comprehensive studies looking specifically at AI and religious education by Papakostas (2025) looks at both the ‘pedagogical advantages of AI, such as personalization and administrative efficiency, and the risks of theological distortion, overreliance, and epistemic conformity.’ (p. 1) and focuses on a number of ‘real-world implementations’ from both the Harvard Divinity School and the Oxford Centre for Digital Theology. The conclusions reached are important in terms of seeing the many positives of AI in religious education, the most important being ‘personalized learning, increased accessibility, and administrative efficiency’ (p. 17). At the application level, AI also ‘enables new forms of theological inquiry through tools such as natural language processing and adaptive learning systems’ (p. 17). The study also highlights the important element of ethical considerations with the use of AI, specifically mentioning three very important considerations of ‘ethical dilemmas regarding authority, privacy, bias, and educational dependence’ (p. 17).

In an extended study by Alkhouri (2024) looking at the role of AI in the study of the psychology of religion, evidence is presented from a detailed review of relevant literature that aims to ‘understand the transformation of traditional religious practices and raise important questions about authenticity, inclusiveness, and the role of technology in the psychology of religious contexts’ (p. 1). Such extensive reviews are essential to further research and investigation, and the Alkhouri (2024) study is an invaluable contribution to the field.

Yao (2024) provides a comprehensive account of both the opportunities and challenges of AI and religion. Covering five main focus areas, including AI in religion; AI in social studies; future landscapes and ethical and societal implications, Yao carefully explores these areas by referencing current research as well as highlighting the need for an ethical framework in a central section looking in more detail at interdisciplinary collaboration and ethical frameworks. Yao concludes that:

Balancing the benefits of technological innovation with the preservation of human values and dignity is essential for the responsible adoption of AI in shaping the future of religion and social studies.

(p. 340)

In two thorough empirical studies by He (2024a, 2024b), the author reaches some important and nuanced conclusions for religious freedom and the use of AI. He (2024b) ‘examines the influences of artificial intelligence and socioeconomic factors on religious freedom across 20 countries from 2000 to 2022’ (p. 1) and concludes that the ‘findings from our research indicate a negative impact of artificial intelligence on religious freedom’ (p. 7). Such a cautionary conclusion leads to the need for the emergence of specific policy implications. The first of these is that ‘governments are advised to develop comprehensive regulatory frameworks for artificial intelligence that carefully consider its potential effects on religious freedom’ (p. 7). The second policy recommendation is ‘to balance the embrace of technological advancements with the protection of religious freedom’ (p. 7).

A two-pronged approach to their research is taken by Tsuria and Tsuria (2024). In the first section they conduct a literature review that ‘aims to explore the evolving discourse in academic literature regarding the relationship between religion and AI, examining various perspectives, concerns, and potential implications stemming from this interaction.’ (p. 3). This review recognizes the limited yet growing research in this area and provides a comprehensive account of current trends based around four themes: new religious metaphors; new religious movements; religious uses of AI; and debates related to human existence and purposes (p. 3).

The second section focuses on two case studies undertaken by the researchers into using currently available AI tools to look at AI chat responses to questions on religion. The first case study used Claude AI in a dialogue focusing on some specific questions of Judaism while the second case study used ChatGPT and Microsoft Bing ‘to ask the AI questions that are more general in nature and bring up commonly held stigmas related to two religious traditions: Christianity and Islam’ (p. 5). The results of both case studies allowed the researchers to tentatively conclude that three main elements emerged from these interactions:

• Inability to represent complex religious issues.
• Emphasizing diversity of opinions.
• Calling the reader to engage with respect and sensitivity (p. 8).

The authors conclude that they ‘are cautiously optimistic that AI-generated-texts do carefully account for religious diversity and complexity’ (p. 10) while also cautioning that they are ‘also apprehensive about this ‘moralizing’ tendency in AI—if AI becomes our moral guide, where is the human intelligence in that?’ (p. 10).

In this comprehensive paper Trotta et al. (2024) investigate the ways in which ‘religious actors’ use and interact with AI in current ways, providing several accounts of specific use cases across a range of religious contexts. The paper concludes with a number of important future developments for possible research in this area. In the first theoretic section of the paper, the authors provide a review of prior ‘narratives and research on religions and artificial intelligence’ (p. 3) and cover a range of writers and researchers and their work and views on AI. The authors conclude that most of the narratives around AI ‘have focused their attention on narratives of imagined ai and its intersections with religious discourses’ (p. 6) and that often these discourses are ‘firmly integrated into techno-utopian ideologies which are gaining traction in today’s political and cultural landscapes’ (p. 6) and, finally, that these narratives ‘are suffused with vocabularies and imageries of salvation or damnation, immortality, omnipresence and omniscience which have their historical roots in religious traditions’ (p. 6).

The research findings are split into two distinct sections: firstly, the use of AI in religious practice and secondly, religious actors engaging with AI beyond religious practices. A range of use cases are introduced in the first section including AI for ‘preservation of “orthodoxy”’ (p. 10) in the Christian faith, giving the Bible Project as an example ‘in which AI-powered technology helps people to get acquainted with the stories of the Bible and relate the scriptural verses to their daily lives’ (p. 10). Other use cases from other faiths (Islam and Judaism) include prayer or devotional apps that ‘provide support with finding the right way to engage in prayers or rituals’ (p. 10). AI-powered robots aiding practice in the Buddhist and Protestant faiths are discussed as ‘the sphere of communion and community-building’ (p. 11), where AI is being deployed to identify locations for worship and track physical and virtual attendance at places of worship.

The second section outlines a number of thematic areas where religious actors are engaging with AI beyond religious practice. In the area of policy and advocacy engagement, the authors highlight the important work that is being conducted at the EU level on the engagement with AI and religious bodies, referencing dialogues the EU has undertaken with religious leaders in 2019 and the White Paper on Artificial Intelligence (European Commission 2020). The authors also focus on specific Catholic policy engagement such as the important 2019 COMECE publication (The Catholic Church in the European Union 2019), which looks at the ethical consequences of a deepening dependence of robotics in life as well as references the vital 2020 Rome Call for AI Ethics by the Pontifical Academy for Life (RenAIssance Foundation 2020). The references to these and other policy and engagement discourses and publications highlight the essential work that is being carried out by supra-governmental bodies in continuing to develop approaches to AI that include religious, ethical and moral voices. In their conclusion the authors state that ‘it is clear that religious actors can provide significant contributions in understanding and orienting the use of AI in development work.’ (p. 18).

Two fascinating and excellent studies that focus on the areas of archeology using AI and the history of robotic technology are deserving of more extended analysis than is available in this current review and are highly recommended. Çelik (2025) reports on the use of generative AI technology to investigate archeological infrastructures. By deploying AI tools, the author demonstrates that AI has ‘the potential for researchers and historians to leverage artificial intelligence technology to revive and comprehend the interior spaces of ancient architectural structures’ (p. 8). A highly readable account of the history of robotic technology and the influence of religion is presented by Geraci (2024). In the introduction, which brings to the fore the possible realities of a more robot-dominated world, Geraci argues that ‘(R)ecognizing the significance of religious communities, practices, and beliefs means accepting that human-equivalent robots would also see these things as significant.’ (p. 730).

A comprehensive book from Singler (2024) uses the sociological approach of a series of in-depth user case studies, covering a wide range of aspects of the connection between AI and religion including rejection, adoption, adaptation, transhumanism, new religious movements, posthumanism and a look to the future in a fascinating chapter on ‘Entanglements, Imaginaries, and Futures.’ This book provides an excellent introduction into the background and relationship between technologies and religion whilst encompassing a deeper exploration into the implications and potential consequences of the closer relationship between AI and religion. Indeed, in this final chapter the author concludes that ‘The future of religion and AI as a field may need a reflexive moment on ethics’ (p. 207).

The research question posed by Sierocki (2024): ‘how algorithms are understood, perceived, and imagined, and how their meanings are negotiated at the level of vernacular theories, represented by the faithful, belonging to the community of the Roman Catholic Church in Poland and participating in discussions on Catholic internet forums’ (p. 5) and the results of the study are particularly illuminating in this developing field of investigating human perceptions of AI as an ‘algorithm’ and an algorithmic presence in daily life. Sierocki found through a qualitative, netnography approach to the research that ‘the category of “algorithm” is used in discourse as a metaphor’ (p. 8) and that there was a widespread view of how algorithms function for the online-forum users. Certain significant issues regarding human conceptions of algorithms and religion are reported in the research findings (pp. 8–15) including discussions of the important difference between the reality of religious practice and the artificial nature of religious online practices such as ‘replacing spiritual guides’ (p. 12) and the role of ‘artificial prayers’ (pp. 12–13). This paper is an important contribution to the debate about AI and religion as is encouraged in the Dicastery for the Doctrine of the Faith and Dicastery for Culture and Education (2025) note by focusing on researching concepts of ‘human dignity’ and AI, as well as urging further research into ‘understanding of algorithms in the context of religion’ (p. 17).

A comparative analysis of AI use in two theological contexts is undertaken by Dimara et al. (2024): the West and East contexts, ultimately attempting to bring light to the question of whether AI is a threat to theology. By looking at examples of ‘a robotic Buddhist preacher to a blessing-delivering robot in a German church’ (p. 1), the authors explore this question from multiple theological perspectives. Ultimately their conclusions are positive, and they come to the conclusion that ‘history suggests that the dynamic interplay between theological doctrines and technological advancements may lead to an enriching dialogue rather than a predetermined conflict’ (p. 8).

In Onyeukaziri’s (2024) paper the author argues for research and development in the use of AI that should be ‘anthropological’ that is, human centred. In a detailed series of arguments, the author clearly states the need for a central role in AI research and discourse for all members of the Catholic community. Looking at both ‘weak’ and ‘strong’ AI and the implications for both, the author sees clearly the need for:

a serious and detailed interrogation of the research in AI as it relates to the meaning of work and the dignity of the human person in the social teachings of the Church. It emphasizes the need for Christian scholars across all fields to engage in serious study and research of AI, paying attention to the promotion of a personalistic and anthropological/ethical-based AI design and development.

(p. 10)

This is an important contribution to raising the need for a Catholic focused, humane and ethical approach to the challenges, and potential, of AI as society moves forward with integrating this technology across cultures and society. Such an approach fits with the guidance offered by the Vatican in pursuing a fundamentally humane approach to integrating AI into our daily lives and practices.

In a fascinating research study, Gruchoła et al. (2024) acknowledge that their work fills an existing gap as ‘(C)urrently, there are no empirical studies directly addressing peoples’ attitudes toward artificial intelligence as a technology supporting prayer practices in the process of developing Christian prayer’ (pp. 1–2). Their research looked at Polish Catholic respondents’ attitudes to AI being used to support prayer practices, and the findings present some fascinating results. Firstly that ‘the higher the level of religious practices (frequency and engagement), the more conservative the attitude towards AI programs/applications supporting prayer.’ And, at a wider population level:

the groups that were most favorable to AI-based prayer applications were respondents/Poles aged 35–65, professionally active individuals, residents of large cities with access to AI technology, and those who frequently used it in their professional work. Notably, there was an increasing interest in prayer supporting applications among older individuals (55–65 years), where the observed level of users was higher than that of younger respondents.

(p. 11)

The identification of such research gaps, leading to a growing collection of data regarding AI use and prayer practice, is an important area for future research work.

Finally, Kozak and Fel’s (2024) study aimed ‘to analyze the correlations between the level of religiosity and emotional responses to artificial intelligence’ by using Welch’s variance analysis tool, ANOVA, to compare university students’ emotional reactions to AI across religiosity levels’ (p. 1). The results indicated that there were significant differences in the emotional responses to AI across the groups with different religious attitudes, ‘particularly in the context of negative emotions such as fear, sadness, and anger’ (p. 11).

What is clear from this literature review is that academic, scientific investigations into how AI’s relationship with religion has vital contributions to make across all areas of society including governance, ethics, economics and education. All the reviewed literature texts meet the criteria of being scientifically valid in terms of Popper’s urging for ‘problem solving’ as well as being concerned with the need for human dignity to be at the centre for research into AI as urged by Pope Francis. Additionally, in many of the background reviews of AI and religion, the authors have looked at the societal impact that AI is having and will continue to have, often referencing previous dystopian or utopian visions by writers from the past. However, what does seem plausible to suggest is that the entire role of religion in the academic discourse needs to be re-centred. Donald (1999) made a prescient observation that:

…secular intelligentsia have become more and more alienated (from high culture) with every generation unable to accept the new order, and uncomfortable with their own deeply religious past. Yeats was right. The center did not hold, there is still no point, and there is no dance. We are in a cultural free-fall, with no ground in sight.

(p. 22)

3. Discussion

• “Turning and turning in the widening gyre
• The falcon cannot hear the falconer;
• Things fall apart; the centre cannot hold…”2

AI is becoming increasingly immersed in all aspects of our lives and, unsurprisingly, is infiltrating religion. From prayer practice to robot ministers to uncovering archeological sites of religious significance, AI has the potential to change humanity’s understanding of and relationship with religion. The 15 academic publications submitted for more detailed scrutiny in this review all met a fundamental, Popperian, falsifiable criteria of being empirically valid. Many of the papers also acknowledge, either in the contextual sections of the work or as a main focus of the work, the importance of the need for a continued ‘governing hierarchy’ (Donald 1999, p. 23) for society as we see the rise in AI use across all aspects of our lives. Of the publications reviewed, seven provide a particular Catholic Christian element or focus (Yao 2024; Singler 2024; Sierocki 2024; Dimara et al. 2024; Onyeukaziri 2024; Gruchoła et al. 2024; Kozak and Fel 2024). Most significantly, the publications reviewed all meet the societal challenges and questions posed in the Antiqua Et Nova note Sections 50–57.

What both Donald and the urgings of the Vatican’s Antiqua note address is the fact that religion, and in this specific regard, Catholicism, provides a ‘governing hierarchy that regulates both individual consciousness and public behavior on much of the planet’ (Donald 1999, p. 23). Donald sees the rapidly fading governance structure of religion for the effective cohesion of society as a grave concern, and in this matter, he is completely correct. What is needed, therefore, is a continued search for the answers to the questions that this new technology is now posing for society as a whole. A search that is rigorous and scientifically grounded, such as that urged by Popper and his correct focus on ‘critical rationalism’ but also grounded in a recognition of the vital importance of religion to humanity. This search is reflected in the review of the literature above, scholarly works that take seriously the need to understand the multi-faceted perspectives that AI is having on all aspects of our lives.

Encouragingly, there are tentative signs of a possible renewal of faith in parts of the globe that have been suffering from declining religious attendance for generations. In the US, the share of Americans who identify as Christians, according to a recent Pew study (Smith et al. 2025), ‘shows signs of leveling off’ (p. 1). In the UK, according to a population of survey of church attendance in 2018 and again in 2024 (Hattie 2025) there are signs of increased church attendance in the younger generations: ‘In 2018, four per cent of the 18- to 24-year-olds reported that they attended church monthly, compared with 16 per cent in 2024’ (p. 1). These are hopeful signs of what may be a turn towards the practices and community of Christian religious observance and a realization of a fundamental human need for a spiritual and communal connection.

4. Conclusions

By shining a light on some elements of the hugely important scientific work that is being currently practised into the role of AI and Religion, as demonstrated by all of the publications addressed in this review, in the words provided to us by John, we may find a way to re-balance our human autonomy in ways that see how technology can benefit and serve humanity’s needs of the future, rather than allow it to dominate humanity, because what really is at the heart of the Vatican Antiqua et Nova note is a recognition of the need for ‘True Wisdom’ (p. 18) a wisdom that ‘is the gift that humanity needs most to address the profound questions and ethical challenges posed by AI.’ In scripture we find this calling for human wisdom as it “lets itself be found by those who seek it and be seen by those who love it; it anticipates those who desire it, and it goes in search of those who are worthy of it” (Wis 6: 12–16). We hope that our review has made a contribution in the appreciation that such words of wisdom may help to guide all who are working in this brave new world.