2. Christianity and The Challenge of Artificial Intelligence

AI, medicine and Christian ethics - Research Handbook on Health, AI and the Law - NCBI Bookshelf An official website of the United States government Here's how you know The .gov means it's official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site. The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely. Log in Show account info Close Account Logged in as: username Dashboard Publications Account settings Log out Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation Bookshelf Search database Books All Databases Assembly Biocollections BioProject BioSample Books ClinVar Conserved Domains dbVar Gene Genome GEO DataSets GEO Profiles GTR Identical Protein Groups MedGen MeSH NLM Catalog Nucleotide OMIM PMC Protein Protein Clusters Protein Family Models PubChem BioAssay PubChem Compound PubChem Substance PubMed SNP SRA Structure Taxonomy ToolKit ToolKitAll ToolKitBookgh Search term Search Browse Titles Advanced Help Disclaimer NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health. Solaiman B, Cohen IG, editors. Research Handbook on Health, AI and the Law. Cheltenham, UK: Edward Elgar Publishing Ltd; 2024 Jul 16. doi: 10.4337/9781802205657.ch13 Research Handbook on Health, AI and the Law. Show details Solaiman B, Cohen IG, editors. Cheltenham, UK: Edward Elgar Publishing Ltd ; 2024 Jul 16. Contents Search term < Prev Next > Chapter 13 AI, medicine and Christian ethics Zachary R. Calo . This chapter examines the uses of artificial intelligence (AI) in healthcare from the perspective of Christian ethics. It gives particular attention to the resources of theological anthropology. At the heart of Christian theological anthropology as applied to matters of medical ethics is the concept of dignity, and thus this chapter can be said to approach the question of AI’s use in health from a personalist and dignitarian perspective. Christian understanding of the person provides a point of convergence for establishing common moral views about the uses of AI within the medical field, but equally is the source of a distinctive Christian critique of certain technological futures. There remain aspects of medical care that are irreducibly human and in which uses of AI encounter limits. This chapter proposes an ethic organised around such ideas as solidarity, suffering and dependency. 1. Introduction Artificial intelligence (AI) is revolutionising the provision of healthcare. 1 It offers immense potential for improving medical diagnoses, treatment and patient care. Precision medicine provides a means of developing personalised treatment plans while predictive analytics can use AI to assess future health risks. AI algorithms can analyse medical images to detect patterns or abnormalities that might be easily overlooked by the human eye. AI also has applications that go beyond patient care. AI-powered virtual assistants, telemedicine and wearable technologies are changing the delivery of health services. AI can aid medical administration and record-keeping in ways that improve efficiency and reduce errors. Of course, some of the most transformative work is occurring in medical research, where AI algorithms can analyse vast amounts of data and scientific literature. 2 Many of these uses of AI within healthcare do not in principle seem ethically problematic. 3 What is objectionable, after all, about making better diagnoses or improving treatment plans? What is problematic about facilitating research and devising more effective drugs? All of these goals accord with the most basic purposes of medicine and do not seem to establish deep points of ethical tension. As is often the case, however, the rapid advance of technology has outstripped sustained ethical reflection. 4 Amid a flurry of new possibilities, it is important to create space for ethical reflection on where technology has come and where it should go. It is also critical to step back from particular innovations in order to reflect on the broader moral implications of AI for the nature of medicine itself. This chapter examines what the Christian ethical tradition can offer emerging debates about the application of AI to healthcare. The perspectives offered are not in all instances uniquely Christian. There are many points of convergence and overlap between Christian perspectives and those of other religious and secular traditions of ethical thought. At the same time, the chapter explores what is and might be distinctive about a Christian approach to using AI in healthcare. In undertaking this task, the chapter focuses on applying the resources of Christian theological anthropology, which is the body of thought that seeks to understand the nature of human personhood in relation to God. At the heart of Christian theological anthropology is the concept of human dignity. Dignity is a concept that is both particular and universal. It is invoked as the foundation for universal human rights that transcend different worldviews. It operates as the dominant moral category within both secular and religious bioethics. It also serves as a theological category that reflects particular views of human nature as illuminated by religious insight. In this chapter, dignity is used in this latter respect, as a particularistic account of human nature, human goods and human ends as illuminated by Christian theology. With theological anthropology as its interpretive lens, the chapter proceeds by first looking at general Christian engagement with AI. It then turns to considering theological perspectives on AI specifically within the context of healthcare and medicine. Finally, it addresses how these perspectives might be applied to the development of legal and policy guidelines. The chapter argues that while there is much about AI’s increased use in healthcare that should be embraced from the perspective of Christian ethics, there are also limit principles that arise. Even more so, from the perspective of Christian theological anthropology, there might be appropriate hesitation about certain uses of AI in healthcare and, even more so, the overall effect that AI has in reshaping the moral nature of medicine. 2. Christianity and The Challenge of Artificial Intelligence Before turning to how Christian ethics might address the uses of AI within healthcare, it is important to consider Christian thought about AI more generally. Much of what Christian ethical thought will have to say about AI and health is derived from these more basic theological assessments of AI. Moreover, while there is little theological literature focused on AI and health, there is a significant and rapidly developing body of work on theology, ethics and AI. 5 This literature offers important resources for sketching the development of a future engagement with health and medicine. It is commonly asserted that Christian thought, and perhaps much religious thought for that matter, is instinctively resistant to AI. So the argument goes, Christianity holds firm to a view about human nature as having been created imago Dei , in the image of God. This nature is fixed and inviolable, unique and special. Efforts to alter and augment it, or mimic and recreate it through artificial means, violate the intrinsic dignity that is the sole province of humanity. It is an arrogation of the Divine prerogative, a supreme act of hubris in which creatures seek to become Creator. As one commentator proposes, there is a basic fault line between ‘those who favor increased human understanding and control of nature, and others who look at such projects as impious, hubristic, or impermissibly defiant of the given order of things’. 6 Most religious thought, it is assumed, will lie on the impious, hubristic and impermissibly defiant side of the divide. This representation of Christian thought about AI, however caricatured and simplistic, is not without some warrant. As theologian Ronald Cole-Turner concedes, advocates for the radical transformation of humanity expect ‘traditional religious institutions to [. . .] stand in the way of science and technology [. . .] and to defend the human status quo as fixed and final’. 7 It is also the case that theological perspectives can inspire intense critiques of AI, especially in its more speculative and fantastical ambitions, whether that be creating conscious machines or uploading minds. The ambition to exercise mastery over human nature and destiny is critiqued as a seizure of the divine prerogative or, in the words of one theologian, a ‘sort of sham human auto-divinisation’. 8 Another concern is that AI as a science is indissociable from AI as a philosophical worldview that stands in necessary tension with the Christian worldview. Along these lines, Robert Song argues that ‘the ambitions of the scientific disciplines related to AI’ rest upon a ‘reductive, naturalistic mentality’. 9 Naturalism, in Song’s understanding, is a reductive philosophy that claims everything can be explained through the natural sciences. A naturalistic account of humanity cannot be reconciled with the Christian idea that ‘every single person is a precious soul loved by God’. 10 There is no doubt that AI threatens to destabilise many inherited aspects of Christian thought, especially claims about human nature, human uniqueness and human communion with God. What, after all, becomes of the central Christian claim that human beings were created for a special relationship with God if machines can replicate the noetic capabilities of persons? What is the meaning of imago Dei if AI and robots display the same capacities as human beings? 11 Just as evolutionary science led to changes in Christian understandings about the nature of human beings in relation to God and creation, so too might AI compel a rethinking of human nature, human uniqueness and what it means for humans to commune with God. 12 While there are criticisms and concerns, the perception that Christianity is resistant to the development of AI is misguided. Christian thought is doing much more than offering a defiant no to AI. For some Christian thinkers, the starting point is the principle that AI in itself is amoral. 13 The moral assessment of AI, as with any technology, does not concern the thing in itself but its particular uses. It might be determined that it is ethical to use AI in the delivery of certain health services but unethical to use it in forms of automated warfare. The ethical task for Christian thought is not to offer a yes or no to AI tout court but rather to assess how AI contributes to the common good in a certain context. 14 From a Christian perspective, the main ethical criteria is whether a given use of AI respects human dignity and facilitates authentic human development. 15 The evaluation of AI, in other words, revolves around considerations of theological anthropology. Does AI elevate or diminish that which is essentially human? The question of whether AI elevates our essential humanity is different from that of whether ‘AI will make us more or less human’. 16 In ‘Artificial Intelligence: An Evangelical Statement of Principles’, a group of Christian scholars stressed that AI will not ‘obtain a coequal level of worth, dignity, or value’ to human persons. In making this assertion, the authors seem to mean that there is a fixity to human nature that will always remain unique and unchangeable. AI, in this sense, can neither acquire nor take away the dignity bestowed on human beings by God. 17 Human persons will always be unique, even as this essential humanity is made more or less dignified in the way it is lived out and expressed. These anthropological considerations, which operate at a high level of abstraction in much of the theological deliberation about AI, assume a more determined meaning when applied to AI in medicine and healthcare. Three anthropological concepts that have resonance within Christian ethics are particularly relevant: embodiment, relationality and vulnerability. Embodiment : For Christian thought, personhood is indissociable from embodiment. To be human is not merely to have thoughts, emotions and memories: a person is not simply a mind, such that if memories could be captured and uploaded onto a supercomputer the self could endure eternally; rather, what makes us human is that we are enfleshed and embodied creatures. The body is particularly central to Christian ethics in light of the doctrines of the incarnation and resurrection. Human destiny is not found in overcoming the body but in the body remade and glorified. We are bodies and the embodied character of persons remains essential to reflecting on the relationship between theology, anthropology and AI. According to Christian theological anthropology, the body possesses an intrinsic dignity that needs to be respected. The focus on embodiment within Christian ethics also leads to an understanding of the body’s limits. It is not just respect for the body as such that matters, but respect for the particular nature humans have as embodied creatures. To be human is to be limited by the givenness of the body. To interfere with or to change this nature generates particular concern, such as with proposals to use AI to augment human capabilities. 18 Relationality : Relationality refers to the capacity to be in a relationship – that is, to encounter others, the natural world and God. But pushed further, relationality also concerns the ways in which persons express and experience freedom with and through others. Not only is the capacity to love and be loved essential to human experience, but the relationality to which it gives rise is one of the riskiest aspects of being human. Ronald Cole-Turner notes that Christians are ‘called [. . .] to lose [their] lives in relationship with God’. 19 It is by losing the self in the otherness of God that one becomes an authentic self. Ted Peters thus concludes that ‘Christian anthropology has an investment in the viability of the person-in-relationship model of the human reality’. 20 The self, it might be said, is achieved through relationship – through being in-communion. To have and to exhibit dignity necessarily then includes the capacity to be in a relationship. This focus on relationality bears especially on the doctor–patient relationship. A central concern for Christian ethics going forward thus needs to be assessing how uses of AI can either enhance or disrupt the doctor–patient relationship. Vulnerability : The limits that define our nature, including the ultimate limitation of death, are essential to our humanness. To respect human dignity is to respect these limits. They are not to be overcome but taken as a site of moral reflection about what it means to live well as limited creatures of a certain sort. The features of human nature that might seem at first to be most undignified – deficiency, debilitation, death – are, in fact, essential to grasping how to live with dignity. As Christian ethics further reflects on the uses of AI in health, it will encounter a tension between embracing the goods that arise from employing new technologies and the impulse to call attention to the limits of all medicine. These are broad anthropological themes, but they prompt further consideration about what a Christian engagement with AI might entail. The ethical themes of embodiment, relationality and vulnerability are also particularly relevant points of engagement with the uses of AI and healthcare, to which the next section turns. 3. Towards a Personalist Ethic of Care: Christianity, Ai, and The Priority of The Person Little has been written specifically about Christian ethics, AI and health. It is a field just beginning to take shape. As such, the remarks that follow are necessarily of a somewhat speculative nature. Nevertheless, there are significant resources in place to consider how Christian ethics might operate in this field going forward. In addition to the rapidly developing conversation around AI, there is a long tradition of Christian thought about healthcare, bioethics and the uses of technology within medicine that bear on the conversation. While developing an ethical discourse around AI and health will involve more than manufacturing an alloy from these existent lines of inquiry, significant resources are in place for identifying the shape of a Christian ethical voice. The central concern for Christian ethics is not with AI as such but rather with how AI impacts the human person who is the object of medicine. ‘Technology, viewed in itself, is ambivalent’, Pope Benedict XVI wrote. 21 The Pope’s point is that technological capability has no intrinsic moral status. What matters is the ways in which technology shapes moral imagination and practice. When it comes to the uses of new technologies, such as AI in healthcare, the operative question is not whether AI should be used but rather whether any given use enhances or diminishes the dignity proper to human beings. To put the question more pointedly: when does the use of AI lead to the evasion or even destruction of that which should remain an irreducibly human element of medicine? It is from this anthropological perspective that Christian ethics will engage both systemic and particular applications of AI to healthcare. A Christian ethical appraisal of AI in healthcare is one in which technology, and its ability to improve the efficiency, accuracy and cost of care, remains subservient to the human person as primary object of concern. This model of Christian engagement with AI might be termed a personalist ethic. At its most basic, a personalistic ethic is one that measures technology from the perspective of the human person. Technology is to be the servant of the human and not its master; it is to facilitate the realisation of human flourishing and not frustrate it. In this way, a personalist ethic is neither pro-AI nor anti-AI but rather seeks to discern how to use AI based on whether it enhances human dignity and the priority of the person as patient. Using AI for medical imaging, the development of personalised treatment plans or to support surgery might be deemed straightforwardly dignity-enhancing. By contrast, using AI for genomic analysis could raise ethical concerns about genetic discrimination, while AI-driven telehealth assessments might raise concerns about the depersonalisation of care. These evaluations are ultimately fact-intensive and not often easy or clear-cut. They will only become more complex. What Christianity offers, however, is an approach to AI and medical ethics that begins with anthropology and a theological view of what it is to be authentically human. Pope Francis offers one example of how to address AI on such personalist terms. He has identified in numerous writings the threat of what he terms the ‘technocratic paradigm’. 22 This paradigm is not only defined by the ever greater use of and reliance on technology but, even more critically, by an embrace of technology as a self-subsistent form of moral order. Scientific progress, the Pope warns, has become an end in itself – a goal that goes unquestioned. Progress is pursued for the sake of progress, mastery for the sake of mastery. Even improvement of the human condition – a worthwhile aim, of course – is sought without question. The ceaseless will to mastery, however, cuts technology off from contact with a deeper moral anthropology. Rather than elevating the human, it becomes paradoxically, in Pope Benedict XVI’s assessment, ‘radically anti-human’. 23 This ‘purely functional rationality’ leads to the ‘[mutilation of] man’. 24 This is not a problem that begins with AI. It is an impulse deeply rooted in modernity itself. But it is an impulse being rapidly furthered and, in some sense, fulfilled through the reality and promise of AI. The project of strong AI and the fantastical projects of uploaded brains and digital immortality represent a quest to achieve total mastery over the ultimate limits of humanity. This line of moral critique is hardly exclusive to Catholic thought. It provides a framework for a general Christian ethic of AI. But what does a personalist approach to AI and medicine entail? What does it mean for technology to be ‘humanized’ rather than humanity ‘technologized’? 25 What is required by Pope Francis’s charge at the 2018 World Economic Forum in Davos to ensure that ‘Artificial intelligence, robotics, and other technological innovations [. . .] contribute to the service of humanity’? 26 Above all, how can this personalist ethic be applied to such areas as health and medicine? A few overarching considerations might give more concrete shape to Christian ethic of AI in health. Most basically, AI must not, in Paul Scherz’s phrasing, ‘disrupt the healing relationship and its possibilities for encounter’. 27 This position can be understood on different levels. Most basically, the concern is that AI be employed to enhance the human element of care in which the practitioner ‘[focuses] on the patient’. 28 It emphasises the need to assess medical care by more than efficiency and outcome, but equally in terms of the relationship forged between the doctor and patient. This critique taps into longstanding concerns about the ways in which technology has decentred the face-to-face aspects of care. Of course, the integration of AI into the provision of health services need not operate in the same dehumanising way as other technologies. Many uses of AI, in fact, have the potential to humanise the provision of health services. AI, for instance, could take over certain routine and bureaucratic tasks, thus freeing professionals to devote more time and attention to patients. 29 This might include using AI to write appeals letters for decisions that deny coverage for a procedure or respond immediately to patient inquiries or requests. Outside of such bureaucratic tasks, however, there are legitimate concerns about integration of AI into medical diagnosis and decision-making that might, for instance, become more transactional. There are equally concerns about how to balance efficiency and empathy. 30 The challenge for Christian ethics will be to assess how AI impacts the healing relationship and can be employed to aid rather than inhibit the human element of caring. This concern with preserving the human element of care goes beyond the doctor–patient relationship in a narrow sense. What matters is not merely the amount and quality of attention that a doctor can offer a patient. The ethic is at base grounded in a deeper sense of the ontology of care in which the doctor ‘encounters’ the patient as a genuine object of concern in her or his particularity. This ‘human-to-human encounter’ cannot be replaced by AI. 31 At the same time, it need not necessarily be obstructed by it either. AI might be integrated holistically into an ethic of care, such that AI-informed recommendations about a plan of treatment might be considered with the same level of humanity as those made without AI. For Christian ethics, the key is that the realities of our human condition, in all its frailties and vulnerabilities, must be met at a human level. The doctor and the patient, both mortal beings who shall die, transcend in their humanity the power of any technology. Yet technology is not intrinsically anti-human. To heal the body and to extend life is praiseworthy, so long as it is done through a genuine encounter that respects the needs and limits of our created nature. This ethic of care is grounded in solidarity, suffering and participation in the healing work of God. In this respect, a Christian personalist ethic locates the work of health and medicine within a larger moral ecology that sees the doctor as offering a form of concern that reflects the love of God for persons. This moral vision is not exclusively Christian, but its language and form are deeply nurtured by the particular resources of Christian ethics and anthropology. To reframe it in non-theological terms would entail loss, for the notion of encounter is informed by the ways in which the work of medicine mirrors the work of God in the world. Christian bioethicist William F. May’s use of the term ‘medical covenant’, a bioethical concept that hearkens to biblical themes, conveys a similar idea. 32 A medical covenant goes beyond a mere doctor–patient relationship, in that the doctor has an obligation to remain committed to the care, flourishing and dignity of the person. The language of covenant connects the work of the physician to God’s fidelity and provides a way to speak of an encounter grounded in loving concern. Christian thinkers have also raised concerns about the depersonalising effect of AI-driven data collection on healthcare. Paul Scherz, for instance, has considered how ‘data intensive technologies like AI [. . .] may form practitioners in the technocratic paradigm so that they see patients in terms of cost containment’. 33 Patients, that is, might become mere algorithmic data points in a push to find more efficiencies within the healthcare system. This is a double-edged sword. Just as AI can provide new efficiencies in the delivery of care that ought to be embraced, the same processes might encourage, even if only subtly, a view of healthcare as a transaction that seeks to maximise savings in the delivery of a product to consumers. The patient, Scherz argues, is thus reduced ‘to what can be captured by quantitative metrics that can be analyzed through machine learning’. 34 This concern might prove particularly problematic from an ethical perspective when such metrics are employed to make decisions about the allocation of resources, provision of care or end-of-life decisions. In other contexts, concern with the depersonalising effects of AI requires a different approach. Are the model for end-stage liver disease (MELD) scores used in determining eligibility for liver transplants more ‘depersonalised’ if derived from linear statistical models as opposed to machine learning? 35 It would seem not, and yet the lack of explainability behind the AI-generated result could create a sense among patients that the result demands explanation and warrant. The challenge is to integrate personhood and technology in ways that humanise machine-based determinations about treatment. Similarly, the task for Christian ethics is to preserve and elevate the normative centrality of the human person even as new technologies are increasingly utilised in the delivery of care. It might be proposed that AI is merely providing information on a neutral basis that can inform decision-making by doctors and hospitals. The data as such contains no ethical judgement as to its own meaning. But to many Christian thinkers, there is a concern that AI necessarily informs a vision of reality. Although the technology might be neutral, its application is not. AI can embody a view of reality and encourage a view of the human person, and the person as patient, that is already normed. AI did not introduce this moral anthropology into modern medicine, but it has certainly accelerated it. This is not to intimate that AI cannot be utilised to advance many goods, but we should be ever more attentive to the ways in which technologies subtly encourage and transmit normative worldviews. For instance, Joseph Davis has argued that the increased use of AI in psychiatry illuminates the failure of much contemporary medicine to meet ‘the unique and authentic human dimension’ of treatment. Referring to mental health chatbots, for instance, Davis remarks that ‘Perhaps the chatbots can be a wake-up call, alerting us by their very artifice and superficiality, to what human care truly is. And to what it should always aim to be.’ 36 Introducing AI into such fields as psychiatry changes the moral nature of care and introduces distortions into medicine. Paradoxically, increased reliance on AI-assisted care might also provoke renewed reflection about what authentic patient-centred care entails and how it can be maintained alongside AI. These are not only Christian concerns but human ones. But these considerations are a place from which Christian ethics, with its attention to the personalist dimension of care, can find a critical voice in discussions about AI. Pope Francis has spoken of the need to develop an ‘algor-ethics’ ordered around concern for human dignity. Giving shape and life to such an ethic in any particular context is the ongoing work of constructive moral thought. 37 The particular challenge for Christian ethics in the area of AI and healthcare is to translate broad principles of theological anthropology into more concrete judgements. The next section considers how such work has and might be pursued. 4. From Personhood to Policy: Christian Ethics and Guidelines for The Use of AI in Healthcare In recent years, several leading organisations have issued guidelines on the ethical use of AI in healthcare and medicine. Christian organisations have done the same. These are nascent attempts to apply ethical principles to emerging challenges in healthcare in a manner that will inform and shape practical decisions about law, policy and practice. In this section, these guidelines are used to assess the current state of ethical discourse around AI and health, while looking especially at how Christian perspectives are positioning themselves in relation to mainstream institutional guidance. What is notable about these initiatives is the high level of synchronicity, albeit at a broad level, that has emerged. In addition, the ethical norms advanced to address AI generally mirror, with only minor adjustments, the same guiding principles that have informed more general discussions of medical ethics. While there is of course some variation at the margins, the ethical guidelines issued by leading medical organisations tend to emphasise principles of privacy, confidentiality, consent, autonomy and bias in the use of AI. 38 This same approach is found in all of the major guidelines and statements issued to date, including those of the World Health Organization, 39 the American Medical Association, 40 the Parliamentary Assembly of the Council of Europe, 41 the Royal Australian and New Zealand College of Radiologists, 42 the Indian Council for Medical Research 43 and the Nuffield Council on Bioethics. 44 The ethical framework set forth in these guidelines follows that which is being advanced in the scholarly literature on AI and health. 45 In fact, a recent scoping study found that the most frequently cited ethical principles in the academic literature are autonomy, privacy, fairness, explicability and prevention of harm. 46 Another review of the literature found that the main ethical issues discussed are bias, control, autonomy, consent and privacy. 47 While consensus is itself not surprising, the degree to which Christian statements have employed similar principles and language is. Indeed, to the extent that Christian ethics have been translated into policy guidelines, they have tended to follow the same approach of those found in the broader secular conversation. For instance, the Rome Call for AI Ethics, a document signed by the Pontifical Academy for Life, endorsed as its guiding principles transparency, inclusion, responsibility, impartiality, reliability and security. 48 There is little about these concepts that adds anything distinctive to the debate; nor, for that matter, is there anything about the statement reflective of a distinctive theological methodology. Similar points can be made about ‘Artificial Intelligence: An Evangelical Statement of Principles’. On the whole, this document offers a richer and more systematic theological engagement with AI than the Rome Call. Yet Article 4 of the Evangelical Statement of Principles, which addresses the uses of AI in medicine, provides only that ‘AI-related advances in medical technologies’ should be ‘guided by basic principles of medical ethics, including beneficence, non-maleficence, autonomy, and justice, which are all consistent with the biblical principle of loving our neighbor’. 49 Although reference is made to biblical principles, and the Statement presents itself as an initiative to ‘equip the church to proactively engage the field of AI’, it ultimately does little to illuminate what Christian ethical perspectives uniquely offer the debate. It is more of a theological gloss on mainstream medical discourse than a distinctive theological intervention. The existence of significant overlap between Christian and secular approaches reflects a common awareness of the ethical challenges of using AI in health. Yet, the methodological and linguistic commonalities found in these policy documents obscure the potential distinctiveness of a Christian ethical perspective. These issuing organisations have missed an opportunity to inject into the debate a perspective that diverges from mainstream positions and challenges certain governing moral assumptions at a fundamental level. This divergence between Christian and secular approaches might occur on two levels. For one, theological insight might alter or qualify the dominant meaning of certain concepts. For instance, bias or equality of access might be matters of common concern and yet take on a different emphasis within a theological framework that emphasises the moral doctrine of the preferential option for the poor. 50 Likewise, Christian ethics might offer a different understanding of the concept of autonomy and its limits. For instance, Christian ethicist Gilbert Meilaender observes that ‘Although a great deal of the best work in bioethics has involved the application of certain ethical principles – such as respect for autonomy [. . .] to particular issues of concern, there is no way to apply principles in a vacuum’. 51 Autonomy is not neutral, nor does it have a universally accessible meaning. Its application depends on background assumptions and worldviews. A notion of autonomy that affirms the right of patients (and not AI) to have final decision-making authority over treatment might be largely uncontroversial, yet autonomy as a deeper anthropological value cannot be unqualifiedly embraced so easily. An endorsement of autonomy as a limit principle against the encroachment of AI into medical decision-making is one thing. But autonomy as a normative value that elevates self-mastery might prove deeply problematic from a perspective that emphasises the limits of human nature and dependency on God. Freedom and will are not ultimate goods, for the self does not exist in isolation. Within the tradition of Christian ethics, the person is never fully autonomous but exists within a web of relations with other persons and God. Self-mastery in a totalising sense is not only morally problematic but a delusion that ignores one’s ultimate reliance upon the grace of God. On a more basic level, divergence might also be found in the general posture taken towards future uses of AI in healthcare. Christian ethical thought has been broadly supportive of using AI to improve the efficiency and accuracy of diagnosis, the effectiveness of treatment and the burdens of management. There are no doubt many other uses that would be roundly endorsed. Yet, Christian ethics has expressed hesitation when AI takes the form of a more totalising worldview. Technology is not an end in itself, nor can technological progress offer an ultimate hope for the future. What Christian ethics can, thus, above all offer debate is a sense of the limits that should bind us, even when technology permits going further. Technological capability cannot override the primacy of human persons as possessors of a unique dignity and moral status. Privacy, confidentiality, consent, autonomy and bias are all important ethical considerations. But what animates Christian ethical engagement with AI is the question of what it means to be human. In healthcare more than any other field, the ethical implications of technology need to flow from this fundamental consideration. When AI is treated as an ethic unto itself, a good that requires no warrant, it becomes a form of ‘idolatry’ that must be critiqued and deconstructed. 52 Christian ethics can critique the idolisation of AI in healthcare by offering a perspective that calls attention to that which is irreducibly human and beyond the grasp of technology. It can also uniquely contribute to the emerging moral discourse in this area by recalling principles of care based not on power and possibility but on weakness. This different sort of ethic might especially call attention to the ongoing realities of suffering and death, solidarity and dependency in healthcare ethics. While seemingly far removed from immediate concerns with the ethical implementation of AI in medicine, it is precisely such considerations that are needed to shape a deeper moral understanding of when, how and why new technologies might be employed. These are broad moral principles that do translate easily into policy proposals, but they also have the potential to pull debate into a broader moral framework that challenges default understandings of the terms that shape the current debate. Suffering : The provision of medical care to those suffering is the most human of actions. It forces a confrontation with human vulnerability and the very limits of human nature itself. A medical practitioner is required to do no harm but is not required to extend life at all costs. As Meilaender observes, ‘At the heart of Christian belief lies a suffering, crucified God’. While suffering is not a good thing, we should equally not ‘suppose that suffering must be eliminated by any means that is available to us’. 53 Human dignity as it relates to healing informs two competing principles. It compels the work of eliminating suffering while affirming the need to be in solidarity with those whose suffering cannot be overcome. Part of the Christian hesitation towards AI is that it might elevate the elimination of suffering above other moral considerations. This might be seen when considering the uses of AI in connection with palliative care. What moral worldview is embedded, implicitly or explicitly, into an AI system and how might this impact provision of care? Schwarting, for instance, has considered how a Christian perspective on end-of-life care can resist the impulse to preserve or extend life at all costs, whereas AI trained with a bias towards ‘vitalism’ might not. 54 Of even greater concern is that AI could have a bias against preserving life. Could mortality predictions lead to the early cessation of treatment on the grounds that a life has diminished value or utility? Could they be used in making decisions about resuscitation or to facilitate use of euthanasia? AI has the potential to disrupt ethical assessments about suffering, care and what constitutes a meaningful life (or a life worth committing resources to sustaining). A Christian understanding can stand in critical judgement of the idea that eliminating suffering is the highest goal of medicine. Christian ethics might also critique uses of AI that allocate care based on judgements about efficiency, resource allocations or quality of life. Solidarity : AI presents a particular challenge to sustaining what John Wyatt calls ‘the role of human solidarity in health care’. 55 AI can augment and assist human care but can never replace it. 56 Solidarity means more than just human-to-human interaction. It encompasses an understanding of medicine as an encounter with the vulnerability and finitude that defines all persons. 57 Christian thinkers have accordingly expressed ethical concerns about employing technology, such as assistive care robots that minimise the centrality of personal encounters. 58 Paolo Benanti, along similar lines, has argued that AI can assist in the provision of medical services but cannot replace the human relationship that must remain at the centre. 59 Solidarity demands that the patient be accompanied in this experience of vulnerability. 60 While there are legitimate concerns that AI can undermine human solidarity by technologising interactions, a challenge for Christian ethics is to think about how AI can constructively enhance the practice of solidarity. For instance, Stanford researchers have trained a deep learning algorithm to predict the mortality of a given patient within the next 12 months. This information could be used to inform a more solidaristic and holistic approach to end-of-life care. It is not only valuable to the patient, but equally so to doctors and family members. Patients could more easily make decisions about spending their final days at home, doctors could have difficult conversations about palliative care that might otherwise be delayed and loved ones could make plans about how to best provide support. In these ways AI and the principle of solidarity become collaborative rather than oppositional. Dependency : Solidarity is not limited to the doctor–patient relationship but also implicates the broader community to which the person belongs. Christian ethical thought might, in this sense, emphasise the ways in which all persons are ultimately dependent creatures, neither fully self-sufficient nor autonomous. It is perhaps in ageing and illness that the illusion of the self is shattered. 61 Yet, for Christian thought, our necessary dependence on others leads to the church, the community of the faithful, whose inner life bears witness to the suffering of Christ in the world. In the mystical body of Christ, the individual and community intersect in the eucharist, the body and blood of the crucified Jesus. Against an approach to AI which sanguinely offers companion robots as a means of solving the epidemic of loneliness and social isolation, Christian ethics presents a more radical vision of a community defined by mutual dependency and weakness. 5. Conclusion Contrary to the charge of some critics, the Christian tradition is not opposed to the advancement of AI. As a tool that can enhance healing in accord with human dignity, Christianity celebrates AI. At the same time, the Christian ethical tradition does offer resources for thinking critically about how AI should be used and within what limits. This chapter has offered the experiences of solidarity, suffering, death and dependency as distinct nodes of reflection around which Christianity might provoke a conversation about the ethical aims of AI. These aspects of life, essential to our common human nature, contain dynamics that cannot be fully addressed through forms of technological intervention. They thus compel an encounter with the ineluctably human dimension of care and concern. It is here, at the limits of technology, that the uses of AI in medicine run into the anthropological vision of Christian ethics. It is through the resulting tension that Christianity might offer a genuinely distinct vision of medicine, health and healing in the age of AI. There is plenty that Christian ethics can say about particular applications of AI in the medical field and elsewhere. There will no doubt be an emerging literature in which theological ethicists opine on new and contemplated uses of AI within healthcare. These are important contributions, to be sure, but the transformative possibility of Christian ethics comes not from its intervention in discrete technical debates but from what it can say about the broader moral purpose of healthcare. Placing debate in this broader framework is critical to addressing ethical concerns often lost when operating at a more granular level. Cultivating an ethic around a theological anthropology of human limitation and vulnerability might seem of modest use to forward-looking concerns of the day, but what is perhaps most needed now is reflection on the moral disposition of medicine as it encounters technologies that upend inherited ideas of human personhood. Christian ethics is well positioned to do this.<TX> Notes 1 John Wyatt and Stephen N Williams (eds), The Robot Will See You Now: Artificial Intelligence and the Christian Faith (SPCK Publishing 2021). 2 See, generally, Peter Lett, Carey Goldberg and Isaac Kohane, The AI Revolution in Medicine: GPT-4 and Beyond (Pearson 2023); see also Varun H Buch, Irfan Ahmed and Mahiben Maruthappu, ‘Artificial Intelligence in Medicine: Current Trends and Future Possibilities’ (2018) 68 British Journal of General Practice 143; Kanadpriya Basu and others, ‘Artificial Intelligence: How Is It Changing Medical Sciences and Its Future?’ (2020) 65 Indian Journal of Dermatology 365. 3 Michael Miller, ‘Catholic Health Care and AI Ethics: Algorithms for Human Flourishing’ (2022) 89 The Lincare Quarterly 152, 153. 4 Michael J Rigby, ‘Ethical Dimensions of Using Artificial Intelligence in Health Care’ (2019) 21 AMA Journal of Ethics 121, 121. 5 An early book in this area was Anne Foerst, God in the Machine: What Robots Teach Us About Humanity and God (Dutton 2004). 6 Russell Blackford, ‘The Great Transition: Ideas and Anxieties’ in Max More and Natasha Vita-More (eds), The Transhumanist Reader (Wiley-Blackwell 2013) 423. 7 Ronald Cole-Turner, ‘Going Beyond the Human: Christians and Other Transhumanists’ (2015) Theology and Science 150, 150. 8 Brandon Gallaher, ‘Godman vs Mangodhood: An Eastern Orthodox Response to Transhumanism’ (2019) 32 Studies in Christian Ethics 200, 201. 9 Robert Song, ‘Robots, AI and Human Uniqueness: Learning What Not to Fear’ in John Wyatt and Stephen N Williams (eds), The Robot Will See You Now: Artificial Intelligence and the Christian Faith (SPCK Publishing 2021) 140. 10 ibid. 11 Simon Balle, ‘Theological Dimensions of Humanlike Robots: A Roadmap for Theological Inquiry’ (2023) 21 Theology and Science 132, 137. 12 AI, one journalist has proposed, ‘may be the greatest threat to Christian theology since Charles Darwin’s On the Origin of Species ’. Jonathan Merritt, ‘Is AI a Threat to Christianity?’ ( The Atlantic , 3 February 2017) < https://www ​.theatlantic ​.com/technology/archive ​/2017/02/artificial-intelligence-christianity/515463/ > accessed 2 May 2024. 13 As Miller notes, ‘AI, a priori, is . . . amoral’. Miller (n 3) 157. 14 ibid 153. 15 Miller (n 3) 153. 16 The Ethics & Religious Liberty Commission (ERLC), ‘Artificial Intelligence: An Evangelical Statement of Principles’ (2019) art 12. 17 The dominant Christian view would hold human nature to be fixed in its essence. By contrast, many secular advocates for employing AI to enhance humanity take this nature to be what Nichols Bostrom terms ‘a half-baked beginning that we can learn to remold in desirable ways’. Nick Bostrom, ‘Transhumanist Values’ in Frederick Adams (ed), Ethical Issues for the Twenty-First Century (Philosophy Documentation Center 2003) 4. 18 Dalia Fahmy, ‘Highly Religious Americans More Skeptical of Human Enhancements Such As Brain Implants, Gene Editing’ ( Pew Research Center , 4 May 2022) < https://www ​.pewresearch ​.org/short-reads/2022 ​/05/04/highly-religious-americans-more-skeptical-of-human-enhancements-such-as-brain-implants-gene-editing/ > accessed 2 May 2024; opposition to enhancement is contested among Christian ethicists. See, for example, Michael Buttrey, Moira McQueen and Tracy J Trothen, ‘Faster, Higher, More Moral: Human Enhancement and Christianity’ (2022) 13(4) Religions 1. 19 Cole-Turner (n 7) 160. 20 Ted Peters, ‘The Soul of Trans-humanism’ (2005) 44 Dialog: A Journal of Theology 381, 394. 21 Pope Benedict XVI, Caritas in Veritate (2009) §14. By contrast, the drafters of the document ‘Artificial Intelligence: An Evangelical Statement of Principles’ expressly ‘deny that the use of AI is morally neutral’. Although AI can be employed in ways that aid human flourishing, the statement goes on to note that ‘Since the Lord Jesus alone can atone for sin and reconcile humanity to its Creator, technology such as AI cannot fulfil humanity’s ultimate needs’. See ERLC (n 16) art 2. 22 Pope Francis, Laudato Si (2015) §101. 23 Caritas in Veritate (n 21) §14; on Benedict’s approach to technology and human development in Caritas in Veritate, see J Brian Benestad, ‘Pope Benedict XVI’s “Caritas in Veritate”’ (2009) 16(2) Josephinum Journal of Theology 1, 4. 24 Joseph Ratzinger, ‘Europe in the Crisis of Cultures’ (2005) 32 Communio International Catholic Review 345, 347 and 351. 25 ‘Address of His Holiness Pope Francis to Participants in the Plenary Assembly of the Pontifical Academy for Life’ ( The Holy See , 25 February 2019) < https://www ​.vatican.va ​/content/francesco ​/en/speeches/2019/february ​/documents/papa-francesco ​_20190225 ​_plenaria-accademia-vita.html > accessed 2 May 2024. 26 ‘Message of His Holiness Pope Francis to the Executive Chairman of the “World Economic Forum” on the Occasion of the Annual Gathering in Davos-Klosters’ ( The Holy See , 23–26 January 2018) < https://www ​.vatican.va ​/content/francesco ​/en/messages/pont-messages ​/2018/documents ​/papa-francesco_20180112 ​_messaggio-davos2018.html > accessed 2 May 2024. 27 Paul Scherz, ‘Data Ethics, AI, and Accompaniment: The Dangers of Depersonalization in Catholic Health Care’ (2022) 83 Theological Studies 271, 290. 28 ibid. 29 Eric Topol makes this claim by arguing that AI can restore the human element of medical care by allowing physicians to focus more on the doctor-patient relationship. Eric Topol, Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again (Basic Books 2019). 30 Aurelia Sauerbrei and others, ‘The Impact of Artificial Intelligence on the Person-Centered, Doctor-Patient Relationship: Some Problems and Solutions’ (2023) 23 BMC Medical Informatics and Decision Making 1. 31 John Wyatt, ‘The Impact of AI and Robotics on Health and Social Care’ in John Wyatt and Stephen N Williams (eds), The Robot Will See You Now: Artificial Intelligence and the Christian Faith (SPCK Publishing 2021) 226. 32 William F May, Testing the Medical Covenant: Active Euthanasia and Health Care Reform (Wipf & Stock Pub 2004). 33 Scherz (n 27) 272. 34 ibid 275. 35 Young-Dong Yu, ‘Artificial Intelligence for Predicting Survival Following Deceased Donor Liver Transplantation: Retrospective Multi-Center Study’ (2022) 105 International Journal of Surgery 1. 36 Joseph Davis, ‘Will ChatGPT Replace Psychiatrists?’ ( Psychology Today , 26 July 2023) < https://www ​.psychologytoday ​.com/us/blog/our-new-discontents ​/202307 ​/will-chatgpt-replace-psychiatrists > accessed 2 May 2024. 37 Miller (n 3) 158. 38 Rigby (n 4). 39 World Health Organization (WHO), Ethics and Governance of Artificial Intelligence for Health: WHO Guidance (World Health Organization 2021). The six core principles identified by WHO are: (1) protect autonomy; (2) promote human well-being, human safety, and the public interest; (3) ensure transparency, explainability, and intelligibility; (4) foster responsibility and accountability; (5) ensure inclusiveness and equity; (6) promote AI that is responsive and sustainable. 40 American Medical Association, ‘Augmented Intelligence in Health Care H-480.940’ (2018). 41 Steering Committee for Human Rights (CDDH), ‘CDDH Comments on the Parliamentary Assembly Recommendation 2185(2020) – Artificial Intelligence in Health Care: Medical, Legal and Ethical Challenges Ahead’ (2021) < https://rm ​.coe.int/recommendation-2185-2020-artificial-intelligence-in-health-care-medica/1680a2dcfa > accessed 2 May 2024. 42 Royal Australian and New Zealand College of Radiologists (RANZCR), ‘Ethical Principles for Artificial Intelligence’ (2019). The RANZCR’s statement identifies the foundational ethical principles as safety, privacy and protection of data, avoidance of bias, transparency and explainability, application of human values, decision-making on diagnosis and treatment, teamwork, responsibility for decisions made and governance. 43 Indian Council for Medical Research (ICMR), Ethical Guidelines for Application of Artificial Intelligence in Biomedical Research and Healthcare (2023). The ICMR’s statement identifies the foundational ethical principles as autonomy, safety and risk minimisation, trustworthiness, data privacy, accountability and liability, optimisation of data quality, equity and inclusiveness, collaboration and non-discrimination. 44 Nuffield Council on Bioethics, ‘Artificial Intelligence (AI) in Healthcare and Research’ (2018). Nuffield identifies the foundational ethical principles as reliability and safety, transparency and accountability, data bias, fairness and equity, effects on patients and data privacy. 45 See eg, Tugba Akinci D’Antonoli, ‘Ethical Considerations for Artificial Intelligence: An Overview of the Current Radiology Landscape’ (2020) 26 Diagnostic and Interventional Radiology 504; Dariush D Farhud and Shaghayegh Zokaei, ‘Ethical Issues of Artificial Intelligence in Medicine and Healthcare’ (2021) 50 Iran Journal of Public Health i. 46 Golnar Karimian, Elena Petelos and Silvia MAA Evers, ‘The Ethical Issues of the Application of Artificial Intelligence in Healthcare: A Systematic Scoping Review’ (2022) 2 AI and Ethics 539. 47 Fan Li, Nick Ruijs and Yuan Lu, ‘Ethics & AI: A Systematic Review on Ethical Concerns and Related Strategies for Designing with AI in healthcare’ (2023) 4 AI 28. 48 ‘Rome Call for AI Ethics’ ( RenAIssance Foundation , 2020) < https://www ​.romecall.org/the-call/ > accessed 2 May 2024. 49 ERLC (n 16) art 4. 50 The concept of the preferential option for the poor, developed most fully within Catholic social thought during the twentieth century, is the idea that particular concern should be given to the weak and disadvantaged in society. It reflects that the poor are given special regard in Scripture, such as the statement of Jesus in Matthew 25, ‘Whatever you did for the least of these, you did for me’. 51 Gilbert Meilaender, Bioethics: A Primer for Christians (4th edn, Eerdmans 2020) 1. 52 Brian Patrick Green, ‘The Catholic Church and Technological Progress: Past, Present, and Future’ (2017) 8 Religions 1, 11. 53 Meilaender (n 51) 7. 54 Marcus Schwarting, ‘A Christian Perspective on AI in Palliative Care’ ( ai and faith , 17 March 2023) < https://aiandfaith ​.org ​/a-christian-perspective-on-ai-in-palliative-care/ > accessed 2 May 2024. 55 Wyatt (n 31) 221. 56 Schwarting (n 54). 57 Andrew Sloane, Vulnerability and Care: Christian Reflections on the Philosophy of Medicine (T&T Clark 2018). 58 Wyatt (n 31) 217. 59 Interview with Paolo Benanti, ‘The Ethics of AI in Healthcare’ ( aboutDigitalHeath ) < https: ​//aboutdigitalhealth ​.com/2021/03/30 ​/the-ethics-of-ai-in-healthcare/ > accessed 2 May 2024. 60 Sherz proposes that ‘the theological analysis of AI and data ethics must ensure that such tools do not impede efforts at encounter and true accompaniment’. Scherz (n 27) 273. 61 Gilbert Meilaender, ‘I Want to Burden My Loved Ones’ ( First Things , March 2010) < https://www ​.firstthings ​.com/article/2010 ​/03/i-want-to-burden-my-loved-ones > accessed 2 May 2024. Copyright Edward Elgar Publishing. This work is licensed under the Creative Commons Attribution-NonCommercial-No Derivatives 4.0 License Bookshelf ID: NBK613212 PMID: 40245226 DOI: 10.4337/9781802205657.ch13 Contents < Prev Next > Share Views PubReader Print View Cite this Page Calo ZR. AI, medicine and Christian ethics. In: Solaiman B, Cohen IG, editors. Research Handbook on Health, AI and the Law. Cheltenham, UK: Edward Elgar Publishing Ltd; 2024 Jul 16. Chapter 13. doi: 10.4337/9781802205657.ch13 PDF version of this page (218K) In this Page Introduction Christianity and The Challenge of Artificial Intelligence Towards a Personalist Ethic of Care: Christianity, Ai, and The Priority of The Person From Personhood to Policy: Christian Ethics and Guidelines for The Use of AI in Healthcare Conclusion Similar articles in PubMed Philosophical foundations for digital ethics and AI Ethics: a dignitarian approach. [AI Ethics. 2021] Philosophical foundations for digital ethics and AI Ethics: a dignitarian approach. Hanna R, Kazim E. AI Ethics. 2021; 1(4):405-423. Epub 2021 Feb 26. Death and dignity in Catholic Christian thought. [Med Health Care Philos. 2017] Death and dignity in Catholic Christian thought. Sulmasy DP. Med Health Care Philos. 2017 Dec; 20(4):537-543. Human Dignity and Artificial Intelligence in Healthcare: A Basis for a Catholic Ethics on AI. [J Relig Health. 2024] Human Dignity and Artificial Intelligence in Healthcare: A Basis for a Catholic Ethics on AI. Gozum IEA, Flake CCD. J Relig Health. 2024 Dec 27; . Epub 2024 Dec 27. The Role of Artificial Intelligence in Nutrition Research: A Scoping Review. [Nutrients. 2024] The Role of Artificial Intelligence in Nutrition Research: A Scoping Review. Sosa-Holwerda A, Park OH, Albracht-Schulte K, Niraula S, Thompson L, Oldewage-Theron W. Nutrients. 2024 Jun 28; 16(13). Epub 2024 Jun 28. Review Xenotransplantation-theological-ethical considerations in an interdisciplinary symposium. [Xenotransplantation. 2015] Review Xenotransplantation-theological-ethical considerations in an interdisciplinary symposium. Sautermeister J, Mathieu R, Bogner V. Xenotransplantation. 2015 May-Jun; 22(3):174-82. Epub 2015 Feb 27. See reviews... See all... Recent Activity Clear Turn Off Turn On AI, medicine and Christian ethics - Research Handbook on Health, AI and the Law AI, medicine and Christian ethics - Research Handbook on Health, AI and the Law Your browsing activity is empty. Activity recording is turned off. Turn recording back on See more... Follow NCBI Twitter Facebook LinkedIn GitHub NCBI Insights Blog Connect with NLM Twitter Facebook Youtube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov