Statement for the International Congress AI and Medicine – The Challenge of Human Dignity
The tremendous development of artificial intelligence systems in recent years generates both wonder at their achievements and concern about the many problems that face them. Also a great question surges about the future we want to build through these technologies.
For an ethical reflection about AI, as with any technology, it is important not to be limited only to a consideration of the performances it enables, however spectacular; the impact it has on personal and social relationships must also be included in the assessment. As Pope Francis has said, technology:
“always represents a form of order in social relations and an arrangement of power, thus enabling certain people to perform specific actions while preventing others from performing different ones. In a more or less explicit way, this constitutive power-dimension of technology always includes the worldview of those who invented and developed it”. This is why Pope Francis reminds us that: “the inherent dignity of each human being and the fraternity that binds us together as members of the one human family must undergird the development of new technologies and serve as indisputable criteria for evaluating them before they are employed”. In new digital technologies what is at stake are not only principles and rights, but the specificity and originality of the human mind, that has to be protected in the context of devices equipped with a specific and unprecedented form of agency.
In the field of health care is therefore crucial that AI be an aid that enhances clinical judgment, supports diagnostic accuracy, and improves patient outcomes, never a substitute for the physician’s expertise, empathy, or accountability.
Key ethical principles for AI in medical practice include:
Clinical Oversight and Judgment
AI must remain subordinate to the physician’s clinical reasoning. While it can assist with pattern recognition, risk stratification, and decision support, “decisions regarding patient treatment and the weight of responsibility they entail must always remain with the human person and should never be delegated to AI”. In the process of using AI, the physician has to be careful not to be hypnotized by the fascination with technological results, leading him to share and to delegate, without enough critical mind, too much powers to a machine. Meaningful and adequate human supervision of AI means also to avoid uncritical use of techniques. AI recommendations has always to be questioned from outside!
Transparency and Interpretability
Physicians should be able to understand and explain how AI-derived recommendations are generated. Black-box algorithms that lack interpretability risk undermining trust and clinical accountability, inducing deskilling and delegation of responsibility.
Bias Awareness and Equity
AI systems trained on incomplete or biased datasets can perpetuate disparities, in care as in other aspects of life. Clinicians must be vigilant in recognizing these risks and advocate for inclusive, representative data in AI development.
Data Privacy and Patient Consent
The use of patient data in AI applications must comply with legal and ethical standards. Patients should be aware that making available their data may be a form of participation in the common good and in improving medical knowledge and practice, but this should be an expression of free responsibility. The ethical profile of medical profession about confidentiality and management of information must be transposed also in the context of AI.
Responsibility and Liability
Errors could be expression of a failure of programming, supervision, clinicians’ action or of the algorithm. Therefore, it would be important to differentiate when the error can be traced back to the doctor for the improper use of these systems, or when it is only and exclusively attributable to the hospital that manages and sets the instrument, or of the AI company. The physician has also the responsibility to warn his/her patient of the dangers to use generative AI as psychological advisers and to help him/her to escape the trap of confinement in “numerical bubbles”.
Access and Fairness
AI should not widen the gap between resource-rich and resource-poor settings. Its deployment must be equitable, ensuring that all patients—regardless of geography or socioeconomic status—benefit from technological advances: “Optimizing resources means using them in an ethical and fraternal way, and not penalizing the most fragile”.
As stewards of patient care, physicians have a critical role in shaping how AI is adopted. By insisting on ethical rigor and patient-centered design, we can ensure that AI strengthens—not compromises—the integrity of medical practice. We are at the same time aware that we need to cooperate with other players working in the field: the power and interests at stake in the research and control of digital technologies make an alliance between all stakeholders essential. This is what “Ethics by design” is all about.
Numerical techniques based on computations, as effective they may be, have many epistemological and logical limits. Therefore, they cannot replace all the facets of human thought and all the dimensions of human relations. Some deep dimensions of patient care cannot be replaced by optimized numerical procedures and autonomous robots. They implies empathic gestures, look full of tenderness, and to take time without any consideration for effectiveness and profitability. AI cannot lead to forget that medicine is not only a science or a technique but a human way to support the patient in his/her suffering, even when any technology is useless. The major risk of AI successes in medicine could well be the insidious way to suggest progressively that medicine is only a technique to cure and not a deep human relation of care. The patient is not a problem to be solved (by AI or other technologies) he/she is a mystery revealing the Christ Himself.
Mons. Renzo Pegoraro,
President of the Pontifical Academy of Life
Prof. Bernard Ars
President of FIAMC
References
1.- Pope Francis, Address at the G7 Session on Artificial Intelligence in Borgo Egnazia (Puglia) (14 June 2024), in L’Osservatore Romano (14 June 2024) 4.
2.- Pope Francis, Message for the LVII World Day of Peace (1 January 2024), 2, in L’Osservatore Romano (14 December 2023) 2; cfr Second Vatican Ecumenical Council, Pastoral Constitution Gaudium et Spes(7 December 1965), 35 ; See also the instruction Dignitas personae on certain bioethical questions (Congregation for the Doctrine of the Faith, 8 September 2008).
3.- Dicastery for the Doctrine of Faith and Dicastery for Culture and Education, Antiqua et nova. Note on the Relationship Between Artificial Intelligence and Human Intelligence (14 Jenuary 2025), 74.
4.- Pope Francis, Address to the Participants at the Meeting Sponsored by the Charity and Health Commission of the Italian Bishops’ Conference (10 February 2017).
5.- Rome Call for AI Ethics (28 February 2020), in https://www.vatican.va/roman_curia/pontifical_academies/acdlife/documents/rc_pont-acd_life_doc_20202228_rome-call-for-ai-ethics_en.pdf.
6.- Cfr Leo XIV, Apostolic exhortation, Dilexi te, of the Holy Father Leo XIV to all christians on love for the poor: “The Christian tradition of visiting the sick, washing their wounds, and comforting the afflicted is not simply a philanthropic endeavor, but an ecclesial action through which the members of the Church ‘touch the suffering flesh of Christ.’ ” (n. 49).
