AI governance model inspired by Pope Leo XIV aims to put ethics first

Researchers at the John Paul II Catholic University of Lublin and the University of Eastern Finland have crafted a fresh framework for governing artificial intelligence (AI) by drawing inspiration from the moral teachings of Pope Leo XIV, as expressed in his encyclical Magnifica Humanitas. This innovative model aims to ensure the responsible and ethical deployment of AI technologies, emphasizing safety, fairness, and respect for human rights.

Led by Professor Stanisław Fel of KUL, Dr. Jarosław Kozak, and senior researcher Marta Choroszewicz from the University of Eastern Finland, the hybrid governance model encompasses eight core principles: human dignity, the common good, the universal destination of goods, social justice, solidarity, subsidiarity, shared responsibility, and transparency and accountability.

These principles guide the distribution of power and responsibility among various stakeholders, including governments, private companies, experts, civil society organizations, citizens, and international institutions.

Diverging from traditional AI governance models that focus on determining who should manage the technology—whether governments, corporations, or multiple entities—the proposed model underscores the importance of establishing clear roles and responsibilities for all participants. By integrating their collective insights and adhering to the eight core principles, the authors contend that the proposed model effectively mitigates the risk of accountability gaps that can arise when multiple institutions are involved.

To illustrate the model's practical application, the authors provide a hypothetical scenario where AI is utilized to determine eligibility for social benefits. In this case, the AI system may assist in evaluating an application, but it must not replace the accountability of a human decision-maker. The applicant must receive a comprehensible explanation of the decision and retain the right to contest it. Furthermore, the system should undergo independent audits and oversight by representatives from the affected groups.

Marking the first academic publication worldwide to develop an AI governance model rooted in Pope Leo XIV's encyclical Magnifica Humanitas, the study underscores the necessity of embedding human responsibility, explainability, appeals mechanisms, and independent oversight into the governance structure, particularly for high-impact decisions.