When Pope Leo XIV released his first encyclical, Magnifica humanitas, on 15 May, it did more than add another document to the Vatican’s archives; it sent a deliberate signal to the global conversation about artificial intelligence. Far from being a purely theological treatise, the encyclical reads as a urgent moral diagnostic, pinpointing who truly steers the direction of AI development and for whose benefit. By framing AI governance as a question of justice and human dignity, the Pope invites policymakers, technologists, and citizens alike to look beyond the hype and examine the underlying power structures that shape everyday algorithmic decisions. This reframing is crucial because it moves the debate from technical specifications to the realm of societal values, asking whether the systems that influence hiring, lending, healthcare, and even democratic participation are serving the common good or consolidating advantage in few hands.
The voice behind this interpretation is Paolo Benanti, a professor of ethics of technology and moral philosophy at LUISS Guido Carli University in Rome, who advises both the Vatican and the United Nations on AI matters. Benanti’s unique position allows him to translate ecclesiastical teachings into actionable insights for secular institutions, bridging the gap between ancient moral traditions and cutting‑edge technological challenges. His work demonstrates that religious ethics can offer a robust framework for evaluating emerging technologies, not by imposing doctrine, but by emphasizing timeless principles such as solidarity, subsidiarity, and the preferential option for the vulnerable. In the context of AI, these principles translate into demands for transparency, accountability, and inclusive design processes that consider the impacts on marginalized communities before deployment.
Papal encyclicals have historically addressed bishops, but modern popes have increasingly used them as universal moral statements intended for all people of goodwill. Magnifica humanitas follows this trend, presenting itself not as canon law but as a reflective document that invites dialogue across faiths, cultures, and disciplines. By choosing the encyclical format, Pope Leo XIV leverages a respected authority to highlight ethical blind spots in the AI ecosystem that technical standards alone cannot see. This approach recognizes that technology never develops in a vacuum; it is embedded in social, economic, and political relationships that shape its effects. Consequently, the encyclical serves as a reminder that ethical foresight must accompany technical innovation, lest we build powerful tools that exacerbate inequality or erode democratic norms.
One of the encyclical’s central diagnoses is the unprecedented concentration of power in the hands of a few private actors who determine the values encoded in AI systems used by billions. Unlike traditional industries where regulation and market competition can diffuse influence, the AI landscape is marked by high barriers to entry, massive data requirements, and network effects that favor incumbents. As a result, a small coalition of multinational corporations and influential foundations effectively decides what counts as fairness, efficiency, or even truth in algorithmic outputs, often without democratic legitimacy or transparent oversight. This privatization of moral judgment raises profound concerns about who gets to shape the societal norms that govern everything from credit scoring to content moderation.
From a market perspective, this concentration has tangible consequences. Startups and smaller players struggle to access the computational resources and proprietary datasets needed to compete, leading to a homogenization of AI products that reflect the priorities of dominant firms. Investment flows disproportionately toward applications that promise rapid returns—such as targeted advertising, predictive policing, or automated hiring—while socially beneficial but less lucrative uses, like accessibility tools or climate modeling, receive less attention. Moreover, the lack of clear regulatory frameworks in many jurisdictions allows these actors to operate across borders with minimal accountability, creating a race to the bottom where ethical shortcuts can be rewarded by market performance. Investors and corporate boards are beginning to notice, however, as reputational risks and potential litigation loom large for firms perceived as neglecting AI’s societal impacts.
The encyclical also warns against the “intelligence illusion,” the tendency to overestimate AI’s capabilities and anthropomorphize its outputs. While modern models can generate remarkably fluent text or recognize complex patterns in data, they lack genuine understanding, intentionality, or moral reasoning. Treating them as infallible authorities can lead to dangerous overreliance, especially in high‑stakes domains like criminal justice, healthcare diagnostics, or financial lending. When decision‑makers defer to algorithmic scores without scrutinizing their assumptions or limitations, they risk automating bias, amplifying errors, and undermining human judgment. The Pope’s message urges a sober assessment: AI should augment, not replace, the nuanced deliberation that only humans can provide, particularly when ethical trade‑offs are involved.
Drawing on the Catholic social tradition, Benanti suggests that AI governance can benefit from principles that have guided ethical deliberation for centuries. The principle of human dignity insists that every person must be treated as an end in themselves, never merely as a data point or a means to profit. The common good calls for arrangements where the benefits of technology are shared broadly, especially with those who are least advantaged. Subsidiarity argues that decisions should be made at the most local level competent to handle them, implying that communities affected by AI should have a meaningful voice in its design and deployment. By translating these concepts into concrete criteria—such as impact assessments, participatory design, and redress mechanisms—policymakers can create regulatory environments that encourage innovation while safeguarding fundamental rights.
For governments and international bodies, the encyclical offers a roadmap for meaningful action. First, legislators should consider establishing AI impact assessments that evaluate not only economic efficiency but also effects on equity, privacy, and democratic participation, similar to environmental impact statements. Second, creating multi‑stakeholder oversight boards that include technologists, ethicists, civil society representatives, and affected communities can help ensure that regulatory decisions reflect diverse perspectives. Third, public procurement policies can leverage purchasing power to demand transparency and ethical standards from AI vendors, thereby shaping market incentives. Finally, investing in digital literacy programs equips citizens to critically engage with AI‑driven services, fostering an informed public capable of holding institutions accountable.
Business leaders, too, have a clear role to play in responding to the Pope’s call. Adopting an “ethics by design” approach means integrating moral considerations from the earliest stages of product development, rather than bolting them on after the fact. This includes conducting bias audits, establishing explainability standards, and implementing robust data governance practices that respect user consent and privacy. Companies should also create internal AI ethics committees with genuine authority to pause or redirect projects that raise red flags. Transparency reports detailing how models are trained, what data they use, and how outcomes are monitored can build trust with customers and regulators alike. Moreover, aligning executive compensation with long‑term societal impact metrics can discourage short‑termist behaviors that sacrifice ethical standards for quarterly gains.
Individuals are not powerless in this landscape. Developing a critical AI literacy—understanding how recommendation engines work, recognizing when a decision is automated, and knowing how to contest unfair outcomes—empowers people to navigate AI‑mediated environments more safely. Supporting advocacy groups that push for algorithmic transparency and participating in public consultations on AI policy can amplify collective voice. When using AI tools in personal or professional contexts, individuals should routinely ask: Who benefits from this system? What assumptions underlie its outputs? Are there avenues for appeal if something goes wrong? By cultivating these habits, users become active participants in shaping a more just AI ecosystem rather than passive consumers of opaque technology.
In sum, Pope Leo XIV’s Magnifica humanitas, interpreted through Paolo Benanti’s expert lens, offers a timely reminder that the challenges posed by artificial intelligence are fundamentally moral and political, not merely technical. The encyclical’s analysis of power concentration, the intelligence illusion, and the need for value‑centric governance provides a framework that resonates across sectors. Market actors, policymakers, and citizens each have distinct levers to influence the trajectory of AI in ways that uphold human dignity and the common good. Ignoring this moral dimension risks entrenching inequities and eroding trust; embracing it opens the pathway to innovations that truly serve humanity.
Actionable advice: For policymakers, launch AI impact assessments that include equity and democratic participation metrics, and create inclusive oversight boards. For business leaders, embed ethics‑by‑design practices, establish independent AI ethics committees, and publish regular transparency reports. For technologists, prioritize explainability and bias testing, and advocate for participatory design with affected communities. For individuals, strengthen AI literacy by learning how common algorithms work, question automated decisions, and support advocacy for algorithmic accountability. By taking these concrete steps, stakeholders can help steer AI toward a future that amplifies human flourishing rather than concentrating power in few hands.