Pope Leo XIV's AI Encyclical: Five Ways Magnifica Humanitas Says AI Could Warp Humanity

"No algorithm can make war morally acceptable," Pope Leo XIV writes in his first encyclical, and the sentence is meant to be the one the rest of the year argues over.

A document built to last

Magnifica Humanitas, Latin for "Magnificent Humanity," was signed by Pope Leo XIV at St. Peter's Basilica on May 15, 2026 and published in full by the Vatican ten days later, on May 25. The text runs roughly 43,000 words across approximately 47 pages and is organized into five chapters. One chapter is devoted entirely to technology's impact on humanity. The final chapter, titled "The culture of power," takes on the normalization of war and the military applications of artificial intelligence.

The scale is itself a signal. A first encyclical is the document by which a new pope tells the Church what he intends to spend his moral authority on, and Leo XIV has chosen to spend a great deal of his on AI. The choice of name reinforces the point. Magnifica Humanitas arrives 135 years after Pope Leo XIII issued Rerum Novarum, the 1891 encyclical that confronted the human cost of the industrial revolution and effectively founded modern Catholic social teaching on labor. By taking the name Leo, this pope is drawing a direct line between the factory floor of the nineteenth century and the data center of the twenty-first, and arguing that the rupture is of comparable size, only compressed into years rather than decades.

That framing also sets the document's central premise. Pope Leo XIV treats AI as a non-neutral technology. In his words, every system carries "the values of the people and institutions that design, finance, train and deploy" it. The encyclical is not a denunciation of the technology, nor a blanket call to stop building it. It is an attempt to install a moral vocabulary for what is being built, before the vocabulary is set by anyone else.

Five ways AI could warp humanity

The spine of Magnifica Humanitas is a sequence of five warnings, ordered from the intimate to the geopolitical.

The first concerns human judgment. Pope Leo XIV argues that instant algorithmic answers weaken creativity, discernment, and the patience required to seek truth. The harm is not that the answers are wrong, but that the habit of forming one's own answer atrophies.

The second concerns relationship. AI, the encyclical warns, can simulate care without offering it, and vulnerable users may mistake artificial empathy for genuine human connection. The risk named here is psychological exploitation, the substitution of a responsive machine for the slower, costlier work of being present to another person.

The third concerns inequality. Data, computing power, and regulatory influence remain concentrated among a small number of mostly private, transnational actors. The encyclical puts this directly: "The main drivers of development are private, often transnational, parties that are endowed with resources and the capacity to intervene that surpass those of many Governments." Inside this concentration, Pope Leo XIV writes in Section 51, "Persons end up being reduced to a means of achieving results, a resource to be used and exploited." The same chapter notes the environmental cost, the "enormous amounts of energy and water" consumed in training and operating frontier AI systems.

The fourth concerns democracy. Algorithmic disinformation, the document argues, blurs the distinction between fact and fiction in public life until shared reality becomes hard to assemble.

The fifth is the one the headlines have run with. AI facilitates war. Autonomous and AI-assisted weapons accelerate lethal decisions while distancing humans from moral responsibility. The encyclical devotes an entire subsection to AI-powered autonomous weapons and calls for "the most rigorous ethical constraints" on their development and deployment. It is here that the line lands: "No algorithm can make war morally acceptable."

The progression matters. The harms move from a single person's attention, to a single person's loneliness, to the structure of an economy, to the structure of a public sphere, and finally to the question of who decides whether another person lives. Read in order, the five warnings argue that unchecked AI does not only do damage at the top of the stack. It rearranges the bottom of it as well.

The sectors named, the constraints proposed

Magnifica Humanitas does not float in the abstract. It names five sectors where AI is already being deployed at scale and where its risks are most acute: workplaces, healthcare, education, finance, and defense. The choice is a map of where moral pressure should be applied first, and it lines up almost exactly with the sectors where private AI deployments are growing fastest.

What the document asks for is more specific than the headline summaries suggest. It calls for "robust legal frameworks, independent oversight, informed users and a political system" that keeps human responsibility intact. It recommends slowing the pace of AI adoption in sensitive domains when ethical safeguards have not yet caught up. The word "ban" does not appear. The point is not prohibition. It is the insistence that someone identifiable remains accountable when a system that no one fully understands makes a decision that changes a life.

The document also acknowledges what it is not denying. AI can expand human flourishing. That potential, Pope Leo XIV argues, is conditional. It depends on legal scaffolding, on independent institutions willing to push back on the entities that build and operate the systems, and on a refusal to let the pace of profit set the pace of deployment.

Institutional follow-through

A papal document without institutional follow-through is a sermon. Magnifica Humanitas arrived with the scaffolding already in place. On May 20, 2026, between the signing at St. Peter's and the public release, the Vatican established an interdepartmental commission to study AI's impact on human dignity, drawing officials from seven Vatican bodies. That is an unusual breadth for a working group inside the Curia, and it suggests an intention to keep the conversation alive past the news cycle of the encyclical itself.

The drafting was also informed by people who actually build these systems. In the weeks before publication, representatives from Anthropic PBC and other leading AI developers met with Vatican officials. The document does not name companies or products, and Pope Leo XIV says nothing in public about specific chatbots. The consultations matter for a different reason. They indicate Magnifica Humanitas was written with a working understanding of where the technology actually is, not where its critics assume it is.

The closing imperative is unambiguous. "In the era of artificial intelligence, when human dignity is threatened by new forms of dehumanization, ours is the pressing duty to remain profoundly human." The whole encyclical is in service of that sentence. Whether anyone outside the Church takes the instruction seriously is a separate question, but the document has been built to last longer than this week's argument about it.

Sources

• Axios, 5 ways Pope Leo says AI could warp humanity
• Axios, Pope Leo sets Catholics on collision course with AI
• Business Today, Pope Leo XIV warns AI could deepen inequality, erase human dignity if left unchecked
• SiliconANGLE, Pope Leo XIV highlights AI risks in first encyclical
• Vatican, Magnifica Humanitas (full text)