The Pope's AI Encyclical: What Magnifica Humanitas Actually Says About Warfare, Labor, and the Meaning of Being Human

The Pope's AI Encyclical: What Magnifica Humanitas Actually Says About Warfare, Labor, and the Meaning of Being Human

Rome Entered the AI Debate Yesterday

Pope Leo XIV stood at the Vatican's Synod Hall on May 25 and personally presented his first encyclical to the world. That was unusual enough. The second anomaly was who stood beside him: Christopher Olah, co-founder of Anthropic and the researcher who built the field of mechanistic interpretability. The first U.S.-born pope, a trained mathematician, chose to launch his most significant teaching document alongside one of the engineers most focused on understanding what is actually happening inside large AI systems.

Magnifica Humanitas runs to 42,300 words and 235 pages, organized across five chapters and an introduction. Its publication date was not accidental. Leo XIV signed it on May 15, 2026, the 135th anniversary of Rerum Novarum, the landmark 1891 encyclical from his predecessor and namesake Pope Leo XIII. That text, issued as factories were dismantling craft labor across Europe and America, argued that workers had dignity and could not be treated as interchangeable production inputs. It became foundational to labor law in the century that followed. The signature date is a thesis statement before the text begins.

The new encyclical applies the same structure to AI. Its central argument is framed in paragraph nine: "Technology is never neutral, because it takes on the characteristics of those who devise, finance, regulate and use it." From there the document builds a case that covers autonomous weapons, labor displacement, AI-synthesized media, children's cognitive development, and the concentration of AI capabilities in a small number of private firms. The position throughout is not that AI should be rejected but that the question of whose AI it is, and in whose interests it operates, is one humanity has not answered.

The Anthropological Frame

The encyclical's diagnostic sentence appears at paragraph four: "Never has humanity had such power over itself." The sentence is not celebratory. It describes vertigo, the awareness that the question of what to do with that power has not been resolved.

The document's core argument follows from that vertigo. AI, in the encyclical's framing, forces a confrontation with questions humanity has been avoiding: what are human cognition, labor, creativity, and relationship if machines can perform them? Leo XIV's encyclical argues the challenge is "not technological, but anthropological. Safeguarding faces and voices ultimately means safeguarding ourselves."

This is a specific theological claim as well as a practical one. In Catholic teaching, human beings are made in the imago Dei, a phrase that carries more freight than intelligence or capability. It encompasses creativity, the capacity for relationship, the dignity of labor as participation in creation, and the irreducible uniqueness of every person. AI systems that simulate those things are therefore touching something the tradition considers fundamental. The encyclical is building from that premise through every practical argument it makes.

On AI-synthesized media, the document goes further than standard concerns about misinformation. "By simulating human voices and faces, wisdom and knowledge, consciousness and responsibility, empathy and friendship, the systems known as artificial intelligence not only interfere with information ecosystems, but also encroach upon the deepest level of communication, that of human relationships." Human faces and voices, in this framing, are not just identity markers useful for authentication. They are the medium through which irreducible human encounters take place. Their simulation is an intrusion into a space that belongs to personhood.

Warfare and the Spiral of Annihilation

The encyclical's most direct political intervention concerns autonomous weapons. Leo XIV named specific conflicts in public addresses preceding the text, describing what is happening in Ukraine, Gaza, Lebanon, and Iran as illustrating "the inhuman evolution of the relationship between war and new technologies in a spiral of annihilation." The phrase is precise: an escalation dynamic, not a policy problem.

The document declares that the Catholic Church's traditional "just war" framework, which has governed Catholic thinking about armed conflict for centuries, is now "outdated" given the capabilities of AI-enabled warfare. In its place, the encyclical calls for categorical prohibition of lethal autonomous weapons systems that operate without meaningful human oversight, framing the demand as a moral position rather than a case-by-case regulatory question. The Vatican's reasoning is direct: decisions to end human life cannot be delegated to systems that cannot understand the moral weight of what they are doing.

The document also demands transparency about command chains when AI is involved in military strikes, insisting that accountability for targeted killing decisions must remain traceable to human decision-makers. The Pope wrote that it is "not permissible" to entrust irreversible, lethal decisions to AI systems, a position that creates direct friction with the Trump administration's push to expand AI use in military and surveillance contexts.

The Labor Question

The Rerum Novarum parallel is most legible on the question of work. Leo XIII argued, when factories were displacing craft workers, that labor was not a commodity. Leo XIV makes the same argument about AI displacing knowledge workers. The document addresses writers, coders, analysts, designers, and educators, noting that AI "has increasingly taken control of the production of texts, music and videos," putting human creative industries at risk of replacement.

The phrase used in pre-encyclical addresses to describe what this displacement produces is "passive consumers of unthought thoughts." It is not saying that AI-generated content is inferior. It is saying that when human creative work is replaced entirely, the act of creation itself is lost along with the output. The encyclical treats that loss as a moral harm, not merely an economic transition.

Paragraph 37 is specific about wages: fair pay is described as "the concrete means of verifying the justness of the entire socioeconomic system," and work is explicitly framed as something beyond a problem to be solved or an income source. The passage maps directly onto debates about whether AI productivity gains should flow to workers whose tasks AI has absorbed or accrue entirely to the firms deploying the systems. "The pursuit of greater profits cannot justify choices that systematically sacrifice jobs, because the human person is an end, not a means, and the economic order must remain subordinate to human dignity and the common good."

Power Concentration and the Limits of Self-Regulation

The encyclical is pointed about where AI capabilities currently reside. OpenAI and Anthropic are each valued at hundreds of billions of dollars, more than the GDP of many nations. Leo XIV wrote that the concentration of power and data in the hands of so few people in the private sector poses a particular danger, especially to children and vulnerable populations, and called for external regulation of their work.

The text is direct about the insufficiency of corporate ethics commitments: "It is not enough to invoke ethics in the abstract; robust legal frameworks, independent oversight, informed users and a political system that does not abdicate its responsibility are required. A more moral AI is not enough if that morality is determined by a few."

The document extends the governance argument into the question of whether AI systems can be understood at all. The encyclical calls for "adequate regulatory tools capable of upholding justice and curbing the distorting effects of technological power," but the requirement for meaningful oversight assumes the systems being overseen are legible. This is where Olah's presence at the launch becomes coherent. His career has been spent on whether AI systems can be understood internally, not just evaluated by their outputs. If governance requires comprehension, and comprehension requires interpretability, then the research field Olah built is the technical face of the encyclical's governance demand.

Children, Cognitive Development, and the Capacity to Think

Chapter Four addresses a concern rarely prominent in AI policy discussions: what AI-saturated environments do to how children develop cognitively and neurologically. Leo XIV has stated publicly that he is concerned about the effect of AI on children's "intellectual and neurological development," and the encyclical calls for a broad rethinking of what it means to educate people in the use of AI, with particular attention to young people.

The formulation he used with a stadium of teenagers before the encyclical's release is worth noting: use AI "in such a way that if it disappeared tomorrow, you would still know how to think." That is not a call to restrict AI for young people but a call to preserve a specific capacity, independent thought, that the Pope argues may require conditions that AI-saturated environments can undermine. Some cognitive capacities, the argument runs, atrophy when they are not used. Children who can always delegate reasoning to a machine may develop differently than those who cannot.

What Happens After an Encyclical

Encyclicals do not become law. Rerum Novarum did not either, and it still shaped labor policy across a century. The Catholic Church reaches 1.4 billion members through parishes, hospitals, universities, and NGOs, and through the formation of legislators and policymakers educated in Catholic institutions worldwide. Magnifica Humanitas establishes a vocabulary. The frames it introduces, centered on human dignity, the sacredness of faces and voices, the dignity of labor, children's development, and the governance of systems no one fully understands, are already moving through debates where Catholic social teaching has institutional weight.

The EU AI Act is being implemented now, in member states where Catholic political culture is influential. Debates about AI-generated media disclosure, autonomous weapons treaties, AI and children, and AI labor displacement are all taking place in the same territory the encyclical covers. The document offers something distinct from technical risk analysis or economic modeling: a claim about what humans are and what their dignity requires, made in a moral register that commands different audiences than white papers do.

The AI industry has largely framed its questions as engineering problems and market opportunities. That framing will continue. But it is now joined, publicly and at scale, by the moral authority of the oldest continuous institution in Western civilization, which looked at the same set of facts and reached a different kind of conclusion.