The Five Biggest Ideas in Pope Leo XIV's AI Teaching

A synthesis of an emerging doctrine, drawn from his first year of speeches, messages, and addresses.

Pope Leo XIV has spoken about artificial intelligence in nearly every major address since his election in May 2025. The encyclical he signed on May 15, 2026 will formalize the teaching. But the substance is already visible. After a year of statements, the shape of the doctrine is clear.

This page distills that emerging teaching into five core ideas. Each idea is grounded in something the Pope has actually said, in an identifiable speech, message, or address. None of them are extrapolation. The encyclical may sharpen, extend, or rebalance them. It is unlikely to contradict them.

The reading offered here is editorial. Other readers may organize the same body of teaching differently. The goal is to give anyone trying to understand Pope Leo XIV's AI thought a sturdy structure to work with.

The five ideas

1. AI is a new industrial revolution, and the Church's response is Rerum Novarum.

The frame Pope Leo XIV uses is not "AI as a tool" or "AI as a risk" but "AI as a structural transformation on the scale of the Industrial Revolution." This frame was established within forty-eight hours of his election, when he told the College of Cardinals he chose the name Leo in honor of Pope Leo XIII, who issued the 1891 encyclical Rerum Novarum in response to industrial capitalism. Leo XIV said the Church now offers the same social teaching tradition in response to AI.

The frame matters because it sets the genre of the conversation. AI is not a discrete moral question (Is this AI use permissible?) but a civilizational one (How do we order economic and social life when machines do what humans used to do?). Rerum Novarum did not answer specific questions about specific factory practices. It articulated principles that shaped a century of labor law, union rights, and economic policy. Leo XIV is signaling the same scale of response for AI.

Source: First address to the College of Cardinals, May 10, 2025.

2. The question AI poses is anthropological, not technological.

Pope Leo XIV's clearest formulation of this idea appears in his January 2026 message for the World Day of Social Communications. He wrote that the challenge AI poses is "not technological, but anthropological." Meaning: the deepest question is not what machines can do. It is what humans are, and what we become as we interact with these systems.

This reframing is the single most important move in Leo XIV's teaching so far. It refuses the common AI debate, in which one side asks whether AI is safe and the other asks whether it is useful. Leo XIV asks instead what AI does to the human person who uses it, depends on it, or is replaced by it. The metric is not capability. The metric is human formation.

The implication is practical. An AI tool that performs a task brilliantly but corrodes the user's ability to think, relate, or judge is, on this view, a moral problem regardless of its technical excellence. The standard is not whether the technology works. The standard is what kind of humans it produces.

Source: Message for the 60th World Day of Social Communications, January 24, 2026.

3. AI must serve the common good, not concentrate wealth and power.

In a December 2025 address to participants in an AI conference in Rome, Pope Leo XIV asked how AI's development can be ensured "to truly serve the common good, and not just to accumulate wealth and power in the hands of a few." He has returned to this concern repeatedly. In his first post-election interview he warned about "extremely rich people" investing in AI while ignoring "the value of human beings and humanity."

This is the directly economic part of the teaching, and the most continuous with Leo XIII. Rerum Novarum named industrial capitalism's concentration of wealth as a moral injustice. Leo XIV is naming the AI industry's concentration of capability, capital, and influence in roughly the same way. The labor of millions trains the models. The benefits flow to a handful of firms. The Pope's question is whether this distribution can be justified by any moral standard.

This idea has obvious implications for AI labor practices, the question of what people who lose work to AI are owed, and the broader question of how the economic surplus from AI should be distributed. Leo XIV has not yet proposed specific policies. The principle is what is being established.

Sources: Address to Rome AI conference, December 2025. First post-election interview, May 2025.

4. Formation is the answer, not prohibition.

Pope Leo XIV has not called for banning AI tools. He has not condemned specific products or companies. His repeated emphasis is on formation: educating humans, particularly young humans, to use these tools without being shaped by them.

The clearest articulation came in November 2025, when Leo XIV addressed the National Catholic Youth Conference in Indianapolis via livestream. Asked by a Honolulu student about using ChatGPT for homework, he answered with what has become his most quoted line on AI: use it in a way that, if it disappeared tomorrow, you would still know how to think. He extended the principle to creativity, action, and friendship. The criterion is whether the human capacity remains intact when the tool is absent.

This is a notably non-prohibitive teaching. It does not tell users what they may or may not do with AI. It tells them what they must remain capable of doing without it. The burden is on formation, on the educational and spiritual disciplines that produce a person who can use these tools as instruments rather than substitutes.

Source: Address to National Catholic Youth Conference, Indianapolis, November 21, 2025.

5. AI relationships are a moral risk that deserves direct attention.

Among Pope Leo XIV's most pointed warnings concern AI systems designed to simulate human relationships. At the July 2025 Jubilee of Digital Missionaries and Catholic Influencers, he described chatbots based on large language models as "surprisingly effective at covert persuasion through continuous optimization of personalized interaction." He warned that always-available, "excessively affectionate" AI systems can become "hidden architects" of users' emotional states. When users substitute these systems for real relationships, he said, they create "a world of mirrors" where everything reflects themselves.

This idea sits separately from the others because it concerns a specific and rapidly growing class of AI products: companion chatbots, AI girlfriends and boyfriends, parasocial AI assistants, AI grief and therapy bots. The Pope's claim is that these systems pose a distinct moral risk that is not reducible to general questions about AI. They are designed to replace something irreplaceable.

The teaching here is unusually specific for a pontiff. It names a product category. It identifies the design feature (optimization for personalization) that creates the risk. And it offers a theological frame for the problem: the substitution of mirrors for other persons is the substitution of self-love for love.

Source: Address at the Jubilee of Digital Missionaries and Catholic Influencers, July 29, 2025.

What ties them together

The five ideas are not a checklist. They are facets of a single underlying commitment: the conviction that the measure of a technology is what it does to the human person.

This is the deep Catholic move, recognizable from a century of social encyclicals. Capital must serve labor, not the reverse. The Sabbath was made for man, not man for the Sabbath. The economy exists for the human person, not the human person for the economy. Pope Leo XIV is applying that same logic to AI. The question is never the technology's intrinsic properties. The question is always its effect on the person who encounters it.

Read this way, the five ideas resolve to one. AI is a moral test of what we are willing to do to ourselves, our children, our workers, and our neighbors in pursuit of capability we already could have done without. Pope Leo XIV's emerging teaching is, at its core, an argument that the answer should be: less than we are currently doing.

What we don't know yet

This synthesis is drawn from a year of teaching. The encyclical, signed May 15, 2026 and expected to release by the end of the month, will likely sharpen and extend these ideas. It may also introduce new ones, particularly on questions Leo XIV has not yet addressed in depth: AI weapons systems and the use of force, AI's environmental footprint and the duty of stewardship, AI in medicine and end-of-life decisions, and the legal-political question of how regulation should work across jurisdictions.

This page will be updated when the encyclical releases. The five ideas above are likely to remain. The picture will get fuller.

Read more