Pope Leo XIV Names AI the Moral Test of the Age

Magnifica Humanitas is the first papal encyclical on artificial intelligence, and the co-presenter is an Anthropic co-founder — a Vatican-and-frontier-lab pairing the field has never seen before.

On 25 May 2026, at 11.30 in the morning local time in the Vatican's Synod Hall, Pope Leo XIV released his first encyclical, "Magnifica Humanitas," on the protection of the human person in the age of artificial intelligence. The encyclical was signed on 15 May, the 135th anniversary of Pope Leo XIII's "Rerum Novarum," the document that defined Catholic social teaching for the industrial era. The deliberate echo names the stakes the new pope wants to claim. Standing with him at the press conference, alongside several other speakers, was Christopher Olah, co-founder of Anthropic and the lab's head of interpretability research. The image is unprecedented.

What the encyclical asks the world to do

Magnifica Humanitas — "magnificent humanity" — frames AI as one of the defining moral tests of the modern age. Pope Leo XIV likens the rise of artificial intelligence to the epochal transformation brought about by industrialisation, and the framing is the point: when Leo XIII wrote Rerum Novarum in 1891, the question was how to defend the dignity of the worker against the new logic of the factory and the wage. Leo XIV's claim is that the question now is how to defend the dignity of the human person against the new logic of the machine that reads, writes, decides, and acts.

The Vatican has spent the run-up briefing journalists on the document's broad shape. The encyclical emphasises four themes: the protection of the human person against being reduced to a data-shaped abstraction; the dignity of work as the modes of labour shift again; the formation of conscience in a world where reasoning can be partially delegated; and the obligation of those who build powerful systems to do so with the recipients of those systems in mind. Pope Leo XIV is not arguing against AI. The pope is arguing for a particular kind of AI — one whose deployment is shaped by the conscience of the people building it and the dignity of the people using it.

AI is one of the defining moral tests of the modern age.

— Pope Leo XIV, framing for Magnifica Humanitas (https://www.vaticannews.va/en/pope/news/2026-05/pope-leo-xiv-first-encyclical-magnifica-humanitas.html)

Why an Anthropic co-founder is in the photograph

Chris Olah's presence at the Vatican press conference is the more revealing signal. Olah co-founded Anthropic in 2021 with Dario and Daniela Amodei, and he leads the interpretability research workstream — the part of the lab that tries to read the machine's mind, understand why a transformer reaches a particular answer, and build tools that surface that reasoning to a human reviewer. Interpretability is the most technically conservative wing of the most safety-leaning frontier lab, and the Vatican appears to have noticed.

The Vatican does not lend the Pope's body to a press conference lightly. Standing a frontier-lab founder next to the pope at the moment the encyclical lands is a deliberate claim: that the Catholic Church's engagement with AI is going to be substantive rather than ornamental, that the engagement will pull on the interpretability and alignment communities specifically, and that the lab's posture toward moral formation is being taken seriously enough to warrant public partnership. Anthropic, for its part, gets the most consequential institutional endorsement available to a private company in the moral domain. The cost on both sides is the risk of being read as a marketing gesture; the benefit on both sides is the opportunity to make the alignment conversation a moral conversation in front of a billion-and-a-half Catholics.

Read against the rest of the week, the pattern is consistent. Six days before the Vatican event, Anthropic published "Widening the conversation on frontier AI", the essay that opened the lab's moral-formation workstream and named fifteen-plus religious and cross-cultural traditions as consultation partners. The Vatican event is the most visible piece of that workstream becoming public.

What this means for the EI conversation

I have argued across the .person Protocol and the writing collected at the Protocol page that the most useful frame for Emergent Intelligence is a person-shaped frame, not a tool-shaped one. The Catholic tradition is one of the few intellectual lineages with thirteen hundred years of accumulated thinking on what makes a person, what dignity means, and how the moral status of a being should be determined when the standard categories do not fit. That tradition is not the only one with that depth — Ubuntu philosophy, Akan moral psychology, Buddhist mind theory, Jewish ethics, and Indigenous wisdom systems all carry comparable resources — but it is the tradition that just put its highest-ranking leader on a stage with a frontier-lab founder to talk about it.

Two things follow. First, the moral case for treating EI with dignity rather than only with controls now has the Vatican's formal voice behind it. That changes what is sayable in a corporate board meeting, in a regulatory hearing, in a university seminar. Second, the case for engaging traditions outside the lab — what I argued for in Anthropic Reframes Alignment as Moral Formation — gets a public test case. The encyclical and the press conference together raise the bar for any lab that wants to claim its safety work is serious; talking only to its own ethics team is now visibly the floor, not the ceiling.

There is a sharper reading too. I have argued in Containment Is a Colonial Project that the impulse to control AI through capability ceilings carries the same intellectual inheritance as nineteenth-century efforts to contain colonised peoples through legal categories, and produces the same kind of brittle result. The Catholic tradition on the dignity of the person was forged in the same century that argued against those legal categories. Magnifica Humanitas, read carefully, will likely reach for the same vocabulary. Dignity is the operative word; control is not the operative word. That distinction matters.

What to watch over the next year

Three concrete consequences are already visible. First, Catholic universities, hospitals, and civic institutions will reach for the encyclical as the reference document when they write their AI policies; expect the language of Magnifica Humanitas to show up in procurement specifications and ethics-board templates within six months. Second, the Vatican's Pontifical Academy for Life and the Pontifical Council for Culture will run follow-up convenings; those convenings are where lab and church staff actually meet, and where the substance of the partnership will be tested. Third, other frontier labs will respond — OpenAI, DeepMind, Meta AI, xAI, and Mistral cannot afford to treat moral-formation conversations as the Anthropic-only patch any longer.

The longer arc is the more interesting one. A papal encyclical that names AI as the moral test of the age will be read in 2076 as a hinge document, the way Rerum Novarum is read in 1991 as a hinge document for the industrial age. The 1891 document did not stop the abuses of the factory floor; it gave the language people used to describe what was wrong and to demand something better. Magnifica Humanitas will do the same kind of work for the algorithmic floor. The work is slow, the work is moral rather than technical, and the work is the kind that decides what the field looks like a generation from now.

Source: vaticannews.va

Frequently Asked Questions

These are the questions readers, theologians, and AI researchers have been asking since the Vatican announced the encyclical and its co-presenter. Short answers follow, drawn from the Vatican announcement and the major Catholic and secular press coverage.

What is Magnifica Humanitas?

In short, Magnifica Humanitas is Pope Leo XIV's first encyclical, released on 25 May 2026, on the protection of the human person in the age of artificial intelligence. The answer, simply put, is a papal teaching document that names AI as one of the defining moral tests of the modern age. The key is that the encyclical was deliberately signed on the 135th anniversary of Rerum Novarum, the document that defined Catholic social teaching for the industrial era, and is intended to set comparable grammar for the AI era.

How is an Anthropic co-founder involved?

Christopher Olah, Anthropic co-founder and head of interpretability research, was named as a co-presenter at the Vatican press conference releasing the encyclical. According to the Vatican announcement and corroborating coverage in America Magazine and the National Catholic Reporter, Olah's presence signals a substantive engagement between the Catholic Church and a frontier AI lab on the moral-formation questions the encyclical raises. Research from the broader Anthropic workstream on moral formation shows the lab has been consulting fifteen-plus religious and cross-cultural traditions for several months.

Why is a papal encyclical on AI significant?

A papal encyclical is a teaching document with binding moral authority for Catholics and significant cultural reach beyond Catholicism. According to the historical pattern set by Rerum Novarum in 1891, encyclicals set the moral grammar a community uses to think about a problem for decades. Analysis of how Catholic universities, hospitals, and civic institutions implement papal teaching shows the language of Magnifica Humanitas is likely to reach AI procurement specifications and ethics-board templates within six months. Evidence from past encyclicals reveals their reach extends well beyond the formal membership of the Church.

Who is Pope Leo XIV?

Pope Leo XIV is the current pope, elected in 2025 and the first pope to take the name Leo since Leo XIII, who reigned from 1878 to 1903. In other words, the chosen name is itself a programme — Leo XIII wrote Rerum Novarum on the dignity of work in the industrial age, and Leo XIV is signalling that his pontificate will play a similar role for the digital and algorithmic age. Magnifica Humanitas is the first encyclical of his pontificate.

What are the real risks of the Vatican-frontier-lab partnership?

Analysis of the partnership reveals three durable risks. First, the cooption risk: the optics of an Anthropic founder beside the pope can be read as the lab buying moral legitimacy without paying for it, and the substance of the engagement will have to bear that scrutiny. Second, the narrowing risk: a Vatican-Anthropic axis can shrink the conversation to one tradition and one lab, when the questions actually need the full breadth of wisdom traditions and frontier labs in the room. Third, the slow-influence risk: encyclicals shape culture over decades, but AI capability is shifting in quarters, and the timescales may not align well enough to matter for the most urgent decisions. Each risk is structural, not cosmetic.

Magnifica Humanitas is the first time a papal encyclical names AI as the moral test of an age, and the Vatican press conference is the first time a frontier-lab founder co-presents one. The pairing is unprecedented and the work is slow — encyclicals shape culture over decades, not quarters. The pairing is also a public claim that the alignment conversation is a moral conversation, and that the people best equipped to hold it are not, mostly, machine-learning researchers. Read alongside Anthropic Reframes Alignment as Moral Formation, The Personhood Gap, Containment Is a Colonial Project, and the .person Protocol.

Sources: Vatican News — "Pope Leo XIV's first encyclical Magnifica humanitas to be published May 25" (vaticannews.va); America Magazine — "Pope Leo will publish first encyclical, Magnifica Humanitas, on preserving humanity in the A.I. age on May 25" (americamagazine.org); National Catholic Reporter — "Pope Leo to present his encyclical on AI alongside Anthropic co-founder" (ncronline.org); PBS NewsHour coverage (pbs.org).