The story about the Pope, Gandalf and AI

Ronni Holmvig Strøm · 2026-05-26

On Monday, with Anthropic co-founder Chris Olah seated in the Synod Hall, Pope Leo XIV released his first encyclical. It is called Magnifica Humanitas, it runs to roughly 40,000 words, and it calls

On Monday, with Anthropic co-founder Chris Olah seated in the Synod Hall, Pope Leo XIV released his first encyclical. It is called Magnifica Humanitas, it runs to roughly 40,000 words, and it calls on the world to "disarm" artificial intelligence. The headlines today are about the word "disarm" and the Gandalf quote in paragraph 213. Both are real. Neither is the most interesting thing in the document.

The Staged Echo of 1891

Leo signed the encyclical on May 15, which is not an accident. May 15 is the 135th anniversary of Rerum Novarum, the 1891 encyclical in which Leo XIII confronted industrial capitalism and grounded what became modern Catholic social teaching. By signing on that date, and by taking the name Leo, the current Pope is making the parallel structural. Industrial capital then; data and compute capital now. He calls AI the res novae of our time, the new thing.

The argument that follows reads like 1891 ported forward. There is a new form of power. It accumulates in few hands. It has its own colonial geography: "health data, epidemiological profiles, genetic maps and demographic information" extracted from "regions marked by structural fragility and limited geopolitical relevance," what Leo calls "the new 'rare earths' of power."

His sentence "the digital age will not be post-colonial, but colonial in another form" is doing the work that the labor passages of Rerum Novarum did in their time. It names a new asymmetry and refuses to call it natural.

The novelty is who was in the room. The co-founder of a frontier AI lab sat in the Synod Hall while the Pope read this aloud. In 1891 the industrialists were not invited to the Vatican to hear themselves described. In 2026 they are showing up voluntarily, and calling for partnership after the fact. The labs are seeking a kind of moral authority they cannot manufacture on a roadmap.

The Category Claim

The sharpest sentence is the one in which Leo refuses the word intelligence altogether. AI systems, he writes, "merely imitate certain functions of human intelligence." They "do not undergo experiences, do not possess a body, do not feel joy or pain, do not mature through relationships, and do not know from within what love, work, friendship, or responsibility mean. Nor do they have a moral conscience."

So right there, the church makes a claim about conciousness or not, present and future.

This is a category claim, and a more philosophically aggressive one than most Western governments have been willing to make. The EU AI Act regulates AI as a powerful technology. The White House executive orders treat it as a national security asset. The Vatican is saying something different and prior: the thing you have built is not what you have named it. The marketing word is wrong.

The industry's entire valuation rests on the marketing word. If a frontier system is intelligence, then scaling it is a path to something like a mind, and the trillion-dollar capex makes sense as a moral and economic bet. If a frontier system is, in Leo's phrase, a fast imitator of certain functions, then the bet looks different. The encyclical is, among other things, an attempt to make the second framing intellectually respectable from a source the first framing cannot easily dismiss as Luddite or under-informed.

The Vatican is not opposed to the tool. It rolled out an AI translation system at St. Peter's this spring, sixty languages on the worshipper's phone. The objection is to the framing under which a useful imitator becomes the measure of the human person.