Pope Leo XIV's First AI Encyclical: What Magnifica Humanitas Actually Says

AI-generated, human-reviewed.

Pope Leo XIV's first encyclical, Magnifica Humanitas, is not the anti-AI broadside many people expected. Its core argument is that technology is never morally neutral, that machines cannot be moral agents, and that responsibility for AI cannot be outsourced — not to the technology, and not to the handful of companies building it. On Intelligent Machines #872, Leo Laporte and Jeff Jarvis were joined by Fr. Robert Ballecer, SJ — joining from Vatican City — to work through what the document says and what it deliberately leaves open.

What the Encyclical Actually Says

The document runs roughly 42,000 words under the subtitle "on safeguarding the human person in the time of artificial intelligence." Ballecer noted it was released on the 135th anniversary of Leo XIII's Rerum Novarum (1891), the landmark labor encyclical of the Industrial Revolution — and that the name "Leo" was a deliberate signal, not a coincidence.

The panel stressed that this is not a Luddite text. Roughly the first half is history, tracing Catholic social teaching from Rerum Novarum through Vatican II to Pope Francis, partly to preempt critics who might dismiss it as a sudden or radical stance. Jarvis highlighted the framing device: a contrast between the Tower of Babel — an impressive but homogenizing project "conceived without reference to God" — and the collaborative rebuilding of Jerusalem. The encyclical's position, as Jarvis paraphrased it, is that technology is neither inherently evil nor a solution in itself, but is never neutral, because it absorbs the priorities of those who devise, finance, regulate, and use it.

Why the Pope Has Standing

Laporte raised the obvious objection: isn't this a centuries-old institution commenting on 21st-century technology? Ballecer's answer was that the Pope speaks as both a head of state and a moral leader, and that the document's hardest questions don't require any particular faith to take seriously. The example he returned to: is the world better or worse if power concentrates in the hands of about a dozen people who increasingly shape what societies see, read, and decide? The panel framed this as a "second Gilded Age" — part oligarchy, part kleptocracy — and tied it to the encyclical's call for transparency in how AI systems are built and what biases they encode.

The Rejection of the "Moral Machine"

The discussion's sharpest point concerned AI alignment. The encyclical rejects the idea of an "artificial moral agent": moral judgment, it argues, requires conscience and responsibility, which originate in humans, not in machines. Jarvis noted the irony that an Anthropic co-founder spoke at the Vatican event, given how central "constitution" and alignment language is to that company's messaging — and argued that ethics cannot be baked into a tool but must inform how the tool is used. The encyclical goes further on autonomous weapons, holding that lethal or irreversible decisions must never be entrusted to AI.

The panel also flagged that the Pope defines human intelligence — self-awareness, morality, love, sacrifice, openness to transcendence — rather than AI, which the document treats as a probability engine that imitates functions without understanding them. Laporte pushed back usefully: people fake those qualities all the time, and some systems fake them convincingly, which is precisely what makes consciousness claims so hard to adjudicate.

Truth, Work, and Responsibility

Jarvis's favorite line — that truth is "a gift to be shared, not a possession to be monopolized" — he read as a challenge aimed at AI companies and at his own field of journalism and its paywalls. The panel noted the encyclical does not endorse universal basic income, arguing instead that work confers dignity, a direct inheritance from Rerum Novarum. It explicitly names transhumanism and posthumanism as ideological dangers; Jarvis connected this to the "TESCREAL" cluster of beliefs and to what he called the stench of eugenics — his framing, not the Pope's. Crucially, the document offers no statutes and no quick fixes, functioning as the start of a broad conversation rather than a regulatory blueprint, and insists that no one — not just governments or labs — is without responsibility.

Key Points

• Magnifica Humanitas is Pope Leo XIV's first encyclical, framed as safeguarding human dignity in the age of AI — not as a condemnation of the technology.
• Its thesis: technology is never neutral, taking on the character of those who build and fund it.
• The Pope rejects "moral machines" and AI alignment as substitutes for human conscience and responsibility.
• It opposes autonomous lethal weapons and warns against concentrating power in a few hands.
• It defends the dignity of work and stops short of endorsing universal basic income.
• It is deliberately non-prescriptive, calling for transparency, education, and conversation rather than specific laws.

The Bottom Line

The Intelligent Machines panel read Magnifica Humanitas as a document built for generations rather than a news cycle. Its lasting provocation is the insistence that AI's ethics can't be engineered into the product or delegated to a dozen powerful actors — they have to be argued out in public, with responsibility shared by everyone who builds, regulates, teaches with, or simply uses these systems.

Listen to the full discussion and subscribe: https://twit.tv/shows/intelligent-machines/episodes/872