Pope Leo XIV Tells the World: Artificial Intelligence Must Be Disarmed

The first American pope arrived at the Vatican press hall on Monday morning carrying a slim, leather-bound book and a message that no one in the room could mistake for diplomatic throat-clearing. Artificial intelligence, Pope Leo XIV declared, must be disarmed. The line, delivered in measured Italian beneath the cameras and the cardinals in scarlet, was the headline of an 82-page document destined to ripple through laboratories in San Francisco, defense ministries in Washington and Beijing, and parliaments still struggling to write their first serious AI laws.

The encyclical, titled Magnifica Humanitas, or Magnificent Humanity, was released on May 25, 2026. It is the inaugural teaching of the former Cardinal Robert Prevost, the Chicago-born Augustinian who took the throne of St. Peter a year ago and chose his name in tribute to Pope Leo XIII, the nineteenth-century pontiff whose Rerum Novarum reckoned with the upheavals of industrial capitalism. The symbolism is deliberate. Where his predecessor confronted steam engines and sweatshops, Leo XIV has decided to confront algorithms and data centers.

A Single Word That Set the Tone

One sentence from the document has already begun circulating in policy circles around the world. "Artificial Intelligence now demands to be disarmed," the pope writes, "freed from logics that turn it into an instrument of domination, exclusion and death." The verb is not accidental. In Vatican language, disarmament is the work of moral courage, the act of refusing to treat a weapon as inevitable. By applying the term to AI, Leo has elevated the ethics of code to the same theological plane as nuclear arms control.

The pope did not appear alone. Flanking him at the lectern were senior cardinals, theologians from the Pontifical Academy for Life, and, in a choice that startled some observers, Christopher Olah, a co-founder of the American AI company Anthropic. The optics were striking. The Vatican was simultaneously condemning the industry's excesses and inviting one of its architects into the conversation, a gesture that Leo allies described as engagement rather than endorsement.

Autonomous Weapons and the End of Just War

The most pointed sections of Magnifica Humanitas address warfare. Leo writes that it is "not permissible" to entrust lethal decisions to autonomous systems, and he goes further than any modern pontiff in arguing that the Catholic Church's centuries-old "just war" tradition has been rendered obsolete by the speed and scale of algorithmic killing. "There exists no algorithm capable of making war morally acceptable," the document states.

That passage lands at a moment when militaries from the United States to Russia to Israel are racing to integrate machine learning into targeting systems, drone swarms and battlefield decision support. The Vatican is now formally on record arguing that any lethal action delegated to software crosses a moral threshold. Diplomats at the United Nations, where talks on regulating autonomous weapons have stalled for years, told reporters they expect Magnifica Humanitas to be cited repeatedly in the next round of negotiations in Geneva.

The Hidden Labor Behind the Machines

If the weapons chapters are the most quotable, the labor sections may prove the most consequential. Leo lingers on what he calls the invisible workforce that powers modern AI: the data labelers in low-wage countries, the miners extracting cobalt and rare earths for chips and batteries, the writers and artists whose work has been ingested without consent. He warns of "new forms of slavery" tied to the AI economy and tells policymakers they have a duty to safeguard workers and prevent child exploitation in the global supply chain.

The encyclical also speaks directly to office workers and creative professionals watching their jobs reshaped by generative tools. Leo argues that productivity gains cannot be allowed to flow only to the owners of the models, and he calls for "robust legal frameworks, independent oversight, informed users and a political system that does not abdicate its responsibility." The line is a rebuke to the laissez-faire stance favored by the Trump administration, which has spent the past year dismantling federal AI safety rules in the name of competitiveness with China.

Surveillance, Power, and the Few Who Control the Code

Throughout the document, Leo returns to a single anxiety: that the future of humanity is being quietly negotiated by a small circle of executives and engineers in a handful of corporations. He warns against "competitive races for ever more powerful algorithms and larger datasets," framing them as expressions of geopolitical and commercial appetite rather than human progress. Data, he writes, cannot remain exclusively in private hands.

Cardinal Michael Czerny, who helped shape the text, was careful at the press conference to argue that the encyclical is not anti-technology. "This document is about the human condition during the time of AI," he said, calling AI "a genuine human achievement" while insisting that the companies behind it cannot "renounce responsibility" for what they unleash. Olah, the Anthropic co-founder, struck a chastened note, acknowledging that his industry operates "inside a set of incentives and constraints that can sometimes conflict with doing the right thing" and welcoming pressure from outside the research community.

From Francis to Leo: A Sharper Edge

Pope Francis spent the final years of his pontificate raising alarms about AI, addressing the Group of Seven leaders in 2024 and publishing a shorter Vatican note on artificial intelligence and human dignity. Leo's encyclical builds on that foundation but sharpens it considerably. Francis tended to appeal to the conscience of individuals and to call for dialogue. Leo names structural problems, demands binding regulation, and explicitly invokes the language of disarmament, a register reserved in Catholic teaching for the gravest threats to human life.

The shift reflects more than personality. In the year since his election, Leo has shown a particular interest in technology's downstream effects on the global poor, a preoccupation rooted in his decades as a missionary in Peru. Aides say he drafted significant portions of the document by hand and circulated it among bishops on three continents before finalizing the text.

What Happens Now

Encyclicals do not have the force of law, but they shape Catholic teaching for generations and often seep into wider public debate. Within hours of the release, European officials in Brussels signaled that Magnifica Humanitas would be referenced in upcoming revisions to the EU's AI Act. Lawmakers in Washington who have struggled to pass even modest AI bills found themselves quoting the pope on cable news. In Rome, theologians began the slower work of unpacking the document's implications for everything from education to end-of-life care.

Leo closed his remarks at the Vatican with a line that read less like a warning than a wager. The pressing duty of this generation, he said, is "to remain profoundly human." Whether the engineers, generals and legislators listening in Silicon Valley and beyond accept that wager will define the decade now beginning, and possibly the century.