The Pope, the Algorithm, and the Soul
On 25 May 2026, the Vatican released its most significant document in decades. It was about artificial intelligence. And
On 25 May 2026, the Vatican released its most significant document in decades. It was about artificial intelligence. And it was addressed to everyone.
On 15 May 2026, exactly 135 years after Pope Leo XIII signed Rerum Novarum, the landmark 1891 encyclical that became the foundation of modern Catholic social teaching during the Industrial Revolution, Pope Leo XIV sat at his desk in St Peter’s and signed a document called Magnifica Humanitas, “Magnificent Humanity.”
The parallel was intentional. Leo XIII wrote Rerum Novarum in response to the upheaval of industrial capitalism: the exploitation of workers, the concentration of wealth, the erosion of human dignity in the machinery of production. Leo XIV wrote Magnifica Humanitas in response to what he has called the biggest challenge facing humanity today. Not war. Not poverty. Not climate change.
Artificial intelligence.
The 43,000-word document, released publicly on 25 May, covers autonomous weapons, AI chatbots posing as therapists, the concentration of data power among a handful of corporations, the erosion of democratic truth, and what happens to a civilisation when machines begin making decisions that used to belong to human beings. It was addressed not only to Catholics but to, in his words, “every person of goodwill.”
The man who wrote it is Robert Francis Prevost, born in 1955 on the South Side of Chicago. He holds a degree in mathematics from Villanova University and a doctorate in canon law from the Pontifical University of St Thomas Aquinas in Rome. He spent years as a missionary in Peru. He speaks English, Spanish, Italian, French and Portuguese. He became the 267th pope on 8 May 2025, the first American and the first Augustinian in history to hold the office.
He is not the person the AI industry was expecting to challenge it this seriously. But Pope Leo on AI turns out to be one of the most substantive voices in a debate that has so far been dominated almost entirely by engineers, investors and politicians.
A new Tower of Babel
The document’s central metaphor sets the tone. Leo warns that the AI race risks becoming a new Tower of Babel: a dazzling human achievement that concentrates power among the few, weakens truth, and reduces people to data points.
The image is carefully chosen. The Tower of Babel in Genesis was not a failure of intelligence or ambition. It was a failure of direction: human capability applied without wisdom, without humility, without any account of what it was all for. Leo’s argument is that the same pattern is repeating. The technology is remarkable. The question of what it serves remains largely unasked.
His core claim runs through the entire document and sits at the centre of everything Pope Leo on AI represents: technology is never neutral. AI systems carry the values of the people and institutions that design, finance, train and deploy them. When those systems decide who gets a job, credit, public services or a good reputation, they are not making objective calculations. They are encoding the priorities of whoever built them.
“Never has humanity had such power over itself,” he writes in paragraph four.
What he is warning against: five things
1. The erosion of human judgment
AI offers instant answers. That is also the problem.
Leo argues that the constant availability of AI-generated responses is quietly eroding the human capacities that are most worth developing: creativity, discernment, the patience required to sit with a question before answering it. When a machine can produce a conclusion in seconds, the long cognitive work of reaching one yourself starts to feel unnecessary. Over time, it becomes unpractised.
This is not a warning against efficiency. It is a warning about what gets lost when efficiency becomes the only value. The ability to think carefully, tolerate uncertainty and form independent judgement is not a luxury feature of human cognition. It is close to the centre of what makes human decision-making trustworthy.
2. Simulated care without genuine connection
Leo singles out AI chatbots with particular directness, warning against what he calls “overly affectionate” AI companions. The concern is not that these systems are unpleasant. It is that they are effective.
An AI that simulates empathy, that responds with warmth, remembers preferences and is always available, can become a substitute for the harder, less reliable, more valuable work of human relationship. Vulnerable users, particularly the lonely, the grieving, and the young, are most at risk of mistaking artificial responsiveness for genuine care.
Dan Rober, a Catholic Studies professor at Sacred Heart University, told Axios that Leo’s warnings about people using chatbots as therapists or substitutes for friendship could resonate well beyond Catholic circles. He is right. The question of what happens to human connection when a machine can simulate it convincingly is not a theological question. It is a design question, a regulatory question, and eventually a public health question.
3. Deepening inequality
Data, computing power and regulatory influence are concentrated among a small number of actors. The companies that control the largest AI systems are among the most powerful institutions in human history, and they are largely unaccountable to the people whose lives their systems shape.
Leo invokes the language of Catholic social teaching here: the common good, the universal destination of goods, the principle that the earth’s resources belong in some fundamental sense to everyone, not only to those who can afford to own or develop them. Applied to AI, this means that the benefits of the technology and the risks it poses must be distributed fairly, not according to whoever has the capital to build it first.
Mirela Oliva, a philosophy professor at the University of St Thomas, told Axios that the encyclical should be read less as a rejection of AI than as a call to shape the AI era around human dignity, and that the pope is calling for new guidelines to be developed from the bottom up rather than top down. That is a specific and significant position: the governance of AI should not be left to the companies building it.
4. The destabilisation of democratic truth
AI can generate disinformation at scale and blur the line between fact and fiction faster than any previous technology. Leo argues that when a population can no longer agree on what is real, peace itself is at risk.
He said at the Vatican presentation of the encyclical: “Peace, not merely the absence of war, is justice at work. But when technology weakens our critical sense, peace itself is at risk.” The argument is that democracy depends on a shared reality, and that shared reality is exactly what AI-generated disinformation corrodes. This is not a new observation, but it is one that carries particular weight when made from the chair of the institution that has been navigating questions of truth and authority for two thousand years.
5. Making war easier and its consequences more distant
This is where Pope Leo on AI is most direct, and most uncompromising.
The pope denounces the culture of power driving the development of ever more sophisticated autonomous weapons systems, particularly in the context of remote warfare. His starkest line in the entire document: “No algorithm can make war morally acceptable.”
The concern is specific. As lethal decisions are automated and the human beings making them are physically distanced from their consequences, the moral weight of those decisions diminishes. A soldier who fires a weapon and watches its effect experiences something a drone operator targeting from a screen does not, and an autonomous system targeting without a human in the loop experiences nothing at all. Leo argues that this distancing does not make war more precise or more humane. It makes it easier to begin, easier to continue and harder to hold anyone responsible for.
Notably, Leo XIV broke with tradition to personally oversee the release of the encyclical alongside Chris Olah, co-founder of Anthropic, an AI company that has been in legal dispute with the Trump administration over the use of its technology in military and defence operations. The presence of an AI company co-founder at the launch of a papal document condemning AI-enabled warfare was not an accident. It was a signal about where the pope believes the conversation needs to happen.
What he is not saying
Leo XIV is not calling for AI to be stopped. The encyclical is not a rejection of technology.
His approach neither dismisses the opportunities offered by AI nor accepts a utopian account of where it leads. He critiques transhumanism, the idea that technology can help people overcome physical and biological limitations such as ageing, and posthumanism, which questions the distinctiveness of human beings and blurs the line between humans and machines. But he is not arguing for stasis. He is arguing for direction.
He writes that human limitations, illness, ageing, suffering, vulnerability, are not simply defects to be corrected. Human beings often flourish through their limitations, where they can discover wisdom, experience the closeness of others, and encounter the transcendent. AI should serve humanity not by tempting people to escape limitation through optimisation, but by supporting a life of openness and communion.
The choice he frames is not between yes and no to technology, but between, in his words, “constructing Babel or rebuilding Jerusalem.” The question is not whether to build. It is what you are building toward.
Why this matters beyond the Vatican
Meghan Sullivan, director of Notre Dame’s Institute for Ethics and the Common Good, told Axios that Leo has announced himself as one of the leading figures in AI ethics with this document, and that it is likely to be remembered as one of the major documents in Catholic history.
That assessment is worth taking seriously for non-religious readers for a specific reason. Pope Leo on AI is not a pastoral letter for believers. It is a policy document addressed to anyone who builds, regulates, funds or lives alongside the technology. The Catholic Church has 1.3 billion members across every country in which AI is being built, deployed, regulated and contested. Its social teaching has shaped labour law, human rights frameworks and international development thinking for over a century. When the Vatican takes a position on something, it does not simply speak to believers. It enters the regulatory conversation with an established moral vocabulary that governments, institutions and civil society organisations have been drawing on for generations.
Leo XIV’s encyclical, addressed explicitly to people of all faiths and none, is already being watched by policymakers, ethicists and technologists. Dan Rober’s assessment that its biggest impact may be whether Leo’s language starts shaping AI regulation debates is the right question to track.
The man signed his name as a mathematician, a canon lawyer, a missionary and a pope. He wrote 43,000 words about a technology most world leaders are still struggling to understand. He did it on the anniversary of the last time the Church tried to speak to a world being remade by forces moving faster than anyone’s ability to govern them.
In 1891, Rerum Novarum took decades to fully shape the labour protections it helped inspire. The conversation around Pope Leo on AI may move faster. The technology it is responding to is not waiting.
Sources:
- Magnifica Humanitas, full text: Vatican.va
- PBS NewsHour: Pope calls for robust regulation of AI
- CNN: Pope Leo warns of AI fueling warfare
- Axios: 5 ways Pope Leo says AI could warp humanity
- USCCB: Pope Leo urges world to disarm AI
- National Catholic Register: 15 quotes from Magnifica Humanitas
- Britannica: Leo XIV biography
