Magnifica Humanitas:
The Encyclical on AI

THE CHURCH AND THINKING TECHNOLOGY

On 15 May 2026, marking the 135th anniversary of the Rerum Novarum, Pope Leo XIV publishes Magnifica Humanitas: the first encyclical in history dedicated to Artificial Intelligence. Often misrepresented by the media, it does not demonize the new technology but raises a fundamental question: how can it be made transparent and put at the service of all, preventing it from becoming a deadly tool in the hands of a few? This article explains what the Pope truly thinks.

( by: Antonio Maria Guerra | date: 03/06/2026 )

Magnifica Humanitas: a mathematician pope facing the AI revolution.

On 15 May 2026, just over a year after his election, Pope Leo XIV signs his first encyclical, the Magnifica Humanitas. The choice of date is no coincidence: it marks the 135th anniversary of the Rerum Novarum (*1), in which Leo XIII had taken a stand on the sweeping social transformations brought about by the Industrial Revolution. A clear gesture of historical continuity: just as his predecessor had faced the ‘new things’ (*2) of his time, the new pontiff chooses to engage with the most profound transformation of ours, Artificial Intelligence.
This may seem a task better suited to a technician than to the Vicar of God, yet it reveals much about who Robert Francis Prevost, the current leader of the Catholic Church, truly is.
An American, born in Chicago in 1955. In 1977 he earns a degree in mathematics (*3) and enters the Order of the Augustinians (*4). From the 1980s onwards, and for a long time thereafter, he serves as a missionary in Peru. This path deeply marks his formation, grounded in scientific rigour and sensitivity toward the most vulnerable.

It’s therefore no coincidence that Magnifica Humanitas, Prevost’s first ‘declaration to the world’, takes a position on a subject, AI, far from alien to him: a clear-eyed analysis, written by someone who knows numbers well but has also lived among the poor, of the enormous risks and equally great potential of this new technology (*5).

Notes:
*1: The ‘Rerum Novarum’ was published in 1891;
*2: ‘New things’ is the English translation of the Latin ‘Rerum Novarum’;
*3: At Villanova University;
*4: 1 September 1977, he entered the novitiate in Saint Louis, in the Province of Our Lady of Good Counsel of Chicago;
*5: Leo XIV, Magnifica Humanitas, Encyclical Letter on safeguarding the human person in the time of artificial intelligence, 15 May 2026.

PODCAST - Magnifica Humanitas: The Encyclical on AI

Babel and Jerusalem: two models for the development of AI.

In his encyclical Magnifica Humanitas, Leo XIV draws on two stories from the Old Testament, one mythical and one real, to show that there are opposing ways of developing Artificial Intelligence: one that leads to fracture, the other to shared construction.

The first is the story of the Tower of Babel: according to the book of Genesis, humanity gathered to build a tower so tall it would touch the sky. A grandiose project, carried forward with a single method, a single language, and the same direction. Yet precisely in this uniformity lay the danger: no room for diversity of views or the sharing of choices. The result was dispersion: languages became confused and people could no longer understand one another. The Pope sees in this story the symbol of every technology built on homogenization and efficiency as the sole criterion, one that reduces the person to a mere cog in the machine.
The second account is diametrically opposite in nature. In the 5th century BC, after decades of Babylonian exile, the Jews returned to Jerusalem to find the city in ruins, the walls demolished and the gates burned. A official named Nehemiah, having obtained permission to organize the reconstruction, divided the city walls into sections, assigning each to a different group of people, from priests to merchants and artisans, and so on. Each one responsible for their own stretch. The city’s defenses were thus reborn not through a single voice, but through the shared responsibility of all.
With these two examples, Leo XIV makes clear that the choice is not between accepting or rejecting Artificial Intelligence, but between two opposing ways of developing it: the way of Babel, vertical and homogenizing, or the way of Jerusalem, where the plurality of voices and visions becomes a resource and dialogue the common ground on which to build justice and fraternity.

The risk of a ‘partisan’ AI.

One of the themes that recurs most frequently throughout the encyclical Magnifica Humanitas is the risk of an Artificial Intelligence that is … anything but neutral. Leo XIV states this clearly: in everyday practice, technology is never ‘above the fray’, because it inevitably ‘takes sides’ with those who design it, finance it, regulate it, and use it. A fact that has consequences as concrete as they are immediate.
In this regard, it is worth recalling that, for centuries, innovation was guided primarily by States, guarantors of a more or less democratic development. Today this is no longer the case: the main drivers of ‘high tech’ are large private actors, often transnational, with resources and capacity for intervention greater than those of many governments, and for this reason difficult to govern or, at the very least, steer toward the common good.

The consequences of such a concentration of power are anything but abstract. When, for example, an algorithm decides who deserves a mortgage, who can access a service, or who is to be selected for a job, the responsibility for the decision, artificially distanced from the ‘powers that be’, dissolves behind a presumed objectivity. The exclusion of the weakest, or rather the ‘least performant’, is thus cloaked in a supposed neutrality against which it is impossible to object.
Officially, no one decides: it’s the machine that does so. Even though, on closer inspection, it always reflects the priorities, the interests… and the prejudices of those who created it (*1).

Note:
*1: Leo XIV, Magnifica Humanitas, ch. 3, nn. 104-105.

Work, truth and freedom: three goods at risk.

Undoubtedly the most ‘concrete’ chapter of the encyclical Magnifica Humanitas is the fourth, dedicated as it is to the everyday repercussions of the digital transformation on three ‘goods’ of fundamental importance for all of humanity, namely work, truth and freedom.
On the work front, Leo XIV does not merely flag the problem of growing unemployment caused by ‘technological substitution’, but warns against something more subtle: an AI that, rather than freeing workers from the most burdensome tasks, risks deskilling them, subjecting them to the pace of the machine and, on top of that, exposing them to an increasingly pervasive and invasive automated surveillance. In this regard, the Pope recalls that the ultimate aim should not be the efficiency of people, but their dignity.

As regards truth, the document denounces how digital platforms and Artificial Intelligence systems are transforming public communication into fertile ground for disinformation. The possibility of manipulating content, images and footage at ever lower costs exposes citizens to distorted narratives, undermining that trust without which democracy cannot hold.

Finally, concerning freedom, Leo XIV identifies two distinct dangers. The first is that of ‘digital dependencies’: platforms deliberately designed to capture attention and exploit users’ vulnerabilities, eroding their independence. The second is the social control made possible by the mass collection of data: when every action leaves a trace, profiling, predicting and steering behaviour becomes an enormous power, often exercised without people being fully aware of it (*1).

Note:
*1: Leo XIV, Magnifica Humanitas, ch. 4, nn. 131-171.

AI and war: when the machine strikes.

For centuries, Catholic doctrine has developed the theory of the ‘just war’: a concept elaborated by Saint Augustine and Saint Thomas Aquinas, according to which an armed conflict can be considered morally acceptable if it meets precise criteria, including the use of weapons as a last resort.
From this standpoint, in his encyclical Magnifica Humanitas, Leo XIV takes a step that several observers have described as momentous: he declares that, in the face of the advent of Artificial Intelligence in the military sphere, any ‘justification’ for war becomes effectively obsolete.

The Pope’s reasoning is clear: autonomous weapon systems, currently operational in various scenarios, make lethal decisions in fractions of a second, escaping human control. Under such conditions, criteria such as proportionality of response and the distinction between combatants and civilians become impossible to guarantee. It’s therefore inadmissible, the Pope writes, to entrust decisions of life and death, often irreversible, to AI-driven systems.
Leo XIV also denounces how these innovations are lowering the ‘psychological threshold’ of potential confrontations: when violent strife becomes more ‘accessible’ because it is the machine that fights it, the resort to force risks no longer being the last option, becoming, in effect, the first (*1).

Note:
*1: Leo XIV, Magnifica Humanitas, ch. 5, nn. 197-200.

“Disarming AI”: a strong word with a precise meaning.

Among all the potentially groundbreaking statements contained in Magnifica Humanitas, the one that generated the most media noise was undoubtedly: “Artificial Intelligence must be disarmed”. An undeniably assertive phrase, which many interpreted as a call to block or curtail technological development, triggering reactions that were, to say the least, heated.
Reactions that, on closer inspection, are largely unjustified.
“Disarming”, according to Prevost, does not in fact mean abandoning technology altogether.
The Pope uses this word rather as an explicit reference to the Church’s longstanding work toward nuclear disarmament.

Just as the use of atomic energy in warfare must not be tied to competition between powers, so too AI cannot remain prisoner to a race for technological primacy aimed exclusively at securing geopolitical or commercial advantages.
To summarize, his appeal translates into three concrete objectives. First: to break the link between technological supremacy and the right to govern, that is, to prevent those with the largest and most powerful servers from automatically setting the rules for everyone. Second: to free AI from monopolistic control, opening its mechanisms to public debate. Third: to ensure that ‘the thinking machine’ serves all of humanity… and not only those who own it (*1).

Note:
*1: Leo XIV, Magnifica Humanitas, n. 110; speech at the presentation of Magnifica Humanitas, 15 May 2026.

Digital assets: a common good.

An important passage of the encyclical Magnifica Humanitas from a legal and economic standpoint is the one in which Leo XIV extends the Catholic principle of the ‘universal destination of goods’ to the digital ecosystem.
For centuries, this principle has held that all natural resources, from land to water and air, are the common asset of the entire human community and cannot be monopolized by a few. The Pope thus takes a historic step, declaring that today, among the goods destined for all, we must also include patents, algorithms, platforms, infrastructures and data.
The implications of this act are, as one might easily grasp, enormous, and stem from the awareness that when ‘digital assets’ remain concentrated in the hands of a few private actors, without adequate forms of sharing and access, new imbalances inevitably arise. Imbalances that divide the world between those included in and those excluded from the technological revolution.

Whoever owns the data, and the tools to process it correctly, owns the future: they can, for example, shape markets, anticipate crises, decide where to direct vast investments. A power that cannot be left unchecked by those who hold it.
The Pope’s suggestion to defuse this dangerous concentration is not nationalization, but ‘shared governance’: transparency over algorithms, equitable access to data, meaningful forms of participation and public oversight.
The digital ecosystem, just like the natural one, can be protected or exploited, shared or monopolized. The choice, Prevost warns, is still in our hands (*1).

Note:
*1: Leo XIV, Magnifica Humanitas, ch. 2, nn. 65-67.

Anthropic at the Vatican: when ethics meets the algorithm.

On 25 May 2026, in the Synod Hall at the Vatican, two historic firsts took place in a single morning. For the first time in two thousand years of Catholic Church history, a Pope attended in person the presentation of one of his own encyclicals: a task that pontiffs have traditionally delegated to cardinals.
Equally unprecedented was the presence, alongside Leo XIV, of Christopher Olah, co-founder of Anthropic, the only representative of the tech world invited to the event.

His presence was anything but coincidental: his company is the only one in Silicon Valley to have sought, since its very foundation, to combine innovation and ethics (*1). Particular interest was also shown in Olah’s research into AI ‘interpretability’, which aims to understand how models think and make decisions from the inside, opening their mechanisms to external scrutiny, an approach that aligns perfectly with the call for transparency and accountability running throughout Magnifica Humanitas.
It comes as no surprise, then, that in this context Olah addressed the Pope with words that left a deep impression on those present: “You cannot expect us to set the rules for ourselves. You must give them to us from the outside, through counterweights.”. An appeal that reflects precisely the vision of the encyclical.
Concrete proof that the dialogue between values and algorithms has already begun (*2).

Notes:
*1: It’s therefore unsurprising that the company has refused to make its models available for unlimited military use, a stance that has caused no small amount of friction with the US administration.
*2: Presentation of Magnifica Humanitas, Synod Hall, Vatican, 25 May 2026. Recording available at: https://www.vatican.va

Not ‘against AI’, but for an ‘AI at the service of humanity’.

The idea that Magnifica Humanitas is a fundamentally ‘hostile’ encyclical toward technology is widespread. Leo XIV is explicit in refuting any such interpretation: his document is not an indictment of Artificial Intelligence, but rather a profound reflection on how it should be oriented. Technology, the Pope recalls, is neither inherently good nor bad: it depends entirely on the use made of it and, above all, on who governs it and to what ends.

The true dividing line is whether AI contributes to making human life ‘more human’ or not.
The Pope does not ask us to halt progress, but to embrace it responsibly. To ensure that the machine remains a tool and does not become a master. To build, as Nehemiah did in Jerusalem, brick by brick, with shared responsibility and attention to the most vulnerable. A construction site open to all: researchers, entrepreneurs, legislators, educators, citizens.
The future of Artificial Intelligence, Prevost warns, is not written: it depends on the choices we make today (*1).

Note:
*1: Leo XIV, Magnifica Humanitas, nn. 85, 129.

Encyclical "Magnifica Humanitas".

It is possible to read the full text of the Encyclical Magnifica Humanitas by following THIS link.”

The images on this page were created using generative Artificial Intelligence tools.

Magnifica Humanitas:The Encyclical on AI