The Vatican

In the [YOU] on AI Field Guide

The cycle that opened with [YOU] on AI keeps returning to a question the technology itself cannot answer: what does it mean to become worthy of the amplifier? Magnifica Humanitas arrives as the encyclical-length version of that same question, voiced by an institution that has watched every prior amplification of human power—the printing press, industrialisation, the atomic age—and developed, across each encounter, a more precise vocabulary for what makes power dangerous. Where [YOU] on AI argues that the AI amplifier carries whatever signal you feed it, the encyclical names the two signals a civilisation can choose: the will to dominate, or the will to build together. Both texts land at the same conclusion: the urgent work of the age is not engineering but moral formation.

The Vatican's reading of concentrated power sharpens one of the cycle's central warnings. A handful of private firms now hold more data, more computational capacity, and more influence over the perceptions of billions than most governments command. The encyclical names this directly: private technological power has acquired a 'predominantly private aspect' that exceeds the capacity of democratic institutions to govern it. This is the dark twin of the democratisation narrative the cycle traces—capability disperses to the individual while control concentrates in the infrastructure owners. The Pope's demand that 'ownership of data cannot be left solely in private hands' is the theological encoding of what the cycle frames as a structural danger to the river of intelligence: a few beavers building a dam so vast that everyone downstream loses agency over the current.

The encyclical's meditation on work resonates across the cycle's treatment of professional identity and meaning. When the Church insists that work is not merely a means of generating income but a fundamental good—the primary arena in which a person cooperates with the world and becomes more fully herself—it supplies the theological grounding for the cycle's observation that the existential vacuum now spreading through knowledge-work communities is not a side-effect of AI but its most urgent human consequence. The machines automate the outputs; the encyclical insists on the irreducibility of the act. Viktor Frankl, also in the cycle's gallery, makes the same argument from psychiatry. The Vatican makes it from two thousand years of reflection on the dignity of labour.

There is a recursion at the heart of the Vatican's intervention that the cycle cannot afford to ignore. The encyclical was drafted with possible AI assistance—at least one detection service suggested as much—and presented with a co-founder of Anthropic in the room. This is not an embarrassment but the thesis in action: the Pope's whole argument is that the tool should be held in a human hand, pointed at a human purpose, with a human remaining the author of the thought. The cycle asks the same question on every page. The Vatican's presence here is the answer materialised as institution.

Origin

The encyclical did not arrive from nowhere. Its genealogy runs through a specific chain of documents, each responding to the 'new things' of its era. Rerum Novarum (1891), from Pope Leo XIII, confronted the factory and the dispossession of the industrial worker. Gaudium et Spes (1965) insisted the Church share the joys and anxieties of the modern world. John Paul II's writings on labour made work itself a theological subject. Francis's Laudato Si' (2015) applied the same logic to the environment and coined the phrase 'technocratic paradigm' for the habit of treating every problem as a matter of technical mastery. Leo XIV's choice of the name Leo was itself a signal: this encyclical stands in explicit lineage with the 1891 response to industrialisation, treating AI as the res novae—the new things—of our own century.

The immediate precursor was a shorter Vatican note, Antiqua et Nova, issued in early 2026 by two dicasteries, which established the foundational distinction the encyclical would inherit: that machine 'intelligence' is powerful information processing, while human intelligence is something wider—embodied, relational, conscientious, searching for meaning. Magnifica Humanitas is the mature, forty-two-thousand-word elaboration of that distinction, citing more than two hundred sources from Augustine and Aquinas through Hannah Arendt and Viktor Frankl, and—in one of its quietest and most precise moves—J. R. R. Tolkien's Gandalf, who counsels that it is not our part to master all the tides of the world but to do what is in us for the good of the years in which we are set.

The launch itself was an argument. Leo XIV presented the document in person alongside the theologians Anna Rowlands and Léocadie Lushombo, and Anthropic co-founder Chris Olah—the first non-cleric ever to speak at such a presentation. The Pope's chosen phrase was deliberate: 'Artificial intelligence needs to be disarmed.' An institution that thinks in centuries had decided this moment could not wait. Alongside the encyclical, the Holy See established an interdicasterial commission on AI drawing together seven dicasteries, signalling not a one-time statement but a sustained presence in the governance conversation.

Key Ideas

Babel or Jerusalem. The encyclical's opening frame refuses the standard debate—yes or no to the technology—and substitutes a choice between two construction projects. Babel is power built without reference to anything larger than the builders' ambition, a uniformity that eliminates diversity, ending in the collapse of shared language. Jerusalem under Nehemiah is distributed, relational, built nearest each family's own home, bound together by purpose rather than by hierarchy. AI systems can serve either project. The encyclical's first claim is that we are currently building Babel, and that the choice is still open.

Infinite dignity, irreducible to data. The load-bearing wall of the entire document is the assertion that every human person possesses infinite dignity, not earned, not revocable, and not expressible as a data point. Against the machine's native tendency to represent persons as vectors and rank them by output, the Church plants a flag: a person is a proper end in herself, never to be instrumentalised. The word 'infinite' is precise: infinities cannot sit on a leaderboard. A civilisation that grades human worth by usefulness will discover, when the machines become more useful, that it has wagered its entire moral foundation on a race it is losing.

The universal destination of goods, extended to data. Catholic social teaching has long held that the earth's resources were given to the whole human family, not to be enclosed by any private party. Magnifica Humanitas extends this principle to algorithms, digital platforms, technological infrastructure, and data. When these goods concentrate in a handful of hands, a new imbalance is created. 'Ownership of data cannot be left solely in private hands but must be appropriately regulated.' This is not anti-market; it is the same logic the Church once applied to land, now applied to the invisible territory of the digital commons.

Work as a fundamental good, not merely a cost. The encyclical reaffirms the primacy of human labour over any logic focused solely on productivity or finance. Work is how a person expresses freedom, cooperates with others, and becomes more fully herself; the objective dimension—what is produced—is always secondary to the subjective dimension—who the worker becomes in the producing. Against this standard, the encyclical does not hesitate to name the hidden labour force of the AI economy—the data labellers, the content moderators, the miners of rare earths—and the conditions the document calls modern forms of slavery. The smoothness of the AI experience is built on exported friction.

The subsidiarity demand. Decisions should be made at the lowest, most local level capable of making them well, and higher authorities exist to support lower ones, not to swallow them. Applied to AI, this is an argument against both unaccountable corporate power and heavy-handed state control. The document calls for transparency in algorithms, equitable access to data, avenues for recourse, and genuine participation—not the democracy of passive users but the Jerusalem model of distributed responsibility.