A Threshold, Not a Tool

There is one sentence, almost thrown in as an aside, that governs the entire reading of this encyclical. Leo XIV does not say that we are living through a period of changes; he says that we are living “a change of era” (§6). The distinction is not rhetorical. A period of changes can be managed with the old categories, simply applied more quickly; a change of era forces us to ask whether those categories still hold. But affirming the threshold is not enough: one must also say why artificial intelligence, among all the techniques humanity has ever invented, warrants so weighty a word. Why not the printing press, the steam engine, electricity, nuclear power? And why this declaration now, in this month of May 2026? It is through these two questions — the why and the now — that we must enter, for it is in answering them that the encyclical ceases to merely postulate the threshold and begins to ground it. And its answer, as we shall see, traces in negative relief the reason the proposed remedy will also be of an unexpected order: interior.

Why AI: a technique that reaches inward

All previous techniques, from flint to turbines, increased humanity’s power over the world. They extended the arm, the eye, the muscle; they acted outward. The decisive argument of the encyclical, taken up from Laudato Si’, is that AI crosses a frontier of an entirely different nature: “never has humanity had such power over itself” (§4). The shift is there, and it is vertiginous. The technique ceases to be a tool held at arm’s length and becomes a medium in which we are immersed. Leo XIV uses a striking word: AI is “an environment in which we are submerged” (§110), which leads him to say that the task is “ecological in the deepest sense.” One does not relate to an environment the way one does to a hammer; one lives within it, and it shapes us in return.

This first reason is sharpened by a second, more radical still: the nature of what AI imitates. The industrial machine multiplied force; AI, by contrast, imitates “certain functions of human intelligence” (§99). It therefore does not touch the periphery of the human, but what was long regarded as its heart — the very seat of its singularity. This is what shifts the debate from the economic to the anthropological. When a technique imitates thought, language, and decision-making, it enters into competition with that by which humanity had defined itself — and this competition takes place, precisely, inward. The encyclical thus observes that artificial imitation of care and communication can, in the user, cause the gradual loss of “the very desire to forge authentic human bonds” (§100). Here is the exact point at which technique penetrates interiority: it threatens not only our jobs or our data, it can reconfigure our desires.

Two further reasons apply, rooted not in what AI does but in what it is. First, its unpredictability: the new technologies open a horizon that is “imaginable but not yet fully foreseeable” (§4), so that their impact can only be assessed after the fact — too late. Second, its constitutive opacity, captured in the most arresting formulation of the text: current systems are “cultivated rather than constructed,” their designers creating a framework in which intelligence “grows” without their mastering the details of how it functions (§98). A technique that its own authors do not fully understand is no longer of the same order as those that came before. Humanity has fashioned a thing whose thread it no longer wholly holds.

There remains a political reason, and it is the one that lifts this from an individual challenge to a challenge of an era. All past technical revolutions were, more or less, steered by states. This one is not: the principal drivers of development are now private actors, often transnational, whose resources exceed those of many governments (§5). Technological power has acquired an unprecedented “private” character (§5). For the first time, therefore, the driver of the change of era escapes largely from political community. A technique that installs itself inside humanity, that neither is foreseen nor fully understood, and that no elected authority truly governs: this is the constellation that, in Leo XIV’s eyes, justifies the word threshold.

Why now: a second social question

If the why is anthropological, the now is historical — and it is written into the date itself. The encyclical is given on 15 May 2026, the 135th anniversary of Rerum Novarum (§3). This choice is anything but decorative. Rerum Novarum, in 1891, was Leo XIII’s response to the “new things” — rerum novarum — of the industrial revolution: the conflict between capital and labour, the condition of the working class, a social question of unprecedented magnitude. By adopting the name Leo and openly playing on the res novae of his own time (§4), Leo XIV draws an explicit parallel: AI is, for our era, what the machine was for his predecessor’s. A second social question opens, of the same magnitude as the first. The date is the argument. And the parallel is not merely analogical: where 1891 confronted the condition of labour born of the steam engine, 2026 confronts the deskilling, automated surveillance, and technological unemployment born of the algorithm (§150–151). Labour, “an essential key” to the social question (§148), remains the terrain on which the threshold is measured in concrete lives.

But the date also says why it is necessary to speak at this precise moment, and not later. The encyclical describes a “rapid phase of transition” in which, while a few dispute the future of technologies, “most people observe from afar, simply hoping for the best” (§6). It is this void of orientation that calls for speech: if we do not discern now, we will let “the succession of urgencies dictate the direction” (§6). To speak late would be to speak after the pattern has been set — and the very unpredictability of AI (§4) forbids waiting to see. Added to this is the note of hope inherited from the 2025 Jubilee (§15): it is from the experience of “pilgrims of hope” that the encyclical intends to speak, which explains its constant refusal of catastrophism. The moment is chosen because it is still open.

AI as a sign of the times

One better understands, from here, the fundamental gesture of the text. Vatican II had introduced a way of situating oneself within history: to scrutinize “the signs of the times” and interpret them in the light of the Gospel. Magnifica Humanitas explicitly claims this heritage (§22). A sign of the times is not a mere current event: it is an occurrence in which something of the human vocation is at stake, and which must therefore be discerned, not merely endured or administered.

This is exactly the status the encyclical confers on AI in affirming that it “challenges from within” the categories of social doctrine and calls for their development (§17). The expression deserves to be dwelt upon: “from within.” AI is not an object to which an already constituted doctrine would be applied, but an event that forces doctrine to grow — just as it forces humanity to redefine itself. One recognizes here the very movement of Gaudium et Spes: faith does not hover above history, it allows itself to be instructed by it. And one understands why the text immediately forbids the two false exits of “naive enthusiasm” and “unfounded fears” (§14): these are the two ways of failing to discern, that is, of suffering the threshold rather than crossing it.

What the threshold illuminates: an anthropology that was already there

The paradox of this text, and its depth, lies here: the radical novelty of AI does not produce a new anthropology — it reawakens the old one. If AI is a threshold because it reaches inward, then what it puts at risk is what the Christian tradition had always held about the interiority of the person — and which a culture of performance had rendered unreadable. The change of era acts as a photographic developer: it brings out in negative what was already there.

At the heart of it is ontological dignity (§52–53). The value of a person depends neither on their capacities nor on their productivity: it is a gift that precedes and transcends each individual. The affirmation is not new — it comes directly from Gaudium et Spes and from the human being created in the image of God. But here is what AI does: by generalizing a logic in which one is evaluated “principally according to the results one produces” (§94), it renders this gratuitous dignity suddenly counter-cultural, almost scandalous. The text calls “particularly insidious” the ideology that makes each person’s worth something they must earn (§51). What was a quiet certainty becomes, in the algorithmic age, a fighting thesis. The threshold did not invent dignity; it made it necessary to reaffirm it forcefully.

The same movement applies to limit. The encyclical holds finitude — illness, old age, failure, vulnerability — not as a defect to be corrected but as the place where humanity matures (§118–122). Nothing new in the tradition, which has always seen in weakness a place of grace. But it is transhumanism, that “archipelago” of currents dreaming of transcending the human condition (§116), that compels the faith to re-state what it already knew: to wish to eliminate all suffering would be to extinguish love and desire as well (§120). The promise of a technological “more-than-human” forces the Church to recall that an authentic “more-than-human” already exists — coming from grace, not from augmentation (§127). The threshold thus reactivates the very centre of faith: the Incarnation becomes the exact counter-proposal to transhumanism, a God who descends into vulnerable flesh rather than a human being who claims to break free of it (§232).

This is the truest meaning of the expression that the title foregrounds: the greatness of humanity. It is not measured by the power of means, but “by the care that a civilization is capable of offering” (§114). AI, in exalting power, throws into relief by contrast what the tradition had always called genuine greatness: the capacity for relationship, for gift, for recognition of the other’s face.

Truth: the first terrain on which the threshold is verified

If proof were needed that the threshold is not speculative, the encyclical finds it in truth. This is the first already visible effect, and the most revealing, because it touches humanity’s very relationship to the real. AI, the text writes, did not invent disinformation, but it offers it “a powerful amplifier” (§132): the capacity to fabricate content, images, and videos blurs the boundary between the true and the false. Yet the truth of facts, Leo XIV reminds us, is both rational — it demands verification and cross-checking — and deeply relational, built on trust (§132). This is precisely the point of attack: a technique that simulates without understanding erodes the fabric of trust through which a society holds together what it can regard as true.

The stakes are therefore not technical but civilizational. The encyclical explicitly binds truth to democracy: when the question of what is true loses its appeal in favour of what merely seems useful or efficient, democratic life weakens (§134). And the text follows the slope to its end, invoking Hannah Arendt: the ideal subjects of totalitarianism are those for whom the distinction between the true and the false no longer exists (§134). One measures here why truth is the first terrain of the threshold. It is not one file among others: it is the condition of all the others. A society that can no longer discern the true can neither govern itself, nor render justice, nor even speak to itself. And it is interiority again that is at stake — for truth, in this analysis, is not first a datum to be validated but a relationship to the real that is cultivated in the heart, where the encyclical will ultimately situate the whole of the struggle.

What the threshold truly changes: power has changed hands

If the anthropology does not change, something does change, and it is here that the encyclical is most novel relative to Vatican II. The Council still thought largely within a world of states. Magnifica Humanitas takes stock of what it could not anticipate: technological power has become “private” and transnational (§5). This displacement obliges a rewriting of subsidiarity. In its classical formulation, inherited from Pius XI, it protected intermediary bodies from absorption by the state. Leo XIV performs a bold gesture: in the digital revolution, he observes, “the highest level is no longer the state, but the major economic and technological actors” (§71). The threat is no longer statism; it is a private power that monopolizes expertise, data, and decision-making while escaping public control — hence the demands for algorithmic transparency, equitable access to data, and avenues for redress (§71).

The same pivoting affects the universal destination of goods. The principle is ancient: the goods of the earth are given to all. The encyclical extends it explicitly to immaterial goods — patents, algorithms, platforms, data (§67) — and names “new monopolies of AI” the concentration of what ought to be shared (§109). It is here that the text converges, by its own route, with a structural critique: it describes the invisible labour that sustains AI — data labelling, content moderation, mining — as a contemporary form of slavery (§173), and speaks of a colonialism of data (§178). But this structural diagnosis never arrives at a programme for seizing the means of production. It remains personalist: the end is not emancipation through collective mastery, but the dignity of the person. The novelty, therefore, does not lie in the doctrine, which remains; it lies in the location of the adversary, which shifts from the state to the private concentration of power.

War, where the thesis is put to the test

There is one terrain on which the wager of interiority is pushed to its extreme point of tension, and the encyclical does not evade it: war. This is the most painful exercise for faith, because it is here that the threshold ceases to be a metaphor. When AI enters weapons systems, it “lowers the threshold for the use of force,” removes accountability, and reduces the enemy to a statistic, the victim to “collateral damage” (§183). The same formula — lowering the threshold — recurs, but it now carries a lethal, literal meaning.

The text then poses a limit it declares non-negotiable: moral judgment cannot be reduced to calculation, and it is therefore not permissible to delegate lethal or irreversible decisions to artificial systems; no algorithm can render war morally acceptable (§198). One recognizes here, pushed to its ultimate consequence, the anthropological argument from the outset: if AI imitates intelligence without possessing consciousness (§99), then entrusting to it the decision to kill is to entrust the irreversible to something that does not know what a face is. The most arresting declaration of the chapter extends this logic: the theory of “just war,” too often invoked to justify any conflict, is now surpassed (§192).

But it is precisely here that the thesis of interior remedy is put to the test. The whole essay has shown that, since evil comes from the heart, the response passes through the conversion of the heart. War resists this elegant symmetry. An automated lethal decision cannot be undone by a change of perspective after the fact: it has already been carried out. The dead do not rise because the operator has been converted. This is why, on this terrain and on this terrain alone, the encyclical ceases to rely on interiority and formulates external, verifiable, almost technical demands: traceability of decision-making processes, the maintenance of effective human control over lethal force, a binding international framework to curb the arms race (§200). The wager of the heart reaches its limit here, and the text knows it: there are thresholds that interior conversion alone cannot guard, because their crossing is without return. War does not refute the thesis, but it compels it to acknowledge that at the extreme, interiority needs institutions if it is not to remain a pious wish — the confession of a faith that knows itself tried by the tragic.

The remedy is interior because the evil is

There remains one trait that, more than any other, marks the coherence of this text: it refuses the apocalypse. There is, in Magnifica Humanitas, no machine escaping its creator, no technological end of times. AI is never demonized — it is even a talent to be put to use, in the sense of the Gospel parable (§9). The danger is not the machine; it is Babel. And Babel is a human work: the idolatry of profit that sacrifices the weak, the uniformity that neutralizes differences (§10). Evil has a face, and it is ours.

Here all the pieces fit together. If AI is a threshold because it acts inward upon humanity — on its desires, its thinking, its relationship to limit — then the response cannot be only outward. A threat that reconfigures interiority calls for a response that passes through interiority. This is why the encyclical can conclude that “the construction of Babel or the reconstruction of Jerusalem begins in the heart of each person” (§130). And this is why the final programme — fidelity to truth, education, relationships, justice (§237–240) — proposes no new technical measure. This is not an oversight or a dodge: it is the strict consequence of the diagnosis. If the threshold is interior, crossing it is interior as well. The response to a change of era is not first a better algorithm, but a conversion of the gaze.

One may judge this displacement insufficient — and war, as we have seen, shows that at the extreme it calls for institutional safeguards that the heart alone cannot provide (§95, §106, §200). But one cannot accuse it of incoherence. It holds together from start to finish, because the remedy exactly matches the shape of the evil: interior for interior, and institutional only where the irreversible demands it.

Deciphering the threshold

What remains, at the end? That artificial intelligence is less a rupture than a revelation. It is a change of era, not because it does something unprecedented to the external world, but because it is the first technique to install itself inside humanity, to imitate its intelligence, to reconfigure its desires, and to do so under the guidance of powers that no one has elected. By pushing to the extreme the logic of performance, optimization, and control, it finally makes visible the alternative that was already traversing all of history — the two loves and two cities of Augustine that the text invokes (§130). The change of era is this: not a future arriving, but a present revealing itself.

The figure of Nehemiah, chosen as the guiding thread of the encyclical (§241), then expresses the whole programme. Confronting the city in ruins, Nehemiah neither flees into nostalgia nor surrenders to resignation: he prays, discerns, organizes, and rebuilds stone by stone, together with all. This is the exact inverse of Babel — not the tower that seeks to reach heaven without God, but the city that rebuilds itself in shared responsibility. Leo XIV asks us to enter into the “building sites of history” — laboratories, technology companies, schools, media (§241) — not to master all the tides of the world, in the words of Tolkien he cites, but to do what is in us, in the years in which we are placed (§213).

This is the strongest message of a text that announces a change of era: the threshold we are crossing is not technological; it is moral and spiritual. The question is not what AI will do to us, but what we will decide to be in its presence. And this question, the Christian tradition — from Gaudium et Spes to the Magnificat of Mary that closes the encyclical (§243) — already knew how to ask. It took a machine that thinks to oblige us to hear it again.