Rome's Answer to the Algorithm

What The Encyclical Attempts

There is a genre problem inherent to the papal encyclical that no pope has yet managed to solve and that Leo XIV has not solved either, though he comes closer than most. The genre demands comprehensiveness, it is, by institutional definition, addressed to the entire human family and must therefore speak to conditions ranging from the subsistence farmer in the Sahel to the artificial intelligence researcher in Menlo Park. This universal ambition tends to produce a universal blandness, a prose of such cultivated generality that it cannot offend anyone because it cannot quite pin down anyone either. Magnifica Humanitas periodically succumbs to this temptation. But periodically, and with growing frequency as the document develops, it does not, and when it does not, what emerges is something genuinely surprising from an institution not widely credited with the capacity for surprise.

The encyclical is structured across five substantial chapters. The first surveys the history of Catholic Social Teaching from Leo XIII's Rerum Novarum of 1891, whose anniversary this encyclical explicitly honours, through the pontificates of Pius XI, Pius XII, John XXIII, Paul VI, John Paul II, Benedict XVI, and Francis. The second articulates the foundational principles of that teaching: human dignity, the common good, the universal destination of goods, subsidiarity, solidarity. The third, the argumentative centrepiece, confronts artificial intelligence directly, invoking Augustine's two cities as its primary conceptual architecture. The fourth addresses the specific threats to truth, work, and freedom posed by digital transformation. The fifth attempts a constructive vision under the banner of what Paul VI called "the civilisation of love."

The governing metaphor, deployed with genuine rhetorical intelligence, is the biblical binary of Babel and Nehemiah. Babel: the grandiose, self-referential technological project that achieves uniformity at the cost of communion, erects towers toward heaven without God, and ends in the cacophony of mutual incomprehension. Nehemiah: the distributed, patient, consultative rebuilding of Jerusalem after exile, in which each family is assigned its own section of wall and the whole is restored through shared responsibility rather than centralised command. That these two images capture something essential about the difference between, say, a social media platform optimised for engagement metrics and a community garden is not accidental. The Pope knows exactly what he is doing with these metaphors, and he does it well.

On the Prose and Its Rhetorical Character

Encyclical prose is a strange and largely unloved literary form. It cannot be lyrical, the Pope is not Blake, even when he gestures toward it, and it cannot be purely analytical, because it must also be devotional. The result is typically something that resembles a very learned sermon delivered to a very large and very heterogeneous congregation: at its best, elevated and clear; at its worst, sonorously empty. Magnifica Humanitas achieves the former more reliably than the latter, though not consistently.

The document's rhetorical strength lies in its diagnostic passages. When Leo XIV writes that the "Babel syndrome" consists in "the idolatry of profit that sacrifices the weak, a uniformity that neutralises differences, and the pretence that a single language, even a digital one, can translate everything, including the mystery of the person, into data and performance," this is not vague gesturing. It is a precise description of the philosophical commitments embedded in the large language model paradigm, in algorithmic content moderation, and in the reduction of the citizen to a demographic. It has the quality of a well-ground lens: you look through it and suddenly things you had seen blurrily come into focus.

The weakness, when it comes, is structural rather than individual. The long historical survey of Chapter One, tracking Catholic Social Teaching from Leo XIII through Francis, is thorough but relentless. One feels the weight of institutional inheritance pressing upon the new pope's own thought. He quotes predecessors not because they resolve a problem but because he is showing continuity, demonstrating that what he says has been said before. This is ecclesiastically necessary and intellectually constraining in roughly equal measure. Nietzsche did not feel obliged to establish his continuity with Kant before criticising modernity. The Pope does. One respects the impulse; one sometimes wishes he had been freer.

The prose brightens considerably when Leo XIV speaks in his own voice, particularly in the extended analysis of the "technocratic paradigm" in Chapter Three and the treatment of truth as a common good in Chapter Four. Here one hears a mind actually thinking rather than cataloguing, and the difference in temperature is considerable. The passage in which he warns that AI-generated disinformation threatens not only individual judgements but the very conditions under which a shared world remains possible, the epistemic commons upon which democracy depends, is among the most genuinely penetrating things written about artificial intelligence by any public figure in any idiom this year. That such a passage should appear in a papal encyclical is, to put it gently, unexpected.

When Leo Meets Leo

The choice to anchor Magnifica Humanitas in the 135th anniversary of Rerum Novarum is not sentimentality. It is a claim about structural analogy that deserves to be taken seriously. When Leo XIII published that encyclical in 1891, industrial capitalism had produced conditions: child labour, the destruction of craft guilds, the immiseration of the urban proletariat, the concentration of productive capital in private hands, that existing moral frameworks were not equipped to address. The Church stepped into that vacuum not with nostalgia for the agrarian order that industrialism had destroyed, but with a new set of analytical categories: the dignity of labour, the living wage, the right of workers to associate, the obligations of property. It engaged the new thing on its own terms and in so doing contributed to the intellectual foundation from which the welfare state, Christian democracy, and the trade union movement would eventually emerge.

The analogical claim in Magnifica Humanitas is that artificial intelligence has created a comparable structural situation. Power over the conditions of human life is being concentrated in private transnational entities, technology companies, whose resources and global reach dwarf those of many nation-states, and which operate according to logics (engagement maximisation, prediction markets, labour automation) that are not oriented toward human dignity as a first principle. The regulatory capacity of the state, historically the guarantor of the common good, has been outpaced. The philosophical categories available to both state and citizen are insufficient to the challenge. Into this gap, Leo XIV is attempting to insert the Catholic Social Tradition as a resource.

Whether the analogy holds perfectly is open to dispute. Industrial capitalism produced visible, measurable, politically articulable suffering, bodies in mines, children in mills, that was amenable to legislative remedy. The harms of artificial intelligence are more diffuse, more epistemological, more difficult to narrate: the slow erosion of epistemic autonomy, the precariousness introduced into middle-class knowledge work, the concentration of data as a new form of capital. The encyclical is better at diagnosing these diffuse harms than at prescribing remedies, which is perhaps understandable given that the prescriptions are still being worked out by everyone.

One should also note what the historical context of 2026 adds. The encyclical is addressed to a world in which the geopolitical competition over AI infrastructure, between the United States and China primarily, but with European regulatory ambitions complicating the picture, is already reshaping the international order. The Pope does not engage this geopolitical dimension as directly as one might wish, preferring the moral register to the strategic. This is partly a function of the form, encyclicals do not do geopolitics, and partly, one suspects, a function of the institutional caution required of a new pontiff in his first major document.

The Papal Philosophical Apparatus

The philosophical framework of Magnifica Humanitas is Thomist-Augustinian in its foundations, enriched by the phenomenological personalism of John Paul II and the integral ecology of Francis. This is not a criticism; it is, in fact, one of the encyclical's distinctive strengths in the contemporary context, precisely because it is so thoroughly unfashionable in the quarters it most wishes to address.

The document's central philosophical wager is that the human person cannot be adequately understood in purely functional terms, as a node in a network, a producer of data, a collection of preferences to be predicted and monetised, because personhood is constituted not by what one does or produces but by what one is: a being made in the image and likeness of God, called to relationship, possessed of ontological dignity that precedes and transcends every contingent circumstance. This is the doctrine the encyclical calls "ontological dignity," as distinct from functional, moral, social, or existential dignity, and it is doing real philosophical work. It is, in effect, a counter-ontology to the utilitarian-computational picture of the person that underlies most AI development: the person as an entity whose value is indexed to their efficiency, productivity, or measurable output.

The encyclical is at its most philosophically interesting when it extends the principle of the "universal destination of goods," the classical teaching that creation's resources belong to the whole human family and not to those who happen to hold title to them, to the digital domain. Algorithms, patents, data sets, digital platforms, and AI infrastructure are, on this analysis, goods whose concentration in the hands of a few contradicts their proper social function. This is a genuinely radical claim, one that has implications no pope has yet been willing to follow to their logical conclusion. It implies, at minimum, that the business model of Google, Facebook, and OpenAI is structurally incompatible with Catholic Social Teaching as now articulated. Leo XIV does not say this plainly. But it follows from what he does say, and anyone reading carefully will notice.

The Augustinian architecture, the "two cities" of the encyclical's third chapter, the City of God and the City of Man, with their corresponding loves (love of God unto contempt of self versus love of self unto contempt of God), is deployed to frame the choice between the Babel model of AI development and the Nehemiah model. This is both illuminating and limiting. Illuminating because Augustine understood better than almost anyone in the Western tradition how civilisations are shaped by their ultimate loves, their fundamental orientations, their latent theologies. A civilisation organised around the love of unlimited growth, frictionless scalability, and the elimination of human limitation will produce different social arrangements than one organised around care, vulnerability, and solidarity. Augustine would have recognised this dynamic immediately. Limiting because the "two cities" framework tends toward a certain Manichean simplicity: goods here, evils there, choose your side. The actual politics of AI governance is considerably messier than Augustine's theological binary admits.

Where Leo XIV's Argument Fails

Intellectual honesty requires that we note the blind spots, and they are real. The most significant is methodological. Magnifica Humanitas is a document produced by a hierarchical institution about the dangers of hierarchical institutions. The Church, which is not governed by synodal democracy despite the gestures toward synodality, which has historically been one of the world's most formidable concentrators of cultural, epistemic, and at times material power, here presents itself as the champion of distributed authority, local community, and protection against concentrated dominance. The principle of subsidiarity, that decisions should be made at the lowest competent level, is invoked to critique Big Tech. The Church does not apply it to itself.

This is not mere hypocrisy; institutions always speak better than they live, and the Church is no different. But it creates a structural weakness in the argument that a hostile reader will immediately exploit: by what right does an institution that claims magisterial authority over the consciences of 1.4 billion people lecture transnational technology companies about their concentration of power? The encyclical does not address this question. It would have been both braver and more persuasive if it had.

The second blind spot is the treatment of work. John Paul II's Laborem Exercens is cited approvingly for its claim that work is not merely an economic activity but a fundamental expression of human dignity, a site of creativity and participation in creation. The encyclical extends this to argue that AI-driven automation must be evaluated not merely for efficiency but for its impact on the dignity of workers. This is correct and important. But the document does not adequately grapple with the possibility, increasingly argued by serious economists, including those sympathetic to the Church's social vision, that some forms of liberating automation might actually enhance human dignity by freeing people from degrading or dangerous labour. The encyclical's treatment of work tends toward a romanticism about certain kinds of labour that is not fully warranted, and which may lead to policy prescriptions (slow automation, preserve existing jobs) that serve neither workers nor the poor of the Global South.

Third, and most philosophically significant: the encyclical assumes throughout that the principal danger of artificial intelligence is its tendency to dehumanise, to reduce the person to data, preference, or function. This is a real danger and the encyclical articulates it well. But there is another danger that receives less attention: that AI systems might turn out to be genuine moral agents, or to have something analogous to inner experience, in ways that would require us to extend the circle of moral concern outward rather than merely protecting the existing human circle. The encyclical's philosophical apparatus, Thomist, personalist, centred on the human person as the locus of infinite ontological dignity, is constitutively unable to address this possibility. It assumes the answer to the question of machine consciousness before the question has been properly posed. For a document that wishes to be taken seriously in the rooms where these questions are actually being worked out, this is a significant lacuna.

Finally: the encyclical's vision of "the civilisation of love" as the constructive alternative to technocratic Babel is beautiful and, in a theological register, genuinely moving. But it is thin on institutional mechanism. How exactly does a "civilisation of love" regulate transformer models? Who enforces the principle of the universal destination of goods against a Delaware-incorporated company with servers in seventeen countries? The encyclical gestures toward international institutions and regulatory frameworks but cannot, because that is not what encyclicals are for, specify what those frameworks should look like. One does not blame the Pope for not writing an AI governance white paper. One simply notes that the distance between the moral vision and the institutional means of achieving it is very large, and the encyclical does not help us cross it.

What the Encyclical Genuinely Achieves

Let us be clear about what is real here, because there is a great deal that is real. Magnifica Humanitas is the first major institutional document from any global body that attempts to think through the AI question not merely as a policy problem or a safety risk but as a civilisational challenge: a challenge to the categories through which we understand what a person is, what work means, what truth requires, and what communities owe to their most vulnerable members. In this, it exceeds anything produced by governments, think tanks, or the technology industry itself, which is perhaps not surprising but is nonetheless striking.

The extension of the "universal destination of goods" to digital assets and data is, as noted above, philosophically decisive and practically radical. It has implications that will take years to work out, and it places the Church squarely in opposition to the prevailing intellectual framework of the technology industry, which treats data as legitimately private property and treats the accumulation of it as a natural and just consequence of market competition. Leo XIV does not accept this. He provides the theological ground for claiming that the concentration of AI capability is not merely a competition concern but a justice concern: that the excluded are not merely inconvenienced but wronged.

The treatment of truth in Chapter Four is the most intellectually rigorous section of the encyclical and deserves extended discussion in its own right. The Pope argues that truth is a common good: not merely a value or a preference or a claim, but a good in the classical sense, something whose availability is a condition of the possibility of human flourishing in common. The privatisation of truth, through algorithmic manipulation of what people see, through AI-generated disinformation that makes the epistemic commons toxic, through the fragmentation of a shared world into incompatible information ecosystems, is therefore not merely a problem for individual epistemology but a structural injustice. This is a contribution to political philosophy that deserves to be read alongside Habermas's account of communicative rationality and Berlin's analysis of epistemic liberty. That it arrives in the form of an encyclical rather than an academic monograph should not diminish its seriousness.

One also notes with admiration the document's insistence that weakness and limitation are not errors to be corrected but constitutive features of human being that deserve honour rather than elimination. This is directly counter to the transhumanist strand of AI enthusiasm, the longevity research, the neural interface development, the pursuit of what the encyclical sardonically calls "unlimited upgrades," and it represents a philosophically rich position with roots in Stoic thought, Pauline theology ("power is made perfect in weakness"), and the phenomenological tradition's account of embodied finitude. Against the dream of the optimised human, Leo XIV offers something altogether more dignified: the image of Nehemiah examining the ruins in silence before beginning to rebuild.

The Document That Still Believes in Souls

There is a passage early in Magnifica Humanitas that I have been unable to stop thinking about since I read it. Examining the story of Nehemiah, Leo XIV notes that before doing anything at all, before commissioning the project or convening the people or assigning sections of wall, Nehemiah "examined the destroyed areas in silence." He did not impose solutions from above. He looked first, quietly, at the ruins. This is offered as a model for how the Church should approach artificial intelligence. It strikes me as something more: a model for how any serious mind should approach any civilisational transformation. Before the advocacy, before the frameworks, before the regulatory proposals and the ethics boards and the impact assessments, there must be a moment of honest looking, a willingness to see what has actually been broken and what it would actually take to rebuild it.

What Magnifica Humanitas does best, in the end, is that silent examination. It looks at the ruins, the epistemic commons poisoned by disinformation, the dignity of workers eroded by automation, the concentration of AI power in entities accountable to no one but their shareholders, the reduction of persons to prediction targets, and it names them as ruins. It does not pretend they are renovation. This may seem a modest achievement; it is, in the present climate of relentless technological optimism, a rare one.

But here is what the encyclical cannot quite bring itself to acknowledge: Nehemiah succeeded because the people of Jerusalem remembered what Jerusalem was for. They had a shared memory of what had been lost and a shared conviction about what needed to be restored. The civilisation now navigating the age of artificial intelligence does not have this. It does not agree on what has been lost, what constitutes a ruin, or what a flourishing human community actually looks like. The Church's tradition offers an answer to these questions. Whether that answer can speak across the fractures of a secular, pluralist, technologically-mediated world — whether the language of dignity, of the common good, of love as a civilisational principle, can be heard in the server rooms and the venture capital offices and the parliamentary committees where the actual decisions are being made — this is the question the encyclical raises and that history alone will answer. Leo XIV has laid his section of the wall. The rest of us must decide what we are building, and for whom.