What Was Surprising About Magnifica Humanitas

One: the Babel and Nehemiah structural frame

Open Magnifica Humanitas at the introduction. Paragraph 1 names a "pivotal choice: either to construct a new Tower of Babel or to build the city in which God and humanity dwell together." Paragraph 7 brings forward the second image directly from the Book of Nehemiah and the rebuilding of Jerusalem's walls. From paragraph 7 forward, these two images are the load-bearing structure of the entire document.

Pre-release commentary anticipated biblical citations. Encyclicals always have them. What was not anticipated was that two specific scenes — one from Genesis 11, one from Nehemiah 2 through 6 — would be used as the explicit organizing frame for everything that follows. Chapter 3 returns to the frame in paragraph 90: "We are called to reflect on the great construction sites of our era and ask: What are we building?" Chapter 5 returns again in paragraph 184: "On the one hand, there is the temptation of constructing the Tower of Babel, relying on power and pride. On the other hand, patience is required in order to rebuild Jerusalem piece by piece." The conclusion at paragraph 241 closes with Nehemiah as "a striking parable of our own vocation."

The choice of Nehemiah matters more than the choice of Babel. Babel was a predictable reference; commentators on AI and hubris reach for Babel routinely. Nehemiah is not predictable. The book is a minor entry in the Old Testament canon, rarely cited in social encyclicals. Pope Leo XIV chose it for specific reasons that paragraph 8 names: Nehemiah "did not impose solutions from above"; he "convened the families, assigned each of them a section of the wall to rebuild, listened to their concerns, coordinated their efforts." This is a particular vision of how to respond to a civilizational challenge — not top-down, not imposed, but distributed, subsidiarist, slow, and shared. The Nehemiah figure is the encyclical's argument for how AI governance ought to work, smuggled in as a biblical image rather than stated as a policy claim.

Once you notice the frame, you can read the rest of the document as an extended commentary on it. Every chapter is a choice point between Babel and Jerusalem. The framing also explains why the encyclical does not collapse into a checklist of AI ethics demands. The Babel and Nehemiah choice is a theological choice before it is a technological one, and that ordering shapes everything that follows.

Two: "Disarming AI"

Paragraph 110 contains the phrase that will outlive most of the document's other contributions. "I would like to employ the expression 'to disarm,' which is close to my heart. Disarming AI means freeing it from the mentality of 'armed' competition, which today is not limited simply to the military context, but is also an economic and cognitive phenomenon."

This is Pope Leo XIV's original contribution to the Catholic vocabulary on AI. Antiqua et Nova did not contain it. The 2026 World Communications Day message did not contain it. No prior Vatican document on AI has used the phrase. The framing makes the construction visible: Pope Leo XIV says explicitly that this expression "is close to my heart," which is encyclical-prose for "I am introducing this term."

The phrase does specific work that the existing vocabulary cannot. "AI safety" frames the problem as a technical risk to be mitigated. "AI alignment" frames it as a calibration problem to be solved by engineers. "AI ethics" frames it as a set of principles to be applied. "Disarming AI" is a different kind of claim. It names the AI development cycle itself as a form of armed competition — commercial, geopolitical, cognitive — and asserts that the competition itself, not just its outputs, is the problem to be addressed.

Paragraph 110 enumerates what disarming involves: "discrediting the assumption that technical power automatically confers the right to govern," "freeing technology from monopolistic control," "opening it to discussion and debate," and "restoring it to the plurality of human cultures and ways of life." Then the paragraph specifies that disarming is "not only ethical or technical" but "ecological in the deepest sense, for it concerns a new dimension of our common home." The closing line: "merely regulating it is insufficient; it must be disarmed, welcoming and accessible."

The phrase will be quoted. It will be on conference panels by August. It is the kind of original magisterial contribution that becomes shorthand for an entire reading of the document, the way "throwaway culture" became shorthand for Evangelii Gaudium. Track its uptake.

Three: the weight of Chapter 5

Count the paragraphs. Chapter 3, on the AI core, runs from paragraph 90 to 130: 41 paragraphs. Chapter 4, on truth, work, and freedom in the digital transition, runs from paragraph 131 to 181: 51 paragraphs. Chapter 5, on the culture of power and the civilization of love, runs from paragraph 182 to 228: 47 paragraphs. Chapter 5 is the second-longest substantive chapter, and it is structurally placed as the climax of the argument before the conclusion.

Pre-release commentary did not place Chapter 5 at the structural climax. The expectation was that war and lethal autonomous weapons would be addressed in a final section, picked up from Antiqua et Nova and Pope Francis at the G7, given magisterial weight, and tied off. What the document does instead is treat war as the place where the technocratic paradigm reaches its most lethal expression. The chapter is not an appendix to an AI ethics document. It is the destination toward which the AI ethics argument has been driving.

Inside Chapter 5, paragraph 197 turns to autonomous weapons systems specifically. Paragraph 198 contains what will become a quoted line: "No algorithm can make war morally acceptable." The argument here is sharper than Antiqua et Nova's version. The earlier document called lethal autonomous weapons a problem requiring careful ethical attention. Paragraph 198 of Magnifica Humanitas says that "it is not permissible to entrust lethal or otherwise irreversible decisions to artificial systems." That is a different register. It is the magisterial form of the claim.

Chapter 5 also moves beyond the autonomous weapons question into broader territory: the normalization of war (paragraphs 189 to 192), the militarization of the economy (paragraph 193), the crisis of multilateralism (paragraphs 201 to 203), and the cultural conditions that make war thinkable (paragraphs 204 to 209). The encyclical reads the AI question through the lens of war and the war question through the lens of AI. Each amplifies the other. The pre-release framing of Magnifica Humanitas as primarily an AI document missed this. It is at least as much a peace document.

This has implications for how the document will be received. Diplomatic and humanitarian audiences will find the document's center of gravity closer to their concerns than was anticipated. AI policy audiences will find that they have to engage the war chapter to understand the AI chapter. The two cannot be cleanly separated.

Four: the Augustinian "two cities and two loves"

Paragraph 130 closes Chapter 3 with a move that pre-release commentary did not anticipate. Pope Leo XIV reaches back to Augustine's City of God and quotes it directly: "Two loves have built two cities: the earthly city, the love of self even to the contempt of God; the heavenly city, the love of God even to the contempt of self." He then applies the Augustinian frame to the AI question: "As throughout history, these two loves continue to contend for dominance in our hearts today. The age of AI is no exception: the construction of Babel or the rebuilding of Jerusalem begins within each one of us."

This connects the Babel and Nehemiah frame from the introduction to Augustine's older and more theologically demanding frame. The introduction's biblical images are accessible and ecumenical. The Augustinian frame in paragraph 130 is specifically theological, specifically Catholic, and specifically about the orientation of the human heart. The encyclical is willing to operate in both registers, and the deeper register is placed at the structural pivot between the AI analysis of Chapter 3 and the social analysis of Chapter 4.

Pope Leo XIV is an Augustinian friar by training. Pre-release commentary noted this biographical detail without anticipating that it would surface so directly in the document. Paragraph 130 is the place where the biography becomes magisterial method. The choice to read AI through Augustine is not a flourish. It locates the AI question inside the longest-running theological argument the Western Church has about the orientation of human love, and it suggests that the AI question cannot be answered without taking a position on that older question.

The Augustinian frame also tightens the encyclical's claim about why AI matters morally. Paragraph 130 closes Chapter 3 by saying that "questioning this alternative path of progress and how we interpret and live it is ultimately a matter of examining our own hearts." The interior turn here is unusual in a document otherwise concerned with governance, economics, and policy. It places the responsibility for the AI question inside each reader before it is placed on institutions. This is consistent with the encyclical's larger argument that institutional reform without interior conversion will not be sufficient.

The "two cities" frame will receive less media attention than "Disarming AI." It will receive more theological attention than any other passage in the document. Track the citations from Catholic universities and theological journals over the next eighteen months.

Five: the direct address to AI developers

Paragraph 111 begins: "I wish to address a special appeal to those who develop artificial intelligence." (For the audience-scoped read of what the document asks of developers, see Magnifica Humanitas for Developers.) The address is unusual in a social encyclical. Most papal social documents are addressed in general terms to the faithful, to all people of good will, and to public authorities. Direct address to a specific professional class is rare. Laborem Exercens addressed workers. Caritas in Veritate addressed financiers in passing. Magnifica Humanitas addresses AI developers directly, by name, in their professional capacity.

The content of the appeal is striking. "In one sense, technological innovation can represent human participation in the divine act of creation. Developers, therefore, bear a particular ethical and spiritual responsibility, for every design choice reflects a vision of humanity." The framing is theological elevation, not condemnation. Developers are not described as ethical risks to be managed; they are described as participants in a creative act with the corresponding responsibility.

The paragraph continues: "Just as the creator of an artistic or literary work must consider the values it conveys, so developers are called to embed values in their projects with due seriousness: with transparency, responsibility toward affected communities and careful attention to ensuring that what is being cultivated is a genuine good." The comparison to artistic creation is consistent with the broader claim. The document is offering AI developers a vocation, not just a list of obligations.

This matters for reception. The encyclical is positioning itself as a document AI developers can read about themselves, not just a document written about them. Catholic AI developers will find an explicit invitation to read their work through a theological lens. Non-Catholic developers will find that the document is taking their professional choices seriously as moral choices, with the dignity that framing implies. The pastoral move is calculated. The encyclical is asking developers to take responsibility, and it is asking by first acknowledging that the responsibility is real and that the work is real.

The address to developers also tells us something about the document's intended audience. Magnifica Humanitas is not primarily addressed to bishops, theologians, or Catholic laity. It is addressed, in part, to the people who are actually building the systems the encyclical is concerned about. The paragraph 111 framing is the place where this intention becomes most explicit.

Six: "new forms of slavery"

Paragraphs 173 through 179 form the sharpest section of the encyclical. The argument runs as follows. The digital economy depends on hidden labor: "millions of people engaged in essential yet largely unseen activities, such as data labeling, model training and content moderation, often involving disturbing material" (paragraph 173). These workers are described as "young people, predominantly women, working under demanding conditions for minimal wages." The argument extends to the extraction of rare earth elements: "children and adolescents work in dangerous conditions, crushing the materials from which rare earth elements are extracted. The bodies of these people are scarred, injured and worn down so that computational flow may continue uninterruptedly."

Paragraph 174 names this directly: "The fight against new forms of slavery is a decisive test for the ethical discernment of AI and digital transformation." Paragraph 178 extends the analysis to data: "Even today, colonialism assumes new forms. It no longer dominates only bodies, but appropriates data, transforming personal lives into exploitable information." Health data, genetic maps, and demographic information are described as "the new rare earths of power."

The passage is unusual in two respects. First, it grounds the AI ethics argument in specific labor conditions in specific places. Most magisterial documents on AI work at a higher level of abstraction. Magnifica Humanitas names data labeling, content moderation, and cobalt mining specifically. The argument is harder to dismiss because it points at conditions that are documentable.

Second, paragraphs 175 and 176 turn the historical lens on the Church itself. "In antiquity and the Middle Ages many individuals and even ecclesiastical institutions had slaves. Already in the early modern period, the Apostolic See of Rome, responding to requests from Sovereigns, intervened several times in order to regulate and legitimize forms of subjugation." Footnote 174 names specific papal bulls: Sicut Dudum (1435), Etsi Suscepti (1442), Dum Diversas (1452), and Romanus Pontifex (1455). Paragraph 176 closes with a formal apology: "For this, in the name of the Church, I sincerely ask for pardon."

This is the move that takes the slavery argument from rhetorical to magisterial. Pope Leo XIV is saying that the Church took eighteen centuries to articulate a formal, absolute, universal condemnation of slavery; that this delay constitutes "a wound in Christian memory"; and that the present moment is a test of whether the Church has actually learned. Paragraph 177: "What we have learned must be translated into discernment and responsibility in the present. If we want to avoid the need to ask for pardon again in the future for having failed to respect the treasure of human dignity that is required by our faith, it falls to us today to denounce, clearly and firmly, trafficking in its many forms."

The structural placement of this argument matters. Paragraphs 173 to 179 sit at the close of Chapter 4, just before the chapter on war. The encyclical is building toward war by first establishing that the digital economy already contains forms of violence the Church is required to name. The progression is deliberate: AI ethics, then labor, then slavery, then war. Each step makes the next harder to escape.

What the six surprises mean together

The six moves are not unrelated. They cohere as a single editorial strategy that the encyclical uses to do something pre-release commentary did not anticipate: to refuse the AI ethics genre.

A standard AI ethics document, including the kind Magnifica Humanitas was widely expected to be, sits inside a particular conversation. It engages technical questions, governance questions, and applied ethics questions in their own terms. It defers to the AI policy vocabulary that has developed over the last decade. It tries to add a Catholic voice to that conversation without unsettling its categories.

Magnifica Humanitas does not do this. The Babel and Nehemiah frame relocates the AI question inside biblical narrative. "Disarming AI" introduces a new term that does not map cleanly onto existing AI policy vocabulary. The weight of Chapter 5 connects the AI question to war in a way the AI policy conversation has been reluctant to do. The Augustinian "two cities" reaches back to a fourth-century theological frame to read a twenty-first century technological one. The address to developers treats them as participants in a creative vocation, not as actors in a regulatory drama. The "new forms of slavery" argument grounds the AI question in specific labor conditions and connects it to the Church's own historical reckoning with slavery.

Each of these moves takes the AI question out of the AI policy register and places it somewhere else — into biblical narrative, into the Augustinian theological tradition, into the history of slavery, into the question of war. The encyclical is saying that the AI ethics conversation, as currently constructed, is too small to hold the question. The document operates on the larger terrain it thinks the question actually belongs to.

Whether this will succeed depends on reception. The AI policy world is not obligated to read the document in its own terms. Many readers will extract the policy-actionable claims (lethal autonomous weapons, data labor conditions, algorithmic accountability) and set aside the theological frame. That is a legitimate reading. But it will miss what is most distinctive about the document, which is the argument that the AI question cannot be answered without engaging frames the AI policy world has been content to leave aside.

The six surprises are the parts of the document that resist reduction to policy. They are what makes Magnifica Humanitas something other than the encyclical version of an AI ethics white paper. Whether that turns out to be an asset or a liability for the document's influence will not be clear for several years. For now, they are what the day-after analysis has to start with.

What to watch in the next ninety days

Three things to track as the reception cycle develops.

First, uptake of "Disarming AI." Does the phrase get adopted by Catholic AI ethics centers, episcopal conferences, and academic discussions? Does it reach non-Catholic AI policy conversations? The trajectory of the phrase will tell us whether Pope Leo XIV's signature contribution survives the news cycle.

Second, reception by AI developers. Paragraph 111 is an invitation. Whether developers actually accept the invitation, including those at the major AI labs, will be visible in industry conferences, internal company discussions, and the public statements of senior figures in the field. Some will engage. Most will not. The shape of the engagement will tell us whether the encyclical succeeds in making itself a document developers feel they have to read.

Third, the war framing. Will commentators read Chapter 5 as integral to the document or peripheral to it? Will the lethal autonomous weapons argument get the attention it deserves, given how sharply paragraph 198 is worded? Will the broader chapter on the normalization of war shape how the encyclical is positioned in diplomatic settings, particularly as the EU AI Act enters its enforcement phase in August 2026?

None of these will be settled in the first ninety days. But the patterns will be visible by then, and the patterns set in those ninety days will shape how Magnifica Humanitas is read for the next decade. The document is now in the world, and its reception is no longer in Pope Leo XIV's hands.