The Encyclical Magnifica Humanitas
Tuesday 6 October — Art of Intelligence · Magnifica humanitas
The table
Executive summary
Moderator introduction — Iva Kapusta
On 15 May 2026, one hundred and thirty-five years to the day after Rerum novarum, Pope Leo XIV signed Magnifica humanitas, an encyclical on safeguarding the human person in the time of artificial intelligence.
It does not ask us to stop the machines.
It asks who must account for what they decide.
In 1458 a merchant from Dubrovnik, Benedetto Cotrugli, wrote that trade exists "with the purpose of preserving humankind and with the hope of personal gain" — in that order — and left the earliest known description of the double-entry ledger, the book in which every merchant could be called to account.
Five hundred and sixty-eight years later the machine is faster than the ledger. In some rooms it is faster than the conscience.
This table does not ask whether AI is good or bad. It asks who answers — in the firm, in the intelligence service, in the army — for a decision taken at a speed no person could follow.
Questions for the table
- What does the encyclical ask of a leader on Monday morning — not in principle, but in the next decision?
- Where does speed stop being an advantage and become an abdication?
- Can technology make responsibility visible, or only make its absence harder to notice?
The case
Subtitle: Who Must Account?
The ledger
Benedetto Cotrugli defined trade as "a skill or a set of rules by which legal persons are to fairly abide by in trading affairs, with the purpose of preserving humankind and with the hope of personal gain." The order is the argument: humankind first, gain second, and rules binding both. The third book of his treatise, on the political life of the merchant, has been called the first ethics of a Renaissance profession.
In the same work he explained how a merchant keeps his books, every entry answered by a counter-entry — thirty-six years before Pacioli, half a century before Machiavelli. The ledger did not make merchants honest. It made them answerable: any line could be traced to the hand that wrote it and to the decision behind it. It was not written for princes; it was written for traders, like the ones coming to CDayZ.
The encyclical
Leo XIII answered the first industrial revolution with Rerum novarum in 1891. Leo XIV has said that he took his name for that reason, because the Church now faces "another industrial revolution" and the developments of artificial intelligence; he signed his first social encyclical on the anniversary of Leo XIII's.
Magnifica humanitas does not reject technology: "To disarm does not mean rejecting technology, but preventing it from dominating humanity" (§110). What it refuses is a decision without an answerable person. AI systems "merely imitate certain functions of human intelligence"; nor do they "have a moral conscience, since they do not judge good and evil, grasp the ultimate meaning of situations, or bear responsibility for consequences" (§99). They are not neutral either: every technical tool "embodies choices and priorities through what it measures, ignores and optimizes" (§104).
Hence the demand at the centre of the letter — accountability as "the possibility of identifying who must 'account' for decisions, justify them, monitor them, and, when necessary, challenge them and remedy any harm caused" (§105). Where data and algorithms decide on credit, hiring or access to services, decisions must be "understandable, contestable and subject to oversight, so that individuals are not reduced to mere profiles" (§164).
And every one of us is given a place: "All are given their own section of the wall: scientists and researchers, entrepreneurs and workers, educators and legislators, civil society, popular movements and faith communities" (§13). In a city whose walls stand on the UNESCO list, the image needs no explanation.
Three rooms
The encyclical names the decisions it fears most, and three of them sit at this table.
The firm. "Important and sensitive decisions — concerning employment, credit, access to public services or even a person's reputation — risk being fully delegated to automated systems" (§102). A loan scored in seconds, a thousand CVs read before breakfast, a price set separately for every customer: each is faster than the person who would have to explain it to the one it refused.
- The letter asks for more than explanations: "every introduction of automation and AI should be accompanied by verifiable measures to protect the employment, retraining and participation of workers" (§156).
- It also asks for "businesses that recognize work and dignity as measures of success" (§181).
The intelligence service. Truth, the encyclical says, "does not arise from centralized or automated control"; it "requires verification, cross-checking of sources and responsible argumentation" (§132) — the first lesson of every analyst. Disinformation "did not begin with AI, yet today it finds a powerful amplifier in AI" (§132). And the race is not only military: it is "also an economic and cognitive phenomenon" (§110).
The army. Here the line is drawn without hedging:
- "it is not permissible to entrust lethal or otherwise irreversible decisions to artificial systems" (§198);
- "the chain of responsibility must be identifiable and verifiable" (§199);
- every system used in war must "guarantee the possibility of retracing and reconstructing decision-making processes, so that accountability and blame are not collapsed into 'the machine'" (§200).
There is a clock in it too: "speed and efficiency should never be the supreme motivating force for the irreversible decisions made in the context of war" (§199).
The Vanguard tension
The moderator's own book pulls the other way — or seems to. Vanguard Leadership was written for tempo. Its Cotruglian AI Framework names two failure modes:
- over-reliance: "Delegate ethical decisions to AI (it optimizes without values)";
- under-use: "Refuse to use AI for fear of losing control (tempo disadvantage)".
It settles the boundary in one line: "AI for operations, humans for judgment." Under prudence it adds a rule the encyclical would sign: "AI requires human approval for high-stakes decisions (doesn't automate judgment)."
The encyclical accepts the cost the book warns against: "Calling for prudence, rigorous evaluation and even, at times, a slower pace in adopting AI does not mean opposing progress" (§106). So the question is not whether to be fast. It is where the line runs:
- between operations and judgment, as the book says;
- at irreversibility, as the encyclical says of war;
- or where each firm must find it for itself, because a dismissal, a refused credit and a ruined reputation are irreversible too, for the person on the other side.
The claim
Reading the encyclical in May, the moderator noted how closely its demands sit on primitives the NEO stack had been built on before the letter appeared:
- a receipt that names who acted and on whose authority;
- a record anchored so that it cannot be quietly rewritten;
- a path to replay a decision and to contest it before a Chamber.
The stack's first rule — the core attests; it never decides — is a technical sentence with a moral shape. Neither side was designed for the other.
Is that convergence a confirmation, or the temptation the encyclical itself names? "A more moral AI is not enough if that morality is determined by a few" (§107). To disarm AI, the letter says, means "discrediting the assumption that technical power automatically confers the right to govern" (§110). A ledger can show who decided. It cannot make the decision good — and that is exactly what the letter says a machine does not have (§99).
The decision
Tomorrow each person at this table will face a decision at machine speed: a credit, a candidate, a report, a target. The encyclical does not tell them what to decide. It tells them that someone must be able to account — to justify, to be challenged, to repair.
The table's task is to say, room by room, what must be in place before the machine is allowed to be faster than the person who answers for it.
Sources: Leo XIV, Encyclical Letter Magnifica humanitas on safeguarding the human person in the time of artificial intelligence, 15 May 2026 (vatican.va), paragraphs as cited; Leo XIV, address to the College of Cardinals, 10 May 2025; B. Cotrugli, The Book of the Art of Trade (1458), ed. C. Carraro and G. Favero, Palgrave Macmillan, 2017, p. 47; D. Balić and D. Papo, Synthesis Philosophica 74 (2/2022), 327–352 (after I. Martinović, on the third book as the first ethics of a Renaissance profession); N. Machiavelli, The Prince (written 1513); Vanguard Leadership, vol. 1, Section 3, "The Cotruglian AI Framework" (pp. 315–317).
Discussion Questions
- What does the encyclical ask of a business leader on Monday morning — not in principle, but in the next hiring, credit or pricing decision?
- Where in your organisation has a decision already been "fully delegated" (§102) — and who would account for it if it were challenged tomorrow?
- Your most important model measures, ignores and optimizes something (§104). Name all three — and say which of them you would defend in front of the person it refused.
- When a competitor or an adversary moves at machine speed, what does the moral timeframe (§199) cost — and who pays it?
- Who in your organisation can stop the machine — and has anyone ever heard that person say "stop"?
- Where is your line of irreversibility?
- Who is paid to tell the truth when the model and the market say otherwise?
- The NEO Turn.
The encyclical asks for an answerable person, not an answerable system. A name that has never used its veto is not yet a safeguard.
The encyclical draws it for war (§198). A firm has its own: a dismissal, a refused credit, a ruined reputation. Draw yours before the model draws it for you.
Truth "requires verification, cross-checking of sources and responsible argumentation" (§132). Intelligence services learned it the hard way; firms are learning it now.
When agents act on their own, a receipt can name who authorised them and a Chamber can hear the one who objects. Is that what the encyclical asks for — or only the part of it a machine can carry?
Moderator Note
The table follows the Archbishop's address and begins from his words, not from this text.
- Akrap: intelligence and security — truth as a common good, verification, the cognitive contest (§110, §132–134).
- Iva: the firm — work, dignity, the criteria for firm action (§148–164, §181).
- Dražen: Cotrugli, Vanguard Leadership and the NEO Turn — how much of the encyclical's demand a ledger can carry, and where it must stop.
CDayZ output: one sentence per room — what must be in place before the machine may be faster than the person who answers for it.
Your comment on this chapter
A question for the table, a disagreement, what you would have done. The case lead reads every comment; the ones the table takes up enter the chapter as questions from the room, with your name.