TL;DR — Magnifica Humanitas prompted me to reexamine the values I carry into my use of AI. Drawing on selected passages from the official English text, this essay considers technological choice, human dignity, labor, data, and governance before turning to my own reflections on faith and work. The encyclical presents the Church’s ethical and social teaching. It is not a technical specification already in force or proof of policy outcomes.

On May 15, 2026, Leo XIV signed Magnifica Humanitas. The Vatican published it on May 25, with the protection of the human person in the age of AI as its theme. The Holy See Press Office’s announcement also notes that the signing date marked the 135th anniversary of the publication of Rerum Novarum.

As someone with theological training who also uses AI extensively at work, I wanted to understand how the document approaches tools, efficiency, and human worth. But my resonance with it does not mean that the encyclical validates every argument I made before.

What follows is a reading of several passages, not a substitute for the complete document or an attempt to compress every paragraph into a declaration either for or against AI. The original can be checked on the Vatican website. The Chinese in the source article is the author’s paraphrase.

Technology Is Not Neutral, but That Does Not Mean a Tool Is Inherently Malicious

Paragraph 9 places the design, financing, regulation, and user choices that technology carries into practice within the scope of ethical judgment. It does not describe technology itself as inherently evil.

The reminder I take from it is that adopting a tool requires asking more than whether it can complete a task. We should also ask what it encourages, what it neglects, and who can adjust it. This requires concrete examination; one default setting cannot establish the intentions of every developer.

For example, the way a platform assigns orders and schedules time, or the way search results are ranked, can affect users’ choices. How large that effect is and what consequences it produces still require their own evidence. A metaphor alone cannot establish them.

The original draft used the claim that “faith is silenced by every model” as an example of this principle, but the evidence was insufficient. I no longer use that generalization here as empirical support for the encyclical; the claims in the essay on the church and AI’s rhythm should also be examined on their own terms.

Five Points I Took from the Text

One: choices cannot hide behind a tool. Paragraph 9 and paragraphs 102–106 drew my attention to the need to consider design, use, and responsibility together. This is an ethical direction, not proof that any particular product has passed.

Two: dignity is not earned through output. Paragraphs 50–51 reject measuring a person’s fundamental worth by capability, efficiency, or production. This resonates with my understanding of faith, but it is not a prediction of every labor-market outcome.

Three: work is more than a capacity number. Paragraphs 150–156 consider labor, participation, and protection during technological change. For me, discussions of AI adoption should also include workers’ circumstances rather than asking only how much time can be saved.

Four: power over data requires examination. Paragraph 178 discusses new relationships of domination through the acquisition and use of data. It prompts me to ask who decides how data is used. It does not allow me to conclude that every form of data analysis is exploitative.

Five: ethical claims still need institutional support. Paragraphs 105–106 discuss responsibility, oversight, and governance. Appeals in a document do not mean that regulations have been enacted or implementation completed. Concrete arrangements and results still have to be examined.

These five points are reading threads I selected. They are not the encyclical’s only important ideas, nor are they a list of verified social outcomes.

Do Not Turn Career Competitiveness and Human Worth into the Same Question

I have previously discussed career advice to “learn to use AI.” Such advice can remind people to pay attention to changes in working methods, but it does not necessarily contain the claim that a person who is replaced has less value. I cannot treat an inference that concerns me as something another speaker already said.

What I actually want to express is my own concern: once we become accustomed to discussing work through speed, output, and replaceability, might we begin to measure people with the same ruler without noticing?

Work skills need to be updated, and people’s circumstances need to be understood. Both can be true at the same time. To me, respecting a person should not depend on whether that person is faster than a tool or can immediately adopt a new way of working.

A tool may help people, and it may make some kinds of work more difficult. Evaluating the effect requires looking at the specific design and institutional context. Neither “innovation” nor “conservatism” should be used to excuse either side from responsibility.

The Encyclical and My Experience Can Enter into Dialogue Without Validating Each Other

I previously discussed verification in From the Royal Society to AI Collaboration: Who Checks After Generation?, and I discussed understanding and accompaniment in the essay about the church’s rhythm. Those essays contain my experience, inferences, and faith, as well as points that later required correction.

Encountering similar concerns in the encyclical makes me want to continue thinking. A similar set of concerns does not mean that the Royal Society’s motto, nineteenth-century social teaching, and today’s AI governance are all repetitions of the same historical event; nor does it mean that Church teaching has comprehensively validated my reflections on work.

I also do not use the label “highest teaching” in place of distinguishing the document’s content and nature. This is an important papal encyclical, but its original claims, my paraphrase, and my own extensions should still remain separate when it is read.

Four-panel comic: But my resonance with it does not mean that the encyclical validates every argument I made before. This requires concrete examination; one default setting cannot establish the intentions of every developer. This is an ethical direction, not proof that any particular product has passed. Work skills need to be updated, and people's circumstances need to be understood. A similar set of concerns does not mean that the Royal Society's motto, nineteenth-century social teaching, and today's AI governance are all repetitions of the same historical event;; Models will change, but I remain responsible for my adoption and judgment.

Bring the Principles Back to Each Adoption Decision

I do not want to end this reading by asking whether the church was late this time. A document’s early appearance cannot, by itself, prove that a problem has been addressed. A document can be valuable without providing every policy and implementation answer.

What I can bring back to my work are several questions that still need answers: What is this tool being adopted to accomplish? Who will be affected? What needs verification? If it is wrong, who can object, and who is responsible for correcting it?

These questions do not require everyone to accept the same theology first. They do, however, make me more aware of the beliefs I carry into my use of tools. Models will change, but I remain responsible for my adoption and judgment.

For more reflections on faith and technology, see Civilization and Humanity.