AI and Education

Education as formation: the claim that changes everything

Before the technology, the framework. The Catholic understanding of education rests on a claim that sounds simple and turns out to be decisive: education is the formation of a person, not the delivery of content.

On this understanding, the point of a writing assignment is not the essay. The point is the student who is formed by writing it, the one who learns to hold a thought across paragraphs, to test a claim against evidence, to discover what they actually think by the labor of putting it into words. The essay is the residue of the formation, not its purpose. The same is true of a mathematics problem, a translation exercise, a lab report, a historical analysis. The work produces a person who can do that kind of thinking, and the person is the point.

This is why the Catholic framework does not evaluate AI in education primarily by asking whether it improves test scores or saves time. It asks whether the use of AI forms the student or lets the student avoid being formed. An AI that helps a student understand a concept they then reason with has served formation. An AI that produces the reasoning the student was supposed to do has replaced formation with its appearance.

Magnifica Humanitas states the principle directly. Education, Pope Leo XIV writes, "is a long journey requiring patience and therefore needs time for development and for engagement with reality beyond appearances." He adds the claim that gives the whole analysis its force: every technology "shapes those who use it." The student who learns with AI is not the same student who learns without it, and the difference is not neutral. The tools of learning form the learner.

This claim has deep roots. The Catholic intellectual tradition, from the medieval universities through John Henry Newman's The Idea of a University to the present, has consistently understood education as the cultivation of intellectual and moral virtues rather than the accumulation of facts. Newman's account of the "philosophical habit of mind" that a true education forms is an account of a capacity, something the student becomes, not something the student has. AI that bypasses the cultivation of that capacity is, in this tradition, not education at all, however much information it conveys.

The deskilling concern: what is actually at risk

The sharpest application of the formation framework is the deskilling concern, and it is worth being precise about what deskilling means.

Deskilling is the loss of a capacity through disuse. It is not the same as ignorance, which is the simple absence of knowledge. A deskilled person once could do something, or could have learned to, and now cannot, because the capacity was never built or has atrophied. The concern about AI in education is that students who rely on AI to perform intellectual tasks never build the capacities those tasks are designed to develop, and that the loss is invisible until the moment the capacity is needed and found missing.

Antiqua et Nova named this in 2025 at paragraph 82, warning that AI programs that "merely provide answers instead of prompting students to arrive at answers themselves or write text for themselves" undermine the development of critical thinking. Magnifica Humanitas develops the concern further and locates it more precisely. The danger, Pope Leo XIV writes, is that "the speed and ease with which answers or summaries can be obtained risk extinguishing the desire to ask questions, which is a process that bears fruit only over time."

Notice where the encyclical locates the risk. It is not primarily that students will fail to learn particular facts. It is that they will lose the desire to ask questions at all. The appetite for inquiry, the willingness to sit with a problem, the tolerance for the discomfort of not yet knowing: these are the dispositions that make a person educable across a lifetime, and they are precisely what the frictionless availability of answers erodes. A student who can get any answer instantly has no occasion to develop the patience that real understanding requires.

The encyclical reaches for an ancient witness to make the point. Pope Leo XIV cites Plato's Seventh Letter, from 353 B.C., on the way the deepest understanding comes not from receiving information but from long engagement with a question. The reference is deliberate. The concern about technology short-circuiting genuine understanding is older than computing; Plato worried about writing itself in the Phaedrus. What is new is the scale and the seductiveness. Pope Leo XIV names it as "the promise of the perfect machine, that subtle temptation which renders human thought seemingly superfluous precisely when it is most needed."

This is the heart of the Catholic concern about AI in education, and it applies up and down the system, from the child learning to read to the doctoral student learning to construct an argument. The skill at risk is not any particular technique. It is the capacity for sustained thought itself.

For students: learning when not to use it

The most useful thing the Catholic framework offers a student is a reframing of the central question. The question is not "what can I get away with?" It is "what am I here to become, and is this use of AI helping or preventing it?"

Magnifica Humanitas puts the decisive skill at paragraph 140: students must learn "to decide when and for what purpose it ought not to be used." This is a more demanding standard than any school's acceptable-use policy, and a more useful one. It asks the student to internalize the judgment rather than follow a rule, which is itself a formation.

The practical distinction is between AI that supports the student's own intellectual work and AI that replaces it. On the supporting side: using AI to explain a concept you are struggling with, to generate practice problems, to check your work after you have done it, to act as a tutor that asks you questions rather than answering them, to get unstuck when you have genuinely tried. On the replacing side: having AI write the essay, solve the problem, compose the argument, or produce the analysis that the assignment was designed to make you produce. The first kind of use builds the capacity; the second kind borrows the appearance of the capacity while leaving the student without it.

The hard cases are real and worth naming honestly. Using AI to brainstorm before writing, to improve a draft you wrote yourself, to summarize a long reading you will then engage with: these sit in the middle, and the framework does not resolve them mechanically. The question to ask in the middle cases is whether the AI is doing the part of the work that the assignment exists to develop. If the point of the assignment is to learn to summarize, having AI summarize defeats it; if the point is to engage with the content and the summary is incidental, AI assistance may be fine. The student who asks this question honestly is already being formed by asking it.

One caution the framework presses, particularly for students who feel behind or anxious: the relief AI offers is real and the cost is deferred and invisible. The essay submitted is done; the capacity not built does not announce its absence until later, often much later, when the thinking is needed and the practice was never put in. Catholic teaching's emphasis on the long view of formation is, in part, a warning against trading a durable capacity for a momentary relief.

For teachers: formation requires the teacher's own engagement

The Catholic framework asks teachers to consider AI's effect on their own work as well as their students'.

Magnifica Humanitas calls for teachers to receive ongoing professional formation so they can help students use AI "responsibly, critically, and creatively." This is a real institutional demand: teachers cannot form students in the wise use of a technology they have not themselves thought through. The encyclical does not ask teachers to ban AI or to master every tool; it asks them to develop the judgment they are then meant to cultivate in students.

There is a parallel here to the question the Catholic framework raises about homilies and pastoral writing. The teacher's engagement with the student's work is part of what the relationship is for. A teacher who uses AI to grade, to generate all feedback, to write reports, and to produce every lesson is mediating the relationship through the machine in a way that can hollow it out. The student's work is an occasion for the teacher to know the student, and feedback is an occasion for the teacher to form the student; AI that fully intermediates both removes the teacher from the formation the teacher is there to do. This does not mean teachers cannot use AI to lighten genuine administrative burden. It means the relational and formative core of teaching is not the part to automate.

On the assessment side, teachers and schools are increasingly deploying AI plagiarism-detection tools, and the Catholic framework presses a specific caution here. These tools have a documented bias against non-native English speakers, whose writing is more often falsely flagged as AI-generated. A false accusation of cheating is not a minor administrative event; it can damage a student seriously, and it falls hardest on students who are already more vulnerable. Catholic teaching's preferential concern for the vulnerable applies directly. Schools using these tools should require meaningful human review of any flag, should be transparent with students about what is being used, and should be willing to abandon tools that produce biased outcomes.

The constructive opportunity is real too. Teachers who understand the formation framework can redesign assignments so that AI assistance and genuine learning are not at odds: assessment that surfaces the student's reasoning process rather than only the product, in-class work that builds the capacities homework can no longer reliably develop, assignments that use AI deliberately as an object of study rather than pretending it does not exist. The teachers who navigate this best are not the ones who police AI most strictly but the ones who understand what their assignments are actually for.

For parents: attention, age, and the formation of the young

The encyclical devotes notable attention to parents and the family, and the guidance is more pointed than much of the document.

Magnifica Humanitas warns at paragraph 141 against unsupervised access to personal mobile devices "at too early an age," stating that this can have "profoundly negative psychological effects, in addition to the moral dangers." The encyclical does not name a specific age, which is wise; the right threshold depends on the child, the supervision, and the use. But it places clear responsibility on parents for the supervision and formation of children in the digital environment, and it does not treat that environment as neutral. Paragraph 142 extends responsibility to service providers, calling for "specific protections against all forms of online sexual exploitation and violence."

The deeper parental concern the framework raises is about attention and its formation. Magnifica Humanitas warns that "the pervasiveness of digital media fosters a culture of immediacy and hyper-stimulation, which gives rise to fatigue, boredom, and apathy concerning the effort required for seeking the truth." The capacities a child needs for a lifetime of learning, the ability to attend to one thing, to tolerate boredom, to sit with a difficult question, to find a book more rewarding than a feed, are formed early and are precisely the capacities a hyper-stimulating digital environment erodes. The encyclical's call for what it terms "digital sobriety" is addressed substantially to families.

The practical implications are not a set of rules but a posture. Parental engagement matters more than any single age threshold. The questions worth asking are formative ones: Is my child developing the capacity to attend, or losing it? Is technology serving my child's formation or substituting for it? Am I supervising the digital environment my child is being formed in, or has it been outsourced to the devices and the platforms? The Catholic framework does not demand that families reject technology. It asks them to take responsibility for the formation that is happening either way.

On the specific homework question that parents face constantly: the same distinction that applies to students applies here. A parent helping a child use AI as a tutor that explains and asks questions is supporting formation; a parent letting AI do the child's work is permitting its avoidance. The parent's role is to help the child internalize the judgment about when the machine helps learning and when it replaces it.

For Catholic schools: the identity raises the standard

Catholic schools and universities face AI whether they engage it deliberately or not. Students are using it. Teachers are using it. The administrative systems are adopting it. The only choice a Catholic school actually has is whether to engage these questions with a framework or by default.

The Catholic identity of the school raises the standard rather than lowering it. A Catholic school claims to form persons in a particular understanding of what persons are. That claim makes the school more accountable for how it integrates a technology that bears directly on formation, not less. A Catholic school that adopts AI tools thoughtlessly, or that lets its formation mission be quietly reshaped by the logic of efficiency the tools carry, is failing its own stated identity.

The concrete need is explicit policy grounded in the formation understanding. Magnifica Humanitas calls for students to be taught when AI should and should not be used and for teachers to receive ongoing formation. A serious Catholic school AI policy should distinguish among subjects (AI assistance means something different in a coding class than in a literature seminar), among grade levels (the formation needs of a 7-year-old and a 17-year-old differ), and among assignment types (some assignments can incorporate AI deliberately; others must exclude it to do their formative work). The policy should be taught to students as part of their formation rather than imposed as a compliance regime, because learning to make the judgment is itself part of what the school is forming.

Catholic universities carry an additional vocation. They are among the institutions best positioned to produce the scholarship and the educated public conversation that the AI moment requires, and the Catholic intellectual tradition has resources for thinking about technology, the person, and the formation of the mind that secular institutions do not always have. The encyclical's call for universities to pursue "the integration of knowledge" is a call Catholic universities are particularly equipped to answer. Internal AI policy at Catholic universities should protect the faculty and the scholarly work engaged in exactly this inquiry rather than pressuring uncritical adoption of the tools that are themselves the object of study.

For the broader treatment of how the Catholic framework applies to Catholic institutions deploying AI across all their functions, including schools, see AI and the Church. For the workplace dimension that affects school employees, see AI in the Workplace.

What the framework asks

The Catholic framework on AI in education does not resolve into a list of rules, and the encyclical is deliberate about refusing to provide one. As Magnifica Humanitas states, the task is to educate people "to decide when and for what purpose it ought not to be used." The judgment is the point, and the judgment cannot be outsourced to a policy any more than the learning can be outsourced to a machine.

What the framework asks, of all four audiences, is a shift in the question. Not "is this allowed?" but "is this forming the person it should be forming?" Not "does this save time?" but "does this build or erode the capacity for sustained thought?" Not "is the work done?" but "is the learner being made?" These are harder questions than the rule-based ones, and they are the questions the Catholic understanding of education has always asked. AI has not changed what education is. It has raised the stakes of remembering.

The encyclical's closing note on this theme is neither alarmist nor naive. Pope Leo XIV does not call for rejecting the technology; he calls for restraint, judgment, and the protection of the young. "We must learn, then, how to exercise restraint in the use of AI and to protect our young people from the promise of the perfect machine." The machine is not the enemy. The loss of the capacity to think is the danger, and that loss is preventable by people who understand what education is for.