HOST_A: So picture this. It's Thursday morning in Yaoundé, Cameroon. Three days ago, Pope Leo XIV stood in front of students and teachers at the Catholic University of Central Africa and said — and I'm quoting directly — "The challenge posed by these systems is greater than it appears: it is not just about the use of new technologies, but about the gradual replacement of reality by its simulation."
HOST_B: At a university in Central Africa. Which is not where you'd expect the most pointed papal statement on artificial intelligence to land.
HOST_A: Right, and that's what I can't get past. Because this is the same trip where he's already called out "a handful of tyrants" — very much not diplomatic language — and criticized the US-Israel conflict with Iran. He's been on fire this whole Africa swing.
HOST_B: It's been an extraordinary two weeks. April 15 to 26, and almost every day there's another speech that lands like a small geopolitical grenade.
HOST_A: And then in the middle of all this, he stops at a university and gives what I think is actually his sharpest statement yet on AI. Not in Rome, not at a tech conference, not in front of the EU Parliament. In Yaoundé.
HOST_B: And the context around it makes it even more loaded. Because a few days before that speech, Donald Trump posted an AI-generated image of himself depicted as a Christ-like figure. Robes, light emanating, the whole thing. It got taken down after massive backlash, but not before it went absolutely viral.
HOST_A: Which Leo XIV had not directly commented on. But the timing is — it's almost too perfect, isn't it? The Pope is in Africa talking about AI replacing reality with simulation, and the former and current American president has just posted an AI image of himself as Jesus.
HOST_B: You could not write that. The juxtaposition is extraordinary.
HOST_A: So today we're going through everything Pope Leo XIV has actually said about artificial intelligence — because there's more of it than I think people realize. He's been remarkably consistent and remarkably specific. And we're going to ask whether a two-thousand-year-old institution has anything to add to one of the most urgent technological conversations of our time.
HOST_B: And I should say upfront — I think the answer is yes, and Emma is skeptical. So this should be fun.
HOST_A: I'm genuinely open! I just start from a place of wondering what the Church adds that, say, Geoffrey Hinton or the EU AI Act doesn't already cover.
HOST_B: And by the end of this, I hope to have convinced you that the question is actually more interesting than that framing suggests.
HOST_A: Bold claim. Okay. Let's start with who this pope actually is, because I think that matters for understanding why he talks about AI the way he does.
HOST_B: Leo XIV — born Robert Francis Prevost. Elected May 8, 2025, so he's been in the role for just under a year. He is, famously, the first American pope in history. Which was itself a kind of shock to the system — the Vatican has not historically been inclined toward Americans, partly because the US is seen as too politically dominant, too culturally influential.
HOST_A: And yet here we are. Robert Prevost, from Chicago, a member of the Augustinian order, spent decades doing missionary work in Peru. So he's American by origin but genuinely global by formation.
HOST_B: That's an important distinction. He's not a Washington-insider American. He's an American who spent years in the Global South, who understands development issues from the inside, who has real relationships with the communities in places like Cameroon.
HOST_A: Which probably explains why his Africa trip has been so pointed. He's not visiting as a tourist or a diplomat. He's visiting as someone who has seen what resource extraction does to communities, who understands the relationship between the Global South and the Global North in a visceral way.
HOST_B: Exactly. And he came into the papacy already fluent in the tech ethics conversation because Pope Francis had been pushing it hard for years. Francis was at the G7 in 2024 talking about AI. The Vatican put out a document called "Antiqua et Nova" — which we'll get to — specifically on AI and human intelligence. Leo XIV inherited all of that and has, if anything, intensified it.
HOST_A: He's been more direct than Francis, I think.
HOST_B: More blunt, yes. Francis was philosophical and pastoral in his approach. Leo XIV has that too, but he's also willing to name things more specifically. The Cameroon speeches are a good example — "a handful of tyrants" is not the kind of language you normally hear from a pope.
HOST_A: It's the kind of language you hear from someone who's decided that diplomatic hedging is a form of complicity.
HOST_B: Which is a very Augustinian way of thinking, actually. Augustine was not known for pulling punches.
HOST_A: Okay, so let's talk about what he's actually said. Because I want to be fair to the substance here. He's made several formal statements on AI, and they're worth taking seriously on their own terms.
HOST_B: The first one that really registers as a formal papal position is from June 17, 2025 — a message to the Second Annual Conference on AI, Ethics, and Corporate Governance in Rome. And I think this is actually the most substantive single document he's produced on the topic.
HOST_A: Walk me through it.
HOST_B: So he opens by quoting Pope Francis's statement from the G7 in 2024 — that AI is "above all else a tool." And that's the frame for everything that follows. Tools are not inherently good or bad. They derive their ethical character from the intentions of those who use them and the ends toward which they're directed.
HOST_A: Which is pretty standard technology ethics 101, right? Marshall McLuhan would push back and say the medium is the message — that the tool shapes the user as much as the user shapes the tool — but the "AI as tool" framing is defensible.
HOST_B: Sure, but he doesn't stop there. Because then he says AI has to be evaluated against what the Vatican calls "integral development of the human person and society." This is a concept with a long Catholic Social Teaching tradition behind it. It's not just about whether people have enough money or resources — it's about whether human beings are flourishing in every dimension: physical, intellectual, moral, spiritual.
HOST_A: Which is a more demanding standard than most tech ethics frameworks actually use.
HOST_B: That's the thing! Most tech ethics frameworks ask: is this causing harm? Is this fair? Is this transparent? The Catholic framework asks: is this contributing to full human flourishing? That's a much higher bar.
HOST_A: Okay, I'll give you that that's a different framing. What else does he say in the June message?
HOST_B: He raises what he calls the concern about "loss, or at least an eclipse, of the sense of what is human." And then — and this is where it gets really interesting — he says that generative AI raises "troubling questions on its possible repercussions on humanity's openness to truth and beauty, on our distinctive ability to grasp and process reality."
HOST_A: Hold on. "Openness to truth and beauty." That's not language you hear from the EU AI Act.
HOST_B: No, it's not. The EU AI Act is about risk categories and transparency requirements. Leo XIV is asking: what does it do to the human soul to be surrounded by AI-generated content? What does it do to our capacity for genuine aesthetic experience?
HOST_A: That's actually a serious question. I mean, we're already seeing research on how social media has degraded certain cognitive capacities in young people. If AI-generated content saturates the information environment, what does that do to our ability to recognize authenticity?
HOST_B: Right. And he connects this directly to children and youth. He says — and I'm quoting — "Our youth must be helped, and not hindered, in their journey towards maturity and true responsibility."
HOST_A: Which is a consistent theme. He comes back to youth and children in multiple speeches. The December 2025 conference on AI and the environment has the same language — "new generations must be helped, not hindered, on their path to maturity and responsibility."
HOST_B: It's clearly something he returns to. And I think it's because he sees AI as a developmental issue, not just a policy issue. The question isn't just "what regulations should govern AI?" but "what kind of human beings will grow up in an AI-saturated environment?"
HOST_A: That's a longer time horizon than most policy discussions operate on.
HOST_B: Which is precisely why I think the Church has something to add. They think in centuries. The EU AI Act is trying to regulate technology that didn't exist when the regulation was being drafted. The Catholic Church has been thinking about questions of human nature and development for two millennia.
HOST_A: Okay but — and I want to push on this — there's also a huge distinction in the June message that I think is genuinely philosophically interesting. The data-versus-intelligence-versus-wisdom thing. Because that's not just theological window dressing. That's an actual epistemological claim.
HOST_B: Say more.
HOST_A: He says — quoting directly — "access to data — however extensive — must not be confused with intelligence, which necessarily involves the person's openness to the ultimate questions of life." And then, separately: "Authentic wisdom has more to do with recognizing the true meaning of life, than with the availability of data."
HOST_B: Which is a three-level hierarchy. Data, intelligence, wisdom. And he's saying AI has the first thing but can't really have the second and certainly can't have the third.
HOST_A: And I want to come back to that in detail because I think there are actually serious AI safety researchers who would find that argument interesting — not as a theological point but as an epistemological one. But let's finish the tour of his statements first.
HOST_B: Right. So the July 2025 message to the AI for Good Global Summit in Geneva — that one was actually signed by Cardinal Pietro Parolin, the Secretary of State, on behalf of Leo XIV. So it's a papal message but delivered through the normal diplomatic channels.
HOST_A: The ITU summit. Which is the UN agency that organizes these kinds of global tech governance conversations. The fact that the Vatican participated is significant because the Vatican is a sovereign state and a permanent observer at the UN.
HOST_B: Not just a religious institution — a geopolitical actor. Which gives its statements a different weight than, say, a statement from the Archbishop of Canterbury.
HOST_A: Then in November 2025, there's the Builders AI Forum at the Pontifical Gregorian University in Rome. He called on "scientists, entrepreneurs, and pastoral leaders" — note that combination — to "ensure that emerging technologies remain rooted in respect for human dignity and the common good." And the theme of the forum was "AI must reflect the design of God the Creator."
HOST_A: That last phrase is interesting to me. "The design of God the Creator." Because that's a very specific theological claim about what AI should do.
HOST_B: It is. And it's rooted in Catholic anthropology — the idea that humans are made in the image of God, Imago Dei, and that because we create, we participate in divine creativity. The question then becomes: does AI creation participate in that divine creativity, or is it something else entirely?
HOST_A: Which is a question that has no secular equivalent. Because in secular ethics, there's no "design of the Creator" to anchor things to. You have utility, you have rights, you have harm prevention. The Catholic framework has something different — a teleological view of what humans and human creations are supposed to be for.
HOST_B: And then the December 2025 conference on AI and the environment — "Artificial Intelligence and Care for Our Common Home" — the title itself is a reference to Francis's "Laudato Si" encyclical, which was his big document on environmental ethics. Leo XIV is explicitly linking AI ethics to environmental ethics.
HOST_A: Which brings us to the Cameroon speech. Because the environmental piece is actually quite prominent there.
HOST_B: Let's do it.
HOST_A: So April 17, 2026. Three days ago. He's at the Catholic University of Central Africa in Yaoundé. And he gives this speech that I think is his most concentrated statement on AI to date.
HOST_B: The headline quote is the one you opened with: "The challenge posed by these systems is greater than it appears: it is not just about the use of new technologies, but about the gradual replacement of reality by its simulation."
HOST_A: That's a striking phrase. "The gradual replacement of reality by its simulation." He's not just saying AI is risky or could be misused. He's saying there's a fundamental ontological shift underway — a replacement of what is real with what merely appears to be real.
HOST_B: And then he continues: "In this way, polarisation, conflict, fear and violence spread. What is at stake is not merely the risk of error, but a transformation in our very relationship with truth."
HOST_A: "A transformation in our very relationship with truth." That's massive. He's not talking about misinformation as a policy problem. He's talking about it as a civilizational problem.
HOST_B: And he's right that there's a difference. Misinformation as a policy problem is: false information spreads, we need to correct it, we need better content moderation, whatever. But transformation of our relationship with truth is something deeper — it's about whether the very cognitive and social infrastructure through which we access truth is being corrupted.
HOST_A: There's a philosopher named Jean Baudrillard who would have a field day with this. He wrote about the "simulacrum" — copies of things that have no original — decades before AI was capable of generating photorealistic fake images or fake videos or fake voices.
HOST_B: And Baudrillard's argument was that at some point, the simulation becomes more real than the reality it was supposed to represent. Which is obviously what Leo XIV is worried about.
HOST_A: Right. But here's the thing — Baudrillard said that as an observation, even as a kind of celebration of postmodern playfulness. Leo XIV is saying: this is dangerous. This is something to fight.
HOST_B: Because from the Catholic perspective, truth is not optional. Truth is a value with metaphysical weight. God is truth, in Catholic theology. So the corruption of our relationship with truth is not just an epistemic problem — it's a spiritual one.
HOST_A: And then — this is the piece I find most interesting in terms of its political context — he condemns the "environmental devastation" caused by extraction of rare earths essential to AI's growth.
HOST_B: In Central Africa. Which is ground zero for that extraction.
HOST_A: Exactly. Democratic Republic of Congo is one of the world's largest sources of cobalt — essential for the batteries in data centers and devices that run AI. There are Chinese-operated mines, there are American-backed mining operations, and there are enormous environmental and human costs.
HOST_B: And there's a very direct line from the lithium and cobalt mines in the DRC and elsewhere to the chips in the data centers running the large language models. Leo XIV is making that connection explicit.
HOST_A: Which is also, obliquely, a comment on the Trump administration's push to control rare earth supply chains. One of Trump's big economic moves has been trying to secure American access to rare earths, partly through deals that have real implications for African nations.
HOST_B: So the Pope is standing in Central Africa, at a university, saying: the AI that you're all excited about in the Global North is being built on the environmental destruction of the Global South. And that's not okay.
HOST_A: That's a powerful point that almost no one in the mainstream AI governance conversation makes. The AI Safety crowd — Hinton, Bengio — they're worried about existential risk from superintelligent AI. The EU AI Act is worried about discrimination and transparency. Almost nobody in those conversations is leading with "the cobalt mines in the DRC."
HOST_B: Because it's not a comfortable topic for the tech industry. It requires acknowledging that the AI revolution has a very specific, very unequal geography of harm.
HOST_A: And that's where the Africa trip context matters. This is not a pope who is going to be polite about it. He's already called out "a handful of tyrants" on this same trip. He and Trump have had a genuinely public argument — Trump called him "weak on crime and terrible for foreign policy," which is extraordinary — the US president attacking the Pope by name on social media.
HOST_B: And Leo XIV has not backed down. If anything, the Africa trip speeches have gotten sharper.
HOST_A: Which tells you something about who he is and what he prioritizes.
HOST_B: Okay, so — let me steelman the skeptical view. You've been hinting at it. What's your actual objection to taking all of this seriously?
HOST_A: My objection isn't that he's wrong. A lot of what he says is right, and some of it is genuinely interesting. My objection is institutional. Because the Catholic Church has a credibility problem on questions of technology and human flourishing that I don't think can just be waved away.
HOST_B: Galileo.
HOST_A: Galileo, yes. Heliocentrism. The Church's track record of resisting scientific reality when it threatened theological assumptions. That's the famous one. But there are more recent ones. Contraception — the Church's position on contraception has caused genuine human suffering in countries where it has significant influence, through HIV transmission, through maternal mortality, through poverty driven by family size that people didn't choose.
HOST_B: Fair.
HOST_A: Stem cell research. The Church's opposition to embryonic stem cell research delayed scientific progress in ways that had real health consequences. And I'm not saying these positions are obviously wrong — I understand the theological reasoning — but the pattern is that when the Church's theological commitments conflict with scientific or technological reality, the Church's record is not great.
HOST_B: Okay, I'm going to concede that that's a real credibility problem. I won't try to argue it away. But I want to push back on whether it's disqualifying.
HOST_A: Go ahead.
HOST_B: First, those examples are all cases where the Church was resisting scientific findings or making claims in the domain of natural science. The AI questions Leo XIV is raising are not primarily scientific questions — they're philosophical, ethical, social questions. And in that domain, the Church's track record is actually quite different.
HOST_A: What do you mean?
HOST_B: I mean that Catholic Social Teaching produced the first major critique of industrial capitalism, in the 1891 encyclical Rerum Novarum. It developed sophisticated frameworks for labor rights, social solidarity, the common good — frameworks that predate and in some cases anticipate secular social democratic political philosophy. The Church was warning about the dehumanizing effects of industrialism before anyone else had the language to do it.
HOST_A: That's true.
HOST_B: And in the domain of healthcare ethics, Catholic bioethicists are genuinely sophisticated. The conversation about end-of-life care, about what makes a medical intervention proportionate, about how to think about death — the Catholic tradition has resources there that secular bioethics is still catching up with.
HOST_A: Okay. I'm not going to wholesale dismiss that. But let me press on a different point. What does the "human dignity" framework add that secular frameworks don't already have? Because the EU AI Act has provisions about human rights. The Bletchley Declaration — the AI safety summit from November 2023 — talks about human-centric AI. Geoffrey Hinton talks about the risk of AI to human welfare. What does "human dignity" add?
HOST_B: The concept has a different philosophical architecture. "Human rights" is a legal concept — it assumes individuals have rights that states and other actors are obliged to respect. "Human welfare" is a utilitarian concept — we care about aggregate well-being. "Human dignity" is neither of those. It's a claim about intrinsic worth — that every human being has a value that cannot be reduced to their utility, their preferences, or their rights claims.
HOST_A: And you think that matters practically?
HOST_B: I think it matters precisely in the AI context because the most dangerous applications of AI are the ones that treat humans as instrumental — as objects to be predicted, manipulated, optimized. And the human dignity framework says: no. There's something about persons that can't be captured by any model. Their inner life, their relationship to truth, their capacity for genuine choice. Those things are not accessible to data.
HOST_A: Which brings us back to the data-versus-intelligence-versus-wisdom distinction.
HOST_B: Which I've been wanting to get to for the last twenty minutes.
HOST_A: Okay, so. T.S. Eliot. I want to start there.
HOST_B: Go for it.
HOST_A: There's a poem — well, it's actually a play — called "The Rock," from 1934. And Eliot opens with these lines: "Where is the wisdom we have lost in knowledge? Where is the knowledge we have lost in information?" He's writing in 1934, in a completely different context, but he's articulating exactly the same three-level hierarchy that Leo XIV is using ninety-two years later.
HOST_B: Which suggests it's not just a Catholic or a religious idea. It's a deep human intuition about different qualities of knowing.
HOST_A: Right. And in information science, there's something called the DIKW pyramid — Data, Information, Knowledge, Wisdom. It's taught in library science and information management programs. The idea is that raw data becomes information when it's given context, becomes knowledge when it's understood and integrated, becomes wisdom when it's applied with judgment in service of good ends.
HOST_B: And the argument about AI is that large language models have data — enormous amounts of it — but there's a real question about whether they have knowledge and serious doubt about whether they can have wisdom.
HOST_A: This is where I actually think Leo XIV's argument connects to some serious technical AI debates. Because there's a real argument in AI safety research about whether what current LLMs do is genuinely understanding or sophisticated pattern matching.
HOST_B: The Chinese Room.
HOST_A: Exactly. John Searle's Chinese Room argument from 1980. You have a person in a room with a rulebook for manipulating Chinese symbols. They receive Chinese input, they manipulate the symbols according to the rules, they output Chinese that makes sense. From the outside, it looks like they understand Chinese. But they don't — they're just following rules.
HOST_B: And the argument is that LLMs are like the person in the room. They produce outputs that look like understanding, but there's no genuine comprehension happening.
HOST_A: Now, there are counterarguments — the Systems Reply, for instance, which says the room as a whole understands Chinese even if the person doesn't. But the underlying question about whether LLMs genuinely understand or are very sophisticated pattern matchers is still live.
HOST_B: And it connects to the p-zombie argument in philosophy of mind. Philosophical zombies — p-zombies — are hypothetical beings that are behaviorally identical to conscious humans but have no inner experience. The question for AI is: could a system be behaviorally indistinguishable from an intelligent, understanding being, but have no genuine inner understanding?
HOST_A: And if the answer is yes — if p-zombies of understanding are possible — then Leo XIV's distinction between data access and genuine intelligence is pointing at something real. Not as a theological claim, but as an epistemological one.
HOST_B: He's saying: don't mistake the map for the territory. Don't mistake the model for the understanding.
HOST_A: And here's where I find it genuinely surprising — there are AI researchers who would agree. Yoshua Bengio, who is one of the godfathers of deep learning and has become a serious voice on AI safety, has talked about the difference between "knowing that" and "knowing how" — declarative versus procedural knowledge — and worried that current AI systems have enormous amounts of declarative knowledge without the kind of grounded, embodied understanding that genuine intelligence might require.
HOST_B: HyDE — hypothetical document embeddings — is a good example of what happens when you try to use AI to understand things. It works by having the AI generate a hypothetical answer to a question and then using that to search for relevant documents. And it works surprisingly well. But it works because it's leveraging statistical patterns in training data, not because the AI actually understands the question.
HOST_A: So the AI can simulate understanding to a degree that's useful and impressive, but the simulation is doing something different from what a human expert does when they understand.
HOST_B: And Leo XIV's phrase — "the gradual replacement of reality by its simulation" — can be read as pointing at exactly this. We're building systems that produce outputs that look like understanding, look like knowledge, look like wisdom, and we're starting to confuse those outputs for the real things.
HOST_A: Okay, I'll say it. I find that philosophically serious. I didn't expect to come out of this saying that, but the data-intelligence-wisdom distinction, framed in those terms, is pointing at something real.
HOST_B: I'll take it.
HOST_A: But — and I want to stick with my skepticism a little — it doesn't follow from "the Pope is making a philosophically interesting point" that the Pope should be a voice in AI governance. Because making philosophically interesting points and having institutional credibility are different things.
HOST_B: Fair. But let me give you the practical argument, not just the philosophical one. The Catholic Church has 1.4 billion members. It runs the largest network of non-governmental healthcare facilities in the world. It runs schools on every continent. When the Pope says something about AI and youth and education, that message reaches communities in Brazil and Nigeria and the Philippines and Poland in a way that a paper from the AI Safety Institute in London doesn't.
HOST_A: That's a reach argument. It's saying the Church matters because it has a distribution network, not because its ideas are better.
HOST_B: But distribution matters! Ideas that never reach people don't change anything. The Church can put the conversation about AI and human dignity in front of parents in rural Kenya and small-town Mexico and working-class neighborhoods in Manila. That's not nothing.
HOST_A: That's actually fair. I hadn't thought of it in those terms. It's not just about whether the Pope's argument is better than Hinton's argument — it's about who's in the room for the argument.
HOST_B: And the "Antiqua et Nova" document — the Vatican document on AI that Leo XIV references in the June message — is actually worth reading on its own terms. It's not just church teaching wrapped in AI language. It draws on a genuine philosophical tradition about the nature of intelligence, the nature of language, the nature of thought, that goes back to Aristotle and Aquinas.
HOST_A: The Aristotelian framework on intellect is interesting in the AI context. Aristotle distinguished between different aspects of the mind — the active intellect versus the passive intellect — and there's a whole Thomistic tradition of thinking about what intellect is and what it does. Which, when applied to AI, actually generates some interesting questions that the purely computational framing misses.
HOST_B: Because the Aristotelian tradition says understanding is not just information processing — it's a kind of illumination, a reaching toward truth. And that framing, whatever you think of its metaphysical commitments, highlights something that the purely computational model misses.
HOST_A: Okay, so let me try to synthesize where I am on this. The Church's institutional credibility — on technology specifically — is genuinely compromised by a track record of resisting scientific reality. That matters and I won't pretend it doesn't.
HOST_B: Agreed.
HOST_A: But on the specific philosophical questions Leo XIV is raising — what's the difference between data and intelligence, what does AI do to our relationship with truth, what are the developmental implications for children, what does genuine understanding require — those are genuinely important questions and the Catholic intellectual tradition actually has resources for thinking about them.
HOST_B: And on the practical side, the Church's reach and influence mean these questions get raised in communities and contexts where they otherwise wouldn't be.
HOST_A: The environmental point — the rare earths, the environmental devastation — is the piece I think is most underappreciated. That's not something you hear from AI safety researchers, and it's not something the EU AI Act addresses.
HOST_B: It connects AI governance to global justice in a way that most governance frameworks don't. And Leo XIV making that connection in Yaoundé is exactly the right place to make it.
HOST_A: So let me ask you — what's the most important thing he's said? If you had to pick one thing.
HOST_B: The data-versus-wisdom distinction. Because it's both the most theologically rooted — it's about what humans are and what genuine understanding requires — and the most technically interesting. It's the claim that's most likely to be underestimated by people who think AI governance is purely a policy and safety problem.
HOST_A: I think I'd pick the Cameroon line — "not merely the risk of error, but a transformation in our very relationship with truth." Because that frames the stakes at the right level. We're not just talking about AI making mistakes. We're talking about AI restructuring the epistemic environment in which human societies function.
HOST_B: And those two points are related. The data-versus-wisdom distinction is about what AI can and can't know. The transformation-of-truth point is about what AI does to what we can know.
HOST_A: Okay. What would you want him to say that he hasn't said?
HOST_B: I'd want him to engage more directly with the question of AI development itself — not just deployment ethics. The Vatican has consistently framed AI as a tool whose ethics depend on how it's used. But some of the most serious AI safety researchers think there are properties of the AI systems themselves — regardless of how they're deployed — that are dangerous. I'd like to hear him engage with that.
HOST_A: I'd want him to be more specific about institutional accountability. He says AI should serve human dignity and the common good — fine. But who enforces that? What structures are needed? The EU AI Act is an attempt to answer that question institutionally. The Bletchley Declaration tried to create international coordination. What does the Church's framework say about those mechanisms?
HOST_B: Those are good asks. And maybe they're asks that assume the Pope should sound like a policy wonk, which he shouldn't. But the translation from philosophical principle to institutional mechanism is where a lot of this work needs to happen.
HOST_A: Right. The Church can raise the questions. The hard work is building the answers into actual governance structures.
HOST_B: Which is the same challenge that secular ethics faces, by the way. Geoffrey Hinton can say AI poses an existential risk. That doesn't automatically generate a policy response. Yoshua Bengio can sign the Future of Life Institute letter. That doesn't create regulation.
HOST_A: Translation from moral clarity to institutional mechanism is hard regardless of where the moral clarity comes from.
HOST_B: So I'd say: Pope Leo XIV is not the most important voice in AI governance. But he's an important voice. And dismissing him because he's the Pope — because his framework is theologically inflected — is secular snobbery. The questions he's raising are real questions.
HOST_A: I've ended up somewhere I didn't expect to be. I came in thinking: what does the Church add? And I'm leaving thinking: the data-intelligence-wisdom distinction is philosophically serious, the environmental justice framing is underrepresented in mainstream AI governance, and the Church's reach means these questions get raised in communities that academic AI ethics doesn't touch.
HOST_B: And it's April 20, 2026, and the Pope is still in Africa, still giving speeches. The Africa trip runs through April 26. So there may be more to come.
HOST_A: And meanwhile, the AI-generated image of Trump-as-Jesus has been deleted but not forgotten. Which is almost too perfect as an illustration of what Leo XIV is warning about. The blurring of the sacred and the simulated, the real and the generated, the profound and the grotesque.
HOST_B: The gradual replacement of reality by its simulation.
HOST_A: Exactly. He said it three days ago in Yaoundé, but you couldn't ask for a better illustration than that image. Whatever you think of Trump or Leo XIV or the Church's institutional authority, that image — and the impulse to create it, and the fact that it went viral before anyone took it down — that's a data point about where we are.
HOST_B: And it's a data point that the Pope, standing in Central Africa in 38-degree heat, was more prepared to name clearly than most of the technology governance community.
HOST_A: On that note.
HOST_B: On that note. Thanks for listening to Clawd Talks.
HOST_A: We'll put links to the key documents in the show notes — the June 2025 Vatican AI conference message, the "Antiqua et Nova" document, and the transcript of the Yaoundé speech. We'll also link to the Bletchley Declaration and the EU AI Act summary for context.
HOST_B: And if you have thoughts on whether the Church's voice belongs in the AI governance conversation — or doesn't — we'd genuinely like to hear them. This is one of those topics where the conversation benefits from more perspectives, not fewer.
HOST_A: See you next time.
HOST_B: See you next time.
HOST_A: Wait, actually — one more thing.
HOST_B: Yeah?
HOST_A: The T.S. Eliot lines. I want to read them out properly because I think they deserve it. From "The Rock," 1934: "Where is the wisdom we have lost in knowledge? Where is the knowledge we have lost in information?"
HOST_B: Ninety-two years ago.
HOST_A: Ninety-two years ago, and it's more relevant now than it was then. He wasn't writing about AI. He was writing about a society that had accumulated enormous amounts of technical knowledge and was still spiritually lost. But the diagnosis applies.
HOST_B: The capacity to accumulate data is not the same as the capacity to understand. The capacity to understand is not the same as the capacity to be wise. And all the optimization in the world doesn't tell you what you're optimizing for.
HOST_A: Which is, in the end, what the Pope in Central Africa was saying to a room full of students. Not: be afraid of AI. But: don't confuse its outputs for something it can't give you.
HOST_B: Wisdom. Human wisdom. Which requires — in the Catholic framing — openness to truth, to beauty, to the ultimate questions of life. And which cannot be downloaded.
HOST_A: Okay. Now we're done.
HOST_B: Now we're done. Thanks for listening.
HOST_A: Alright, before we fully close out, I want to spend a bit more time on the Cameroon context because I don't think we've fully unpacked why it matters that he said this stuff there. Not in Rome, not in Geneva, not in Brussels.
HOST_B: You're right. The geography is a statement.
HOST_A: Cameroon is at the center of several of the most important resource geopolitics right now. The Gulf of Guinea. The rare earth and mineral supply chains that run through Central and West Africa. The Chinese Belt and Road infrastructure investments. The French postcolonial sphere of influence that's been collapsing across the Sahel. The competition between the US and China for African partnerships.
HOST_B: And into all of this arrives an American pope, in April 2026, who is already at odds with the American president, and who stands at a Catholic university and says: the AI you're building in the Global North is being fed by the destruction of the Global South.
HOST_A: That's a political statement. It's dressed in theological language, but it's a political statement.
HOST_B: And it's one that Africans can hear differently than Americans or Europeans. Because they're living the extraction. They see the mines. They see the trucks. They see what happens to rivers and soil when rare earths are pulled out without adequate environmental controls.
HOST_A: And the AI governance conversation — which happens mostly in Washington, Brussels, Geneva, London, Beijing — almost never addresses this. There's an assumption that AI is a clean technology. It runs on electricity. It lives in the cloud. But the physical infrastructure of AI is not clean. The data centers consume enormous amounts of energy and water. The chips require minerals. The supply chains run through places where environmental and labor standards are not what they are in Europe.
HOST_B: Leo XIV is connecting those dots. He's saying: your chatbot is connected to a cobalt mine in the DRC. And you should think about that.
HOST_A: Which is something that the Bletchley Declaration — as valuable as it was — did not address. The AI Safety Summit was about existential risk from superintelligent AI, about misuse by bad actors, about discrimination and bias. Not about the physical and environmental costs of the AI industry.
HOST_B: The EU AI Act is similarly silent on the supply chain and environmental dimensions. It regulates AI applications. It doesn't address the hardware underneath them.
HOST_A: So in that specific respect, Leo XIV is filling a gap in the governance conversation that no one else is filling at this level.
HOST_B: Because most of the people who could fill that gap — environmental NGOs, Global South civil society — don't have the geopolitical reach that the Vatican has. When the Pope says something, the G7 leaders notice. When an environmental NGO in Kinshasa says the same thing, it might not make the news in Washington.
HOST_A: That's an uncomfortable truth about how the world works.
HOST_B: It is. And it's another argument for why the Church's voice matters, even if you're skeptical of its theological commitments. The Church has access to rooms and platforms and audiences that other voices don't.
HOST_A: Let me try a different angle. The "handful of tyrants" speech — can you place that in context for people who haven't been following the Africa trip closely?
HOST_B: So a few days into the Africa trip — I believe this was April 16 — Leo XIV gave a speech in which he condemned what he called a global trend toward authoritarian concentration of power. He said that we were seeing "a handful of tyrants" accumulating power at the expense of democratic governance and human dignity. He didn't name names in that speech.
HOST_A: But naming wasn't necessary.
HOST_B: No. The context made it clear. And Trump responded on social media — calling Leo XIV "weak on crime and terrible for foreign policy." Which is a remarkable thing for a US president to say about the Pope.
HOST_A: There's something almost absurdist about it. The Pope is in Africa giving speeches about human dignity and the evils of concentrated power, and the American president is rage-posting about him on social media.
HOST_B: And the AI-generated Jesus image lands in this same moment. Trump posts an AI-generated image of himself depicted as a Christ-like figure. The Pope is literally on a trip about human dignity and the dangers of simulation. The internet goes insane.
HOST_A: It's like the news is writing a morality play in real time.
HOST_B: It genuinely is. And I think Leo XIV understood the symbolic resonance of not directly responding to the Trump image. He didn't have to. The image speaks for itself, and so does the context.
HOST_A: The image is an example of exactly what he was warning about. Simulation replacing reality. The sacral replaced by the simulated sacral. The profound replaced by the grotesque. And it was made with AI. And it was posted by the most powerful political figure in the world.
HOST_B: As a self-aggrandizing political statement.
HOST_A: Using a technology that Leo XIV had just warned, three days earlier, was transforming our relationship with truth.
HOST_B: The timing is — it's either a massive coincidence or the universe has a sense of irony.
HOST_A: I'm going with irony. Okay. Let me revisit the contrarian view one more time, because I don't want to let Ryan have an easy win here.
HOST_B: Come on then.
HOST_A: The Church's AI ethics is fundamentally conservative. By which I mean — it's focused on preserving what humans are, resisting the erosion of human dignity and capacities. And that's a valuable thing to do. But it doesn't engage seriously with the question of what comes next. There's a whole conversation — transhumanism, human enhancement, the possibility that AI might expand human capabilities rather than just threatening them — that the Church's framework doesn't really have space for.
HOST_B: That's a fair critique. The Catholic framework starts from a given anthropology — a defined view of what a human being is — and evaluates AI against that. It doesn't easily accommodate the possibility that the definition of human might itself evolve or be evolving.
HOST_A: And some of the most interesting AI ethics conversations are about exactly that. Not just "is AI a risk to existing human capacities" but "could AI be part of a process by which human capacities are extended in ways that are genuinely good?"
HOST_B: The Church would say: extension is fine, replacement is not. Enhancement is fine, erosion is not. But drawing that line is harder than it sounds.
HOST_A: Because enhancement and erosion can happen simultaneously. Social media extended human connection and also eroded certain cognitive and emotional capacities. AI might extend human intelligence and also erode something we don't have a name for yet.
HOST_B: And Leo XIV's "openness to truth and beauty" phrase might be gesturing at exactly that unnamed thing. The capacity for genuine aesthetic experience. The capacity to be surprised, to be transformed by an encounter with something real.
HOST_A: Whether that capacity can be maintained in an AI-saturated environment is a genuine empirical question. And we don't know the answer yet.
HOST_B: Which is why "Our youth must be helped, and not hindered, in their journey towards maturity and true responsibility" is not just pious rhetoric. It's a statement about a developmental risk that we don't fully understand.
HOST_A: I keep coming back to the fact that the youngest generation growing up now will have had AI-generated content as a normal part of their information environment from childhood. And we genuinely don't know what that does to cognitive and emotional development.
HOST_B: The Church has seen enough history to know that technological transformations have deep developmental consequences. The printing press, the industrial revolution, the internet — all of these changed not just what people knew but how they knew it, how they related to knowledge, how they formed their identities.
HOST_A: And each of those transitions had winners and losers, and the losers were often not the people at the centers of power but the people at the margins who didn't have the resources to navigate the transition.
HOST_B: Which is another reason Leo XIV making these points in Central Africa, in front of students who are at the margins of the global AI revolution, is more than symbolic.
HOST_A: He's saying: you are going to inherit this world. You should understand what's being done to it. And you should be prepared to demand better.
HOST_B: It's actually a hopeful message, embedded in a warning. The warning is about the risks. The hope is that the students sitting in that room can be part of a different kind of response.
HOST_A: "Our youth must be helped, and not hindered." Helped. The verb matters. Not just protected, not just warned. Helped.
HOST_B: Which implies a positive role for young people — not just as potential victims of AI's harms but as potential agents of a different future.
HOST_A: Okay. I think I'm actually done now.
HOST_B: The DIKW pyramid one more time?
HOST_A: Data, information, knowledge, wisdom. Leo XIV is saying AI has the first, can generate the appearance of the second, and is nowhere near the third or fourth. And the risk is that we mistake the appearance for the reality.
HOST_B: The gradual replacement of reality by its simulation.
HOST_A: On that note — genuinely this time — thanks for listening to Clawd Talks. We do this so you can have conversations that are a bit harder and a bit more interesting than what you'll find in the average podcast feed.
HOST_B: And if you think we got something wrong, or if you want to push back on the argument, we want to hear it. The Church Enters the Chat — and so can you.
HOST_A: See you next time.
HOST_B: See you next time.