Pope Leo XIV's AI encyclical is really about power, not algorithms

Pope Leo XIV has called on governments to "disarm" artificial intelligence, issuing a first encyclical that frames the technology not primarily as a tool but as a structural force reshaping power at scale. The document, released on 25 May 2026, urges international regulators to slow the deployment of military and surveillance AI systems and warns that unchecked automation threatens democratic governance itself. The Vatican described it as the first major teaching document of his papacy — and the timing signals deliberate intent, landing as the G7 digital governance summit approaches.

The encyclical does not oppose AI development outright. Its core argument is subtler: that the architecture of artificial intelligence, as currently configured, tends to consolidate decision-making authority away from public oversight and toward a narrow class of private actors. That reading — not a blanket tech-skeptic screed — is what makes the document analytically notable.

What the document actually says

The encyclical draws a direct line between AI systems and the concentration of geopolitical power. According to Reuters, Pope Leo warned that AI could "spread misinformation, fuel conflict and risk perpetuating inequalities" on a civilisational scale. The language of "disarmament" is deliberate: the Vatican is not calling for a ban on AI, but for binding international constraints on autonomous weapons and mass-surveillance applications — the military-adjacent use cases that have most alarmed ethicists, arms-control advocates, and a growing coalition of Global South governments.

Middle East Eye reported that the document represents the first major teaching document of the new papacy, signalling that AI governance will be a defining theme of Pope Leo's pontificate. That framing matters. Papal encyclicals set the intellectual agenda for one of the world's oldest institutions; this one is explicitly aimed at shaping the terms of a global debate still dominated by Silicon Valley norms and great-power competition.

The geopolitical weight of a papal intervention

The Vatican's influence in multilateral spaces is real but asymmetric. It holds observer status at the UN and has used that platform consistently to push for weapons restrictions, humanitarian corridors, and the rights of marginalised peoples. When it speaks on AI, it does so as a non-aligned institutional actor — which gives its voice a particular quality in a debate otherwise conducted between Washington, Beijing, Brussels, and the major technology companies themselves.

The encyclical arrives at a moment when the regulatory landscape is fragmented. The EU's AI Act has set the most comprehensive binding framework to date, but its implementation remains uneven. The United States has relied primarily on executive orders that are easily reversed between administrations. China has moved quickly on AI governance within its own borders but faces questions about how its frameworks apply to Belt and Road infrastructure exported abroad. Across the Global South, states are largely consuming AI systems designed elsewhere, with little input into their architecture or governance rules.

Pope Leo's intervention is not neutral in that landscape. A papal encyclical carrying moral authority — rather than legal binding force — is a specific kind of soft power. It shapes the vocabulary that smaller states can use when pushing back against great-power AI dominance. It gives civil society advocates a doctrinal anchor. And it creates an upstream argument that governments and tech companies cannot easily dismiss without reputational cost.

Why this is really about power, not algorithms

TechCrunch's analysis of the document — published before the full text was publicly available — caught the essential frame: the encyclical uses AI as a lens to diagnose older problems, specifically "concentrated power, eroding democracy, and a tech elite that shapes the world to its own advantage." That reading holds when the full text is read in context.

The underlying argument has two parts. First, that the institutions currently governing AI — national regulators, private platforms, and the occasional multilateral body — are structurally incapable of the kind of binding coordination that the scale of the technology demands. Second, that the Vatican's own history of navigating great-power competition gives it a standing to call for what it frames as a form of collective restraint.

Whether that standing is recognised by the relevant actors is a separate question. The major technology companies have institutional muscles for absorbing and neutralising criticism of this kind. Great powers have strategic interests that do not bend easily to moral arguments. The Vatican's leverage is rhetorical and reputational, not economic or military.

But the alternative — treating AI governance as a purely technical or commercial matter — has already produced a set of outcomes the encyclical identifies as problems. Algorithmic content moderation shaped elections. Predictive systems determined access to credit, social services, and legal representation. Autonomous weapons redefined the battlefield faster than international humanitarian law could track. None of those outcomes were inevitable. They were the product of decisions made by actors who faced few binding constraints. The encyclical's argument is that the window for inserting constraints is narrow, and that the institutional architecture currently available to do so is not adequate to the task.

The stakes ahead

The encyclical does not create immediate legal obligations for any government or company. What it does is reframe the AI governance debate from a competition over technical standards to a question about the kind of political order that technology is building. That is a harder argument to dismiss, and a harder one to resolve.

The practical test will come in the next eighteen months: at the G7 digital summit, at UN AI governance talks that have been stalled since 2025, and in the EU's implementation review of the AI Act. The Vatican's voice will be present in those conversations. Whether it shifts the balance depends on whether the coalition of smaller states, civil society groups, and institutional investors who find the document's arguments useful can translate doctrinal argument into negotiating positions.

Pope Leo XIV has made the Vatican's position clear. The harder question — whether the great powers and the technology companies that actually shape AI development will engage with it seriously — remains open.

This publication covered the encyclical through the lens of institutional power and governance architecture, framing the Vatican's intervention as a structural challenge to the current distribution of AI authority rather than as a technology-skeptic document. Reuters and Middle East Eye led with the "disarm AI" framing; this article treated that as the Vatican's chosen rhetorical hook and focused on what the argument is actually about.

Wire provenance

This editorial synthesis draws on the following public wire/social posts:

• https://x.com/Polymarket/status/1982309918272348161