Pope Leo XIV published his first encyclical on May 25, 2026 — a Vatican document exceeding 42,000 words — addressing the risks of unconstrained AI development. Titled Magnifica Humanitas, the document calls for new legal and ethical frameworks for AI, protecting human dignity against the automation of labor, the militarization of algorithms, and the systemic concentration of technological power.
Key takeaways
- Magnifica Humanitas exceeds 42,000 words — the first papal document of its kind on AI
- Leo XIV warns against the "Babel syndrome" — the uniformization and reduction of humans to data and performance metrics
- Proposals include social criteria for AI deployments, worker retraining programs, and a ban on autonomous lethal force decisions
- Anthropic co-founder Christopher Olah was present at the Vatican presentation of the encyclical
- Representatives from Amazon, Meta, and Google met with Vatican officials ahead of the publication
The encyclical: 42,000 words on the crisis of dignity
A papal encyclical is the highest-ranking open letter the Catholic Church issues. Leo XIV chose AI as the first major declaration of his pontificate, explicitly referencing his predecessor — Leo XIII — who published Rerum Novarum in 1891 in defense of workers during the industrial revolution. The choice of papal name was itself a programmatic declaration.
The central concept of Magnifica Humanitas is the defense against what the pope calls the "Babel syndrome": the idolatry of profit that sacrifices the weak, a uniformity that neutralizes differences, and the pretense that a single language — even a digital one — can translate everything, including the mystery of the person, into data and performance. This warning is directed at technology companies and governments alike.
"To disarm means discrediting the assumption that technical power automatically confers the right to govern. To disarm does not mean rejecting technology, but preventing it from dominating humanity."
Pope Leo XIV, Magnifica Humanitas, 2026
Concrete proposals — not only moralizing
The encyclical goes beyond philosophical lament. The pope articulates a set of specific regulatory and social proposals:
- Social criteria for introducing automation and AI, along with protections and retraining programs for workers
- Humans, not opaque autonomous systems, must make decisions about the use of lethal force
- Support for teachers and students to engage with new technology in responsible, critical, and creative ways
- Transparency and accountability when algorithms are used to make decisions around hiring or access to services
- Development of more environmentally sustainable AI technology
These proposals align closely with the demands of EU regulators under the AI Act and NIST initiatives. The difference is that the encyclical frames them as moral requirements, not technical ones — which gives them a different weight in the public debate.
Tech industry at the Vatican
Christopher Olah, co-founder of Anthropic — the company behind the Claude model — was present at the document's presentation. This is a revealing presence: Anthropic has consistently positioned itself as a safety-first AI company, making it a natural dialogue partner for the Vatican.
Politico reports that representatives from Amazon, Meta, and Google met with Vatican officials before publication. The goal: to influence the shape of Church positions before they become part of the global regulatory debate. The fact that the world's three largest technology companies treat the Vatican as a significant player in the AI governance debate is itself noteworthy.
The encyclical does not explicitly mention AGI (artificial general intelligence), though a subset of the AI community has been attempting to shape the pope's views on that frontier.
Why it matters
The Leo XIV encyclical can be dismissed as a religious voice in a technical debate — or taken seriously as a signal of a broadening coalition against unregulated AI. The Catholic Church has 1.4 billion members. Its moral framework has real influence on policy in dozens of countries across Latin America, Africa, and Central Europe, where AI regulation is still being formed.
The timing of publication is also significant. The encyclical arrives three months after the EU AI Act took effect and weeks after a wave of AI-driven layoffs (ClickUp, Intuit, among others). The papal position reinforces the narrative that AI creates real social costs requiring systemic responses — not merely individual adaptation.
The proposal to 'disarm' AI — understood as stripping technology of its pretense to an automatic right to govern — is philosophically consistent with accountability debates in academic and regulatory circles. Leo XIV doesn't introduce a new language, but gives existing arguments global reach and moral authority.
What's next
- The Vatican announced a series of inter-religious conferences on AI for 2026–2027 to develop the encyclical proposals into specific policy recommendations
- The European Commission is monitoring the encyclical's influence on the debate around Article 22 of the AI Act (automated decisions with consequences for individuals) — possible tightening of provisions in the 2027 revision
- Anthropic and other companies present at the Vatican will face pressure to publicly address the encyclical's proposals, particularly on autonomous weapons systems
