Version 1

Pope Leo XIV published 'Magnifica Humanitas' on May 25, 2026—the Catholic Church's first encyclical devoted to artificial intelligence—framing AI development as a civilizational binary: 'Babel,' where AI concentrates power and enforces uniformity, or 'Jerusalem,' where technology is co-built for the common good.[3][1] The document calls for explainable AI in consequential decisions, data governance carrying social obligations, and regulation preventing power concentration.[3][6] A secondary controversy surrounds the revelation that Anthropic co-founder Christopher Olah influenced the document, prompting accusations that a tech company effectively used the Vatican to canonize its own product framing.[4] The Verge further reports evidence that AI-generated text may appear within the encyclical itself.[8]

The Vatican's entry into AI governance carries moral weight across roughly 1.4 billion Catholics and introduces a Babel/Jerusalem vocabulary that reframes AI development as a spiritual and democratic referendum rather than a technical or economic one. The Anthropic connection raises a harder question that extends beyond this document: whether the world's most influential moral institution can serve as an independent check on the tech industry, or whether it is being instrumentalized as a legitimacy engine for particular companies' worldviews.

On May 25, 2026, Pope Leo XIV released 'Magnifica Humanitas,' positioning the Catholic Church as a direct participant in debates over AI governance.[1][2] The encyclical's organizing metaphor is stark: AI can become Babel—a system that produces uniformity, erases local culture, and concentrates decision-making among a narrow elite—or Jerusalem, a shared rebuilding where every community contributes to and owns a piece of the future.[3] The Pope's foundational claim is that technology is 'never neutral,' because it absorbs the values of those who devise, finance, regulate, and deploy it.[3] On practical governance, the document argues that any AI system affecting jobs, credit, or access to services must provide understandable reasoning and mechanisms for appeal, and that data itself should carry social obligations including worker protections and participatory rights.[3]

The encyclical dropped into a charged context. Goldman Sachs projects annual AI infrastructure spending will reach $800 billion by end of 2026, and data centers have become flashpoints in communities contesting land use, power consumption, water rights, tax deals, and noise.[3] At the same time, the U.S. Department of Homeland Security and FBI have compiled over 1,000 pages tracking 'anti-technology extremists'—a designation critics warn could expand to cover peaceful protesters and local residents opposing data center construction.[3] The Pope's test—whether ordinary people believe the future is being built alongside them, not around them—lands directly on this fault line.[3]

The document's independence came under immediate challenge when tech commentator Corey Quinn, amplified by Simon Willison, pointed to Anthropic co-founder Christopher Olah's reported influence on the text.[4] Quinn's formulation was blunt: getting the Pope to canonize a specific company's framing of its own product limitations as a spiritual treatise is 'the single greatest act of vendor lobbying I have ever seen.'[4] The charge implies that Anthropic's safety-oriented positioning—treating certain AI constraints as principled rather than merely technical—has been laundered through religious authority into something harder to contest. Whether this represents legitimate intellectual cross-pollination or a conflict of interest remains unresolved.

Reception has been varied. The Atlantic characterized the encyclical as presenting an 'unsettling vision'[5]; CBC focused on its concrete regulatory prescriptions[6]; The Neuron offered the most sympathetic deep-dive, connecting it to data center politics and the broader question of AI's democratic legitimacy[3]; while Transformer News published a critical counter-analysis arguing the Pope got something substantively wrong.[7] The Verge added a meta-layer: it investigated whether AI tools were used to draft portions of the document warning about AI, citing a suspicious pangram as textual evidence.[8] A companion website, encyclical.ai, launched tools for humans and AI agents to engage with the text directly.[9]