The Vatican just published 42,000 words about AI, and the real story is what it buried in the other 41,500.
The Summary
- Pope Leo XIV released his first encyclical, framed around AI but actually diagnosing concentrated power, eroding democracy, and tech elites shaping the world to their advantage
- The document calls for AI to serve humanity rather than disempower it, though critics say achieving that vision will be tough
- AI functions as a lens for older structural problems, not as the primary subject of Vatican concern
The Signal
Pope Leo XIV used artificial intelligence as the hook for his first encyclical, but the 42,000-word document spends most of its pages on something more fundamental. The Vatican isn't worried about AGI timelines or alignment problems. It's worried about who holds the keys to the machine, and what they're building while the rest of us watch.
The encyclical frames AI as a diagnostic tool for examining concentrated corporate power, democratic erosion, and a tech elite that increasingly determines the shape of human life without democratic input. AI didn't create these problems. It's making them impossible to ignore.
"The Vatican isn't providing a technical framework for AI safety. It's asking who gets to decide what safe means."
This matters because the Catholic Church is one of the few institutions old enough to pattern-match on concentrations of power across centuries. When the document calls for AI to serve humanity rather than disempower it, it's not making a technical argument. It's making a political one about who builds the infrastructure of the future and whose interests it serves.
The encyclical arrives as AI companies move from research labs to infrastructure providers. The same week the Vatican published, Anthropic expanded enterprise contracts and OpenAI negotiated data deals with three more governments. The Pope isn't talking to the engineers. He's talking to everyone else, the people watching technical decisions become fait accompli before public debate even starts.
Key tensions the encyclical surfaces:
- Corporate AI development velocity vs. democratic deliberation speed
- Technical efficiency metrics vs. human flourishing as an outcome measure
- Concentrated decision-making by tech elites vs. distributed sovereignty over digital infrastructure
Critics note the vision is ambitious but light on mechanisms. The document describes the world it wants, human agency preserved, democratic input meaningful, power distributed. It's less clear on how a 42,000-word encyclical moves those outcomes forward when the companies building AI infrastructure operate on quarterly cycles and the regulatory bodies tasked with oversight are staffed by former employees of those same companies.
The Implication
The encyclical won't change what AI labs build this quarter. But it does something more interesting. It reframes the AI conversation from technical safety to structural power. That shift matters because technical safety debates stay inside the industry. Structural power debates go wider.
Watch for this document to surface in regulatory hearings, labor negotiations, and international AI governance talks. Not because it offers policy specifics, but because it gives non-technical stakeholders language to articulate what they're seeing. When your priest can explain concentrated AI power in a Sunday homily, the conversation stops being technical and starts being political. That's when things get harder for the companies who prefer the technical frame.