On May 25, Pope Leo XIV released Magnifica Humanitas — “Magnificent Humanity” — his first major encyclical, and the entire document is focused on artificial intelligence. This isn’t a vague moral statement. At over 42,000 words, it’s one of the most detailed critiques of AI’s social impact ever written by a major world institution.
Leo’s headline: AI must be “disarmed.” He wasn’t talking about science fiction — he was talking about autonomous weapons, algorithmic manipulation of elections, and AI systems replacing human judgment in decisions that can’t be undone.
What he actually said to builders
Leo made a direct appeal to AI developers specifically: “Those who create AI bear a unique ethical and spiritual obligation, as every design decision reflects a vision of humanity.” He also called out the concentration of power among a small group of private companies shaping the rules.
His biggest demand wasn’t philosophical — it was structural. “Ethics talk is not enough. We need strong legal frameworks, independent oversight, and informed users.”
Why this matters if you’re building
The Vatican publishing a 42,000-word AI document — presented alongside Anthropic co-founder Christopher Olah — signals that AI regulation pressure is coming from every direction now, not just governments. If you’re building AI products, the window for self-regulation is shrinking. Start thinking about transparency, explainability, and accountability before they’re required.