“The Pope picked a lab.” That's my line, not a headline, and I know it's a provocation. He didn't endorse a tech company, that's not a thing that happens, and encyclicals teach rather than do product placement. But it's honestly how the whole thing landed for me, and the more I read, the more the phrase held up. He handed one frontier lab the microphone and let everyone notice which one showed up.
That detail isn't small to me, because of what 2025 felt like from where I sat. I was using every model I could get my hands on, and Claude was the only one that could actually do tool use consistently, the part where the model stops talking and starts doing things, calling functions, editing files, chaining steps without falling apart halfway through. 3.5 Sonnet especially. Other models would describe what they'd do; Claude would just do it, and get it right.
I rate that model so highly it stopped feeling normal. My only half-serious theory is that someone came back from the far future, grabbed a distilled low-level coding model that was ordinary in their time, brought it to ours, and quietly made it worse, so Anthropic could pull the Apple move and ship a slightly better version every few months. I don't actually believe that. But the fact that it's the explanation my brain reaches for tells you how far ahead it felt. So when Anthropic was the lab standing on that stage and the others weren't, it didn't read as random to me. The choice was the statement.
What the letter actually says
Magnifica Humanitas is Leo XIV's first encyclical, and he signed it on May 15, 2026, exactly 135 years to the day after the Church's famous answer to the Industrial Revolution. Same date, same papal name, on purpose. The message: AI is this generation's version of the machine that rewrote what human work is worth.
It's blunter than I expected a Church document to be. It argues technology is never neutral, that it carries the values of whoever builds and funds it, and it asks for real regulation and independent oversight instead of vague promises to “be ethical.” You can read Anthropic's own account of the day if you want their framing of it.
We need moral voices that the incentives cannot bend.
It's a great line. A 33-year-old atheist co-founder standing next to the Pope, telling the world to trust the labs less and watch them more. And I want to believe it. But I'm going to be honest about where I land: it's still a company, on a stage, managing how it's seen. “Watch us more closely” is a wonderful thing to say when being watched closely is exactly the reputation you want. I think the sentiment is real and I think it's good PR. Both can be true, and I'd rather keep one eye open than clap.
Who actually gets this
The encyclical worries that AI's gains will pool in a few rich countries. In my last post I made the optimistic case, that AI levels the playing field for people who could never afford a tutor, an editor, a developer. I still believe it. But the honest, grown-up version of that idea is that access is becoming a map.
The frontier models cost more every year. At the same time, cheap and capable ones like DeepSeek are rising fast. So here's my actual prediction: once African countries can host and serve cheap models at a good price, China's low-cost offerings get very attractive there. Europe leans toward the American labs. The Americas get ChatGPT, Codex, and Claude. “Shared globally” is a beautiful phrase, but in practice it's going to be drawn along lines of who hosts what, for how much, and where. That's the part the headlines skipped, and it's the part I can't stop thinking about.
Does the Pope even belong here?
My answer is yes, and I don't have to reach for a reason, because I'm already living it. I'm building Bible apps that use AI to translate scripture so it isn't locked to English, and to break passages down so people actually understand them instead of being handed a book and told to figure it out alone.
That is the whole thing the encyclical is about: technology pointed at human dignity instead of at concentrating power. When a 1,400-year-old institution says AI can't be ignored, it isn't overreach. Faith communities are already using these tools to reach people who were being left out. The Church showing up to the conversation is just it admitting out loud what builders like me already figured out quietly.
So I'll keep building. Skeptical of the polished lines on the stage, serious about the work underneath them. The tools are turning into a map of who gets access to what, and I'd rather spend my time making sure the people on the edges of that map are on it, one app at a time.