Pope Leo XIV's AI encyclical: a Catholic writer unpacks the pope's message to Silicon Valley

Pope Leo XIV's AI encyclical

Pope Leo XIV's Magnifica Humanitas, released on Memorial Day, was written with Silicon Valley in mind. That was a deliberate choice. Christopher Hale, founder of the Substack Letters from Leo and a former Catholic outreach lead for the Obama campaign, argues the document landed — Jack Dorsey tweeted the full encyclical the morning it dropped.

The pope who wrote it is not a distant theologian. At 70, Leo XIV is the youngest pope in forty years, the first to own a cell phone, send an email, and wear an Apple Watch. He plays Wordle daily and consumes Western media. Hale says that context matters for reading the document: Leo understands AI is inevitable. The encyclical doesn't argue against it.

The central concern

The document's core argument is that human responsibility must remain in the loop. Leo is not a doomer on AI broadly, but he is alarmed by any system design where a decision is made and no human can be held accountable for it. His most pointed example is autonomous weapons. The first bombing of Gaza killed 168 schoolchildren in a single strike, and Leo received a letter from their parents. He finds that horrific; he would find it more horrific still if no human decision-maker had authorized it.

The encyclical's frequently quoted line captures the wider argument: for an algorithm, an error is a flaw to be corrected, but for a person, an error can be a catalyst for profound change. Leo is skeptical of the drive to eliminate human limitation rather than dignify it. Hale frames this as the pope's sharpest implicit critique of Silicon Valley's optimizing instinct — the idea that every problem is solvable and every constraint is temporary. In Catholic theology, limitation is not a bug. It is what proves the need for something beyond the self.

“What he wants Silicon Valley to keep in mind is this question of the human person, whether our projects are advancing the dignity of the human person or whether they are not... He's especially concerned about any time where human responsibility is abdicated to machines, and there's no one that you can look back to.”

On GDP and labor

Leo invokes his predecessor Leo XIII, whose 1891 encyclical Rerum Novarum supplied the intellectual foundation for labor unions and the 40-hour work week. The parallel is intentional. Magnifica Humanitas questions whether GDP is the right measure of progress — a framing Hale notes is arriving from multiple directions simultaneously, citing economists like Kyla Scanlon's work on the gap between economic growth and consumer sentiment.

The Catholic answer to what should replace pure output measurement is leisure — not as idleness but as recreation in the literal sense: time for family, rest, and renewal. The church's practical ambition, as Hale describes it, is modest and concrete: that working 40 hours a week should be enough for one person to support themselves and their dependents.

Anthropic, Christopher Olah, and the Vatican's decade-long outreach

Christopher Olah's presence at the Vatican event was the result of roughly a decade of engagement between the Holy See and Silicon Valley, engagement that Hale says was periodically rebuffed. Anthropic became the company the Vatican chose to platform, in part because it took those conversations seriously, and in part because its public friction with the Pentagon in February demonstrated, in the Vatican's reading, a willingness to draw ethical lines on defense applications.

Hale says Olah's remarks struck people in Rome. Olah acknowledged openly that Anthropic is driven by innovation and profit, not by the ethical principles the Vatican was raising — a candor Hale describes as something close to a public confession. The implicit ask was that outside institutions serve as advocates for human-centered AI development, even where the companies themselves cannot guarantee it internally.

On the question of whether Claude or any LLM could be considered "a child of God" — a question Anthropic reportedly put to 15 faith leaders in March — the Catholic answer is unambiguous: human dignity is not reducible to behavioral mimicry. You can replicate a human being's outputs at 99% or 100%, but the theological claim is that personhood is not a property that can be instantiated in a lab.

Regulation and international governance

Leo backs international regulatory frameworks and coordinated constraints, which puts him in direct tension with the dominant posture in Washington. Hale notes that JD Vance's Paris speech in early 2025 called for more focus on innovation and fewer guardrails. Leo's answer to that, as Hale reads the encyclical, is the opposite: build the guardrails alongside the technology, and do it through agreed international norms.

Some in Silicon Valley will call that naive, particularly given competitive pressure from China. Hale doesn't fully dismiss that objection, but argues Leo's ask is narrower than it sounds — not a specific regulatory architecture, but a genuine engagement with the question of whether what is being built serves the common good. The Catholic church prescribes principles, not policies. What those principles require of any specific product or law is left to policymakers and builders to work out.

Whether Silicon Valley engages seriously or files it as decel noise is, in Hale's view, still open.