Pope Leo XIV will publish his first papal encyclical, 'Magnifica Humanitas,' on May 25, 2026, a document squarely aimed at the rise of artificial intelligence and its threat to human dignity. The encyclical, subtitled 'On the Protection of Human Dignity in the Age of Artificial Intelligence,' marks the most direct engagement yet between a sitting pope and a frontier AI lab. That lab is Anthropic, whose co-founder Christopher Olah will speak as a lay presenter at the launch event — a first for a papal encyclical rollout.
Why the encyclical now
The Vatican says the document responds to AI’s rapid reshaping of labor and society. Pope Leo XIV signed the encyclical on May 15, 2026, exactly 135 years after Pope Leo XIII's landmark Rerum Novarum, which addressed workers' rights in an industrial age. The timing draws a deliberate parallel: just as that encyclical tackled the dignity of labor in the 19th century, this one confronts the automation of work and decision-making in the 21st.
It's not just a statement. On May 16, the pope approved a new Vatican commission on AI ethics, pulling representatives from seven dicasteries. That commission will likely shape how Catholic institutions — from hospitals to schools to charities — adopt or reject AI tools. The scope suggests the Vatican sees AI not as a niche tech issue but as a moral question touching governance, education, health care, and human rights.
An atheist at the Vatican
Christopher Olah leads interpretability research at Anthropic, the company behind the Claude AI model. He also describes himself as an atheist. Yet he'll address the gathering at the encyclical's launch, alongside the pope, who will then offer a final blessing. That breaks Vatican precedent: encyclical launches typically feature only cardinals and theologians. Inviting a non-believing AI researcher signals a willingness to engage directly with the tech industry's top minds, not just talk past them.
Olah's research focuses on making AI models transparent — understanding what they 'think' and why they produce certain outputs. That aligns neatly with the encyclical's theme of protecting human dignity: if AI systems are black boxes, it's hard to hold anyone accountable for their decisions. The pope appears to be leaning on technical expertise to ground his moral arguments in real engineering challenges.
The commission and what comes next
The new AI ethics commission pulls from seven Vatican dicasteries — the departments that run the Holy See's operations. That breadth means it won't be a one-off advisory board. It's a permanent body charged with translating the encyclical's principles into practical guidelines. The pope signed it into being the day after he signed the encyclical itself, suggesting a coordinated push.
What remains open is how the commission will interact with governments and companies. The fact that Anthropic's co-founder is speaking at the launch hints at ongoing dialogue, but the Vatican hasn't announced any formal partnerships. The encyclical itself will set the tone: how strongly does the pope condemn unregulated AI? Does he call for a global treaty? Those specifics will land on May 25.
For now, the world's first American pope is using his first major document to draw a line in the sand. The launch event, with an atheist AI researcher sharing the stage, is the most visible sign yet that the Vatican intends to be a player — not a bystander — in the debate over artificial intelligence.