Pope Leo XIV Releases First AI Encyclical Magnifica Humanitas, With Anthropic Co-Founder at His Side

At 11:30 a.m. Rome time on May 25, 2026, Pope Leo XIV walked into the Vatican’s Synod Hall and presented his first encyclical, Magnifica Humanitas — "Magnificent Humanity" — to the world. The document, signed on May 15 to mark the 135th anniversary of Leo XIII’s landmark labour encyclical Rerum Novarum, is the first papal teaching letter devoted to artificial intelligence, and Leo broke with centuries of protocol by reading and blessing it himself rather than handing the presentation to a cardinal.

Even more striking was who stood beside him. Among the panel of presenters was Christopher Olah, co-founder of Anthropic and head of the AI safety lab’s interpretability research — the team trying to crack open the black box of large language models. The Vatican’s decision to feature a frontier-lab researcher at the launch of a papal encyclical is a public signal that Leo intends the Church to engage AI builders directly, not just regulate from the sidelines. Olah shared the panel with Cardinals Víctor Manuel Fernández, Michael Czerny and Pietro Parolin, and theologians Anna Rowlands of the University of Durham and Léocadie Lushombo of Santa Clara University.

The encyclical’s central claim is that AI “poses not a technological challenge, but an anthropological one” — a question of what it means to be human in an age of machines that can imitate human reasoning, voices and faces. Leo reserves some of his sharpest language for lethal autonomous weapons, writing that “we must keep a watchful eye on the development and application of artificial intelligence in both military and civilian contexts, to ensure that they do not absolve humans of responsibility.” The document also addresses the displacement of workers, the proliferation of synthetic personas designed to mimic real people, and the duty of wealthy AI nations to share the technology’s benefits with the global South.

The choice of Anthropic, rather than any of its larger rivals, is politically loaded. The lab has spent the past year arguing for binding safety testing of frontier models and was singled out earlier this year by the Trump administration for restricting some military uses of Claude. By inviting Olah, Leo is implicitly endorsing the camp inside the AI industry that wants firmer guardrails — a posture that aligns with the encyclical’s emphasis on accountability over speed. Coverage from CNN, PBS NewsHour and the America Jesuit review framed the document as the Church’s most consequential intervention in a technology debate since John Paul II’s writings on bioethics in the 1990s.

Magnifica Humanitas does not call for a moratorium or specific regulation, but it asks Catholics — and, by extension, the engineers, executives and policymakers reading along — to treat every deployment of AI as a question of human dignity first. With WWDC, a wave of agent launches and a 0 billion Anthropic funding round all landing in the same week, the Vatican’s timing was deliberate: a reminder that the people building these systems are now being read alongside encyclicals, and judged on the same scale.