Pope Leo XIV 将于 5 月 25 日推出他首部聚焦人工智慧的通谕，届时的意外嘉宾是 Anthropic 共同创办人 Christopher Olah。梵蒂冈表示，Olah 这位机器学习专家将出席这场备受期待的发表会；同场发言者还包括梵蒂冈国务卿 Cardinal Pietro Parolin 与 Cardinal Víctor Manuel Fernández。

这份名为 Magnifica Humanitas 的通谕预期将阐述教宗如何在 AI 时代保护人类。Pope Leo 在一年多前成为史上首位美国籍教宗后，便把 AI 列为其任期的核心议题，并在多次讲话与公开场合反复提及。

Olah 负责 Anthropic 的 AI interpretability 工作，重点在于理解先进 AI 模型内部如何运作；他出席梵蒂冈活动，可能带来讨论 AI 安全核心问题的机会，也就是最强大的系统是否能被充分理解，从而值得信任。这份通谕将以教宗 5 月 15 日签署的日期为准，正值 Pope Leo XIII 通谕 Rerum novarum 发表 135 周年；现任教宗以 Leo 为名，正是为了向在第一次工业革命期间引导教会的 Leo XIII 致敬。

Pope Leo XIV’s first encyclical focused on artificial intelligence will be launched next week, with an unexpected guest: Christopher Olah, co-founder of Anthropic. Olah, a machine-learning expert, will join Pope Leo on May 25 at the Vatican’s much-anticipated presentation, alongside Cardinal Pietro Parolin and Cardinal Víctor Manuel Fernández.

The encyclical, titled Magnifica Humanitas, is expected to set out the pope’s views on preserving humanity in the age of AI. Pope Leo, the first-ever US pope, has made AI a central theme of his pontificate and frequently mentions it in speeches and public appearances.

Olah leads Anthropic’s work on AI interpretability, which aims to understand how advanced models operate internally. His appearance may help frame a key AI-safety question: whether the most powerful systems can be understood well enough to be trusted. The encyclical is dated May 15, marking the 135th anniversary of Pope Leo XIII’s Rerum novarum, and Pope Leo chose his name to honor Leo XIII and his role during the first industrial revolution.