Anthropic co-founder Christopher Olah joins Pope Leo XIV at the Vatican for the AI encyclical presentation

Filed retrospectively. Before the Specola kept its nightly watch, these were the observations that turned the telescope.

On May 25, 2026, Christopher Olah, co-founder of the AI company Anthropic, appeared at the Vatican alongside Pope Leo XIV for the presentation of Magnifica Humanitas, the pope’s encyclical on artificial intelligence. The appearance followed months of dialogue between Anthropic and the Vatican on technology and ethics.

Speaking at the event, Olah said that in studying the internals of AI models his team keeps “finding things that are mysterious, even unsettling,” including what he described as “internal states that functionally mirror joy, satisfaction, fear, grief, and unease.” He argued that “religious communities, civil society, scholars, and governments” should be involved in preventing AI from dominating humanity, and acknowledged that Anthropic operates “inside a set of incentives and constraints that can sometimes conflict with doing the right thing.”

The pairing was notable in both directions. A frontier AI lab’s co-founder stood at the launch of a papal document warning about his own industry, and a pope used his first encyclical’s platform to address the people building the technology directly, not only the faithful.

Coverage noted the tension in the arrangement, since Anthropic continues to develop increasingly capable systems while presenting itself as safety-focused. The company’s interpretability research, which Olah leads, is the source of the observations he described in Rome.