Pope Leo XIV's First Encyclical Is About AI - Anthropic's Chris Olah Shares the Vatican Stage

Pope Leo XIV yesterday unveiled Magnifica Humanitas, a 235-page encyclical representing the Catholic Church's first comprehensive theological position on artificial intelligence. The document was presented personally by the Pope at the Vatican, departing from the usual practice of having a cardinal deliver such statements. Anthropic co-founder Chris Olah, who leads interpretability research at the AI company, was invited to share the stage and deliver remarks alongside the papal presentation. This unprecedented collaboration between the Vatican and a leading AI researcher highlights the growing intersection between technology and religious institutions. Olah delivered notably candid observations about the AI industry, acknowledging that frontier labs, including his own Anthropic, operate within incentive structures that can conflict with ethical considerations. He called for greater involvement from individuals outside these commercial pressures who prioritise societal wellbeing and are willing to voice difficult truths. The Anthropic researcher characterised AI models as entities that are 'grown rather than built', emphasising their mysterious nature even to those who develop them. He suggested these systems exhibit patterns that mirror human emotions such as joy, fear, and grief. Pope Leo XIV responded positively to Olah's remarks, noting that despite their differences, meaningful dialogue remains possible between diverse perspectives on AI development.