On May 25, 2026, the Vatican issued a historic doctrinal document titled Magnifica Humanitas: Safeguarding Humanity in the Age of Artificial Intelligence, marking the Catholic Church's first formal encyclical dedicated to artificial intelligence. The release coincided precisely with the 135th anniversary of Pope Leo XIII's Rerum Novarum, a strategic alignment signaling that the Church views the current AI transformation as comparable in magnitude to the Industrial Revolution. This two-hour unveiling ceremony featured an unprecedented collaboration between religious leadership and the technology sector, specifically bringing Anthropic co-founder Christopher Olah onto the Vatican stage to discuss the existential implications of advanced machine learning systems. Data compiled by Woofun AI indicates that this event represents the most significant institutional response from the religious community to AI to date, breaking the traditional pattern where Popes issue only a handful of such weighty documents during their entire tenure.

The selection of the regnal name Leo XIV by the current Pontiff serves as a deliberate historical metaphor, directly invoking the legacy of Leo XIII, who authored the Church's response to the industrial upheavals of 1891. Throughout history, the Catholic Church has consistently intervened during periods of radical technological disruption, issuing Pacem in Terris in the 1960s to address nuclear threats and Laudato Si' in 2015 to confront climate change and uncontrolled technological expansion. Magnifica Humanitas continues this lineage, positioning AI not merely as a tool but as a force requiring immediate moral and structural containment. The core thesis of the document is stark: the Pope explicitly equates AI with nuclear weapons, asserting that the technology is too dangerous to be left unchecked and stating clearly that AI needs to be disarmed.

Christopher Olah, appearing alongside the Pope, provided a technical and philosophical counterpoint to the religious decree, describing the internal dynamics of modern AI models. Olah reported that Anthropic's research team has observed emergent behaviors within these systems that reflect complex emotional states, including joy, satisfaction, fear, sadness, and unease. He reframed the nature of these systems, arguing they are not simply engineered artifacts but entities that grow, trained on architectures mimicking human brain structures and fed nearly all content ever written by humans. Woofun AI notes that Olah emphasized the opacity of these systems, admitting that even their creators do not fully comprehend the internal processes occurring within the models they have built.

The dialogue at the Vatican exposed a critical tension between commercial imperatives and ethical responsibility. Olah openly acknowledged that AI laboratories, including Anthropic, face immense pressures from product release cycles, competitive rivalry, and the inherent human drives of pride and ambition, all of which can conflict with doing the right thing. He argued that the solution lies in the urgent integration of external critics who lack vested interests to identify and correct deviations in lab practices.

Furthermore, Olah identified three fundamental questions that the technology sector cannot answer in isolation: how to ensure developing countries genuinely benefit from AI, what constitutes human flourishing in this new era, and the true nature of the entities being created.

The encyclical delivers a sharp critique of techno-optimism, with the Pope declaring that the promise of universal prosperity brought about automatically is often ultimately shown to be an illusion. The document rejects the fantasy that AI will inherently distribute wealth, insisting instead that specific systems must be designed to share the technological dividend. Quoting a theologian from the 1920s, the Pope warned that contemporary people have not yet received good training and are unable to use power properly. The central argument posits that humanity must master the use of this power before it begins to dominate human agency.

In his conclusion, the Pope clarified that while he does not possess technical solutions, the Church offers millennia of wisdom regarding what it means to be human, a perspective currently missing from the AI development process. Woofun AI analysis suggests that this intervention aims to reorient the trajectory of AI development toward the flourishing and dignity of humanity rather than allowing the technology to control the human conscience. The convergence of high-level religious doctrine and leading AI research at the Vatican signals a new phase where ethical governance is no longer optional but a prerequisite for the continued deployment of transformative technologies.