Analysis Reveals First Papal Encyclical on AI Was Substantially Written by AI

Summary

A detailed technical analysis of Pope Leo's inaugural encyclical, 'Magnifica Humanitas,' reveals that significant portions were likely generated using artificial intelligence—specifically Anthropic's Claude model. The analysis combines multiple evidentiary approaches: linguistic pattern analysis, AI detection tools (particularly Pangram), and statistical comparison with previous papal encyclicals. The author identifies consistent markers of AI authorship across English, Italian, and other translations, with some paragraphs registering 40-100% AI-generated content while others show 0%. Notably, no AI signals were detected in historical papal encyclicals, making the finding unlikely to be a false positive.

The encyclical, released on Memorial Day 2026, addresses safeguarding human dignity in the age of artificial intelligence—making the apparent AI authorship deeply ironic. Cardinal Parolin had emphasized the document's human-centered framing, focusing on 'the human person in the time of AI' rather than technology itself. However, the analysis suggests that some contributors used AI assistance while others did not, as evidenced by significant variation in AI usage rates across different sections. The identification of Claude as the likely source is supported by both textual patterns and circumstantial evidence specific to Anthropic's model characteristics.

This discovery raises critical questions about institutional transparency, AI adoption within traditional authorities, and the gap between public AI governance positions and private AI integration—especially when the institution in question claims moral authority over technological ethics.

• AI signatures are preserved consistently across English, Italian, and other language versions—ruling out translation artifacts
• No AI signals were detected in historical papal encyclicals, indicating this is unlikely to be a false positive
• The finding creates tension between the encyclical's message about protecting human dignity and its apparent reliance on AI authorship without disclosure

Editorial Opinion

The irony is impossible to ignore: the world's first papal encyclical on safeguarding human dignity in the age of artificial intelligence appears to have been substantially written by artificial intelligence. While the Vatican's theological framework on AI ethics deserves serious engagement, the apparent AI authorship without disclosure or acknowledgment creates a credibility crisis that fundamentally undermines the moral authority of the institution's position. This reveals a critical gap between the principles leaders espouse and the tools they actually use—a gap that demands transparency and institutional integrity, especially from those claiming ethical stewardship over transformative technologies.