Pope Leo XIV's Encyclical May Have AI-Written Sections

It's becoming increasingly common for AI to find its way into unexpected places, and the latest instance involves the Vatican. A recent analysis by Linch Zhang, posted on the forum LessWrong, suggests that portions of Pope Leo XIV's latest encyclical, titled 'Magnifica Humanitas,' may have been authored by AI.

The encyclical, which focuses on the impact of AI on humanity, has drawn attention not just for its subject matter but for its potential origins. According to Zhang's findings, popular AI detection tool Pangram identified certain paragraphs as being between 40 percent and 100 percent AI-generated. This is a significant development, blurring the lines between human and machine authorship in a highly influential context.

Zhang's analysis points to specific linguistic markers that are often associated with AI writing, particularly from Anthropic's Claude models. One such marker is an increased usage of the word "genuinely," a frequency that reportedly surpasses that found in previous papal encyclicals. This subtle linguistic shift could be a key indicator of AI involvement.

Further testing by another individual who ran sections of the document through Pangram confirmed these suspicions. The results indicated that a substantial 62 percent of the first chapter might be AI-written. While the Vatican has not officially commented on these findings, the possibility of AI contributing to such a significant document raises profound questions about authorship, authenticity, and the future role of artificial intelligence in even the most traditionally human domains.

This situation is a fascinating case study for us tech enthusiasts, showcasing how AI is not just a tool for creating content but is now potentially co-authoring deeply meaningful texts. It's a stark reminder that as AI capabilities grow, discerning the source of information and creative output will become an ever more complex challenge.