The Vatican's AI encyclical lands harder than expected

Pope Leo XIV released a 42,300-word encyclical called 'Magnifica Humanitas' focused substantially on AI, and HN ran at least five separate threads on it, including one featuring Anthropic co-founder Chris Olah's public response. The document's core argument is that AI must be 'disarmed' of the assumption that technical power automatically confers the right to use it. Olah's response, posted on HN, described AI systems as 'made from us, from our words,' which is a notably non-technical framing from someone at the frontier of AI research.

The Vatican-Anthropic connection is being discussed separately in another thread noting that the relationship is 'reshaping the AI ethics debate,' though the HN comment section was skeptical, calling it theater. What's notable is not whether the ethics debate is genuine but that Anthropic is investing in positioning itself as a responsible actor at the exact moment it is also developing Mythos-class models for automated vulnerability scanning, which another thread called out as nearly an exploit-finding tool.

The pattern here is that the legitimacy fight around AI is intensifying. Who gets to define what 'responsible AI' means, what role moral authority plays, and whether governance can move faster than capability are now live questions with real business implications for anyone building in this space.