Anthropic's Mythos and Project Glasswing found thousands of high-severity software vulnerabilities, and the finding that matters is what happened next: the bottleneck shifted from discovery to human triage and patching. Governments are moving fast in response to AI capability, with the US reportedly requesting $9 billion for GPUs and classified inference infrastructure.
Pope Leo XIV issued Magnifica Humanitas, a formal papal document that rejects AI personhood, defends human labor and dignity, and calls for data governance frameworks that keep humans as the central metric of success. This is not a vague spiritual statement. It is a policy position from an institution with 1.4 billion followers.
The piece is worth reading in full because the three threads, vulnerability discovery at scale, government infrastructure buildout, and a major institutional framework for AI ethics, are happening simultaneously and shaping each other. The Pope's document in particular deserves close reading for what it demands, not just what it opposes.
[WATCH ON YOUTUBE →]