A push for external AI governance is gathering momentum as pope-led discourse and major AI researchers align on the limits of self-regulation. In a new intervention, Pope Leo XIV’s encyclical Magnifica Humanitas argues for AI to be disarmed and regulated in the service of humanity, while Anthropic co-founder Christopher Olah and Anthropic CEO Dario Amodei have warned that AI companies face incentives that conflict with public safety. The commentary argues that internal governance frameworks cannot solve collective-action failures in a competitive frontier model marketplace where unilateral restraint can be punished. It also frames the analogy for governance as governance architecture rather than technology-only fixes. For universities, the message raises fresh pressure on research ethics training, AI procurement rules, and student-facing academic integrity policies as AI capabilities outpace institutional oversight.