CFR President Michael Froman assesses the U.S. government’s de facto ban on Anthropic’s most powerful AI model and what it reveals about the U.S. regulatory framework for AI.