GOVERNMENT KILLS ANTHROPIC’S TWO MOST POWERFUL AI MODELS WORLDWIDE IN UNPRECEDENTED NATIONAL SECURITY CRACKDOWN The U.S. government dropped a bomb on Anthropic late Friday, ordering the AI company to immediately shut off access to Claude Fable 5 and Claude Mythos 5 for all users worldwide. Not just foreign nationals. Everyone. The directive arrived at 5:21 PM ET and Anthropic complied, though it made clear it thinks Washington got this badly wrong. Here is the backstory. Mythos is Anthropic’s most capable model, so powerful at finding security vulnerabilities that the company never released it publicly. It found flaws in every major operating system and web browser it tested. Fable 5, released just three days before the shutdown, was Anthropic’s answer to commercial pressure: a guardrailed version of Mythos finally deemed safe enough for general use. Benchmarks ranked it the most capable AI model available anywhere at the moment of its release. The government’s stated concern is a claimed jailbreak of Fable 5. Anthropic disputes this characterization strongly. It says the underlying capability is already present in OpenAI’s GPT-5.5 and is used routinely by cybersecurity professionals for legitimate defensive work. Anthropic published a lengthy blog post making clear that if this standard were applied industry-wide, it would essentially halt all new model deployments from every frontier AI provider. The irony is brutal. Anthropic spent months telling the world its models were uniquely dangerous. The government listened. Sam Altman had called it fear-based marketing. Now Anthropic’s own safety messaging has become the most serious threat to its business, right as the company prepares for a public offering that could value it near one trillion dollars.