The global AI race has entered a dramatic new phase. Anthropic has disabled access to its most advanced artificial intelligence models, Fable 5 and Mythos 5, after an unprecedented order from the Trump administration. The US government directed the company to restrict access for all foreign nationals, citing national security concerns linked to potential jailbreak vulnerabilities. The move marks the first time Washington has taken such sweeping action against frontier AI software itself, rather than hardware like chips or supercomputers. Anthropic argues that the identified vulnerability was limited and warns that applying such standards across the industry could severely slow future AI innovation. The decision could have major implications for the global AI ecosystem, affecting developers, enterprises, researchers and investors. With companies like OpenAI, Google and Meta also developing increasingly powerful models, the restriction raises a critical question: Is artificial intelligence becoming a strategic national asset similar to advanced military technology?