CIA chief likens cutting-edge AI to nuclear weapons

What happened

CIA Director John Ratcliffe compared the capabilities of advanced AI models to digital nuclear weapons during a speech at the AWS summit in Washington. This came after the Trump administration forced Anthropic to cut off access to its two most powerful models, Mythos 5 and Fable 5, by imposing export controls; Mythos was later partially restored for a restricted circle of U.S. partners, while Fable 5 remains offline. OpenAI simultaneously launched GPT-5.6 with very limited access and agreed to let the U.S. government vet authorized partners on a client-by-client basis.

Why it matters

The U.S. government is treating frontier AI models as strategic national security assets comparable to nuclear technology. These export controls and vetting agreements mark the first time a government has forced withdrawal of a frontier AI model, establishing a de facto licensing scheme that critics have flagged. For companies and developers, this signals that access to the most powerful AI systems will be subject to government oversight and restricted to approved partners.

What to watch

AWS announced a $1 billion（約1600億円） credit program for U.S. intelligence agencies and a classified cloud service for American defense contractors. The analogy between cutting-edge AI and nuclear weapons has become increasingly common in U.S. national security circles, with multiple think tanks describing a technological "arms race" between the United States, China, and Russia.