Washington has banned foreign nationals from accessing Anthropic's two newest AI models, and some in tech say the company's own public warnings helped make it happen.
Last week, the US government barred foreign nationals from using Mythos and Fable, Anthropic's latest models. An FT analysis of Anthropic's 2026 public output — statements, social posts, and articles by the company and its CEO Dario Amodei — found that five in every 1,000 words touched on risk, regulation, or restrictions. The equivalent rate for OpenAI and Sam Altman was roughly eight times lower, at 0.6 per 1,000. Critics in the tech community now argue that Anthropic's vocal stance on AI danger, particularly concerning Mythos, shaped the political climate that produced the ban.
Anthropic, valued at $965 billion, built its identity around being the cautious alternative to Silicon Valley's move-fast culture. If its own language helped trigger a government restriction on its flagship products, it faces an uncomfortable irony: the safety messaging may have invited the very regulatory action it spent years warning others about. The full scope of the restriction — which products, access tiers, and markets are affected — was not publicly specified; requests for clarification sent to Anthropic and US officials were not answered.
OpenAI, which says roughly eight times less about AI risk in public, currently faces no equivalent ban on its models abroad — a quiet competitive edge nobody at either company is eager to discuss openly.