FRANCE PERSPECTIVE

Paris reads the shutdown as the opening shot of an 'AI war' that directly threatens European digital sovereignty. The French press details the mechanics of the order — a directive received at 5:21 p.m. ET (11:21 p.m. in Paris), with no stated reasons, demanding access be cut for 'any foreign national, inside or outside the United States,' including Anthropic's foreign staff — and stresses what that formula implies: a French user, whether living in Paris or San Francisco, is now treated as a national-security risk by the country that builds the tool. The dominant angle is not the tech news item but the strategic warning. Many politicians saw an 'alarming signal': the episode shows that reliance on an American model can be severed overnight, by a single letter from the Commerce Secretary, with no notice and no recourse. This reading echoes a long-standing French discourse on strategic autonomy and distrust of the extraterritorial reach of US law — the same reflex triggered by extraterritorial sanctions laws or the Cloud Act. The press recalls that Mythos 5 is known for detecting software flaws, and Fable 5 is its public, restricted version, placing the affair in the cyber register, not just the commercial one. For Paris, the episode validates the case for European players like Mistral: the question is no longer whether the best model is American, but what happens the day Washington unilaterally decides to shut the door.