US blocks foreign access to Anthropic's top AI models, pushing AI sovereignty into the French presidential campaign
Washington ordered Anthropic to shut off access its most powerful AI models for all foreign nationals, immediately drawing France's presidential contenders into a charged debate on digital sovereignty.
The order
On Friday 12 June, the US government ordered Anthropic to suspend access to its most advanced AI models, Fable 5 and Mythos 5, for "any foreign national, inside or outside the United States," including the company's own foreign employees. The directive was issued under export controls, with Washington citing national security risks. Anthropic disclosed the move in a corporate statement, calling it a step without precedent.
The US government has directed us to suspend access for any foreign national, inside or outside the United States, to our most powerful models.
A pre-existing rift
The order did not come from a vacuum. In early March, the Pentagon cancelled its contracts with Anthropic, labelling the firm a "supply chain risk." Anthropic had been the only AI company with models accredited for classified defence work. It sued, alleging it was punished for refusing to allow its AI to be used for mass surveillance or autonomous weapons.
- Pentagon cancels all contracts with Anthropic, designating the company a supply-chain risk.
- US orders Anthropic to suspend access to its most powerful models for all foreign nationals.
- French presidential candidates react, turning the order into a sovereignty debate.
France's political class reacts
On Saturday 13 June, heavyweights from across the French presidential field seized on the decision. Jordan Bardella of the Rassemblement National called it proof that AI is already a major national sovereignty issue.
This sudden decision reminds us that artificial intelligence is already a major national sovereignty issue. France must accelerate its support for the gem Mistral AI and the entire AI ecosystem.
Jean-Luc Mélenchon, of La France Insoumise, framed the decision as a wake-up call for digital independence while also denouncing it as a political score-settling by the US government against a company known for its ethical-AI stance.
Since 2016, LFI has classified the digital realm as the new frontier of humanity and refuses that France be a digital colony of the USA. This decision proves the urgency of being independent and sovereign.
A warning of vassalisation
Renaissance candidate Gabriel Attal declared that "the AI war has already begun" and warned of a risk of "total vassalisation of France." He drew a geopolitical parallel between the AI chokepoint and the Strait of Hormuz.
We cannot count on others because that makes us vulnerable, the United States' decision shows it. Anthropic is their Strait of Hormuz.
Édouard Philippe of Horizons said Washington was submitting AI development to its power logic, stressing that France controls neither the models nor the computing infrastructure, which he described as essential as electricity or the internet.
A campaign inflection point
The US action has abruptly pushed digital sovereignty into the centre of the French presidential race. Across the political spectrum, candidates are now invoking the need to back domestic AI ventures, especially Mistral AI, and to reduce dependence on American-controlled AI infrastructure. What began as an export-control manoeuvre has become a defining issue in a national election.
