"The AI War Has Begun"! Shock in Europe After US Restrictions on Artificial Intelligence
The order given by Washington to the company Anthropic to restrict the use of its most powerful models sparked reactions in French politics.
On Saturday, the order taken by Washington to suspend access to artificial intelligence models for "all foreign citizens" sparked reactions in French politics, according to AFP.
France alarmed by US decision
Several important figures in the French presidential race warned of an "artificial intelligence war." They also raised the alarm about the need for independence from the United States.
The American AI giant Anthropic was forced, on Friday, by its government to cut off access to its most powerful models. Washington cited a risk to national security.
Thus, it ordered, under export control measures, that access to these models be cut off for "any foreign citizen, inside or outside the United States." Access to these models was to be interrupted including for "foreign employees" of Anthropic, according to the company statement cited by the original source.
"This sudden decision reminds us that artificial intelligence is already a major issue of national sovereignty," said Jordan Bardella, leader of the National Rally party. He echoed the opinion of other officials that "France must accelerate its support for the promising startup Mistral AI and the entire AI ecosystem."
"Since 2016, LFI (France Unbowed party) has classified digital technology as the new frontier of humanity. The party refuses to allow France to become a digital colony of the USA," argued Jean-Luc Mélenchon, leader of LFI.
Friday's decision "proves the urgent need for independence and sovereignty," even if it is also a "political settling of scores" by the US government with Anthropic, "a supporter of ethical AI," added the LFI leader, according to the source.
At the beginning of March, the Pentagon terminated contracts with Anthropic. It designated the company as a "supply chain risk." The company, whose models were the only ones accredited with top-secret status, filed a lawsuit. It claimed it was penalized for refusing to allow the use of AI for mass surveillance or autonomous weapons.
"The AI war has already begun," said Gabriel Attal, leader of the Renaissance party, raising the specter of "complete subjugation of France."
"We cannot rely on others because that makes us vulnerable, as the US decision demonstrates. Anthropic is their Strait of Hormuz," he added.
Washington "decided to subject AI development to its power logic," said Édouard Philippe, former Prime Minister of France. He mentioned that "we control neither the models nor the computing power" of AI. This is an infrastructure "as essential as electricity or the internet." Urging Europe to "wake up," the mayor of Le Havre called for prioritizing "European tech markets." The mayor also called for simplifying "standards that benefit large American tech companies," the source further states.
Interior Minister Bruno Retailleau advocated for a French and European "awakening." According to him, the decision shows that "a nation that depends on others for its technology is a nation that can be disconnected overnight." Calling for AI to be considered "part of our sovereignty," like nuclear energy, he argued for redirecting public procurement towards sovereign solutions. He also argued for "rearming our technological power," according to the same source.
US limits access to Anthropic's new AI models
Anthropic announced it would "abruptly disable" its most advanced AI models for all users, after the US government ordered it to suspend access to these models for foreign citizens, citing national security reasons.
The company received the export control directive requiring it to suspend access to Fable 5 and Mythos 5 for all foreign citizens. No concrete details were provided about the national security reasons invoked, Anthropic stated in a release, according to Reuters.
The company Anthropic understands that the government believes there is a method of circumvention, or "jailbreaking," of a safeguard that would prevent the use of Fable 5 in identifying software vulnerabilities.
The order comes precisely when a previous dispute between Trump administration officials and Anthropic, which is about to go public, was showing signs of easing in some parts of the US government.
Anthropic refused US military access to AI models
Anthropic's relationship with the government broke down this year after it refused to allow the US military to use its AI models for internal surveillance and fully autonomous weapons systems.
The government responded by placing Anthropic on a supply chain blacklist, set to take effect later this year.
The measure also marks a major escalation of US efforts to stop the AI capabilities of foreign adversaries.
For years, US export controls have focused on chips and tools that power AI, rather than restricting foreign access to AI itself.
Anthropic said the government only provided "verbal evidence of a limited, non-universal potential jailbreak."
"We disagree that the discovery of a limited potential jailbreak should constitute a reason to withdraw a commercial model deployed for hundreds of millions of people," the company stated.
The government directive and Anthropic's response highlight the growing tension between AI developers and regulators over how to assess the risks of so-called "jailbreaks" or methods used to bypass model safeguards.
Just on Wednesday, Anthropic requested stricter AI oversight from the US, including the ability to block models that pose unacceptable risks.
However, the company stated that the government's measure on Friday did not follow the principles of fair and fact-based regulation.
The Pentagon's Chief Information Officer, Kirsten Davies, said in a post on X that the Department of Defense supports prioritizing national security.
"Some things are simply more important than revenue cycles, clickbait, and pre-IPO valuation. America First. Always," Davies said.
Anthropic confidentially filed an application for a US stock listing last month, beating rival OpenAI in the race to reach public markets.
