French spies break with data-analysis giant Palantir, wary of relying on US tech

France's government took another step toward ensuring "strategic autonomy" for its intelligence services with the announcement on Tuesday.

The move comes as European governments are increasingly uneasy about relying on US-controlled technologies.

"We cannot accept new strategic dependencies in the digital sphere," French Prime Minister Sébastien Lecornu said as he announced the DGSI (General Directorate for Internal Security) intelligence agency would break with Palantir.

Relying on others for access cannot continue, Lecornu said.

France must instead "build real autonomy" and "not depend on the goodwill of certain partners, who are capable of turning off the access tap" for artificial intelligence, he said.

Tech giant Anthropic abruptly cut off access to its most advanced AI models on Friday after the US government ordered it to suspend foreign nationals, citing unspecified national security concerns.

French company ChapsVision will now become the "technological foundation" for "many public agencies for their critical data processing needs", the company said in a statement.

ChapsVision, founded in 2019, is relatively small compared to American AI behemoths, reporting €200 million ($232 million) of revenue in 2025 compared to Palantir's $4.5 billion.

Its tools, which are designed to gather, prepare and analyse vast swaths of data, have also been selected by Germany's VS internal security service, according to specialist media.

For its part, Palantir said in a statement that its contract with the DGSI "remains fully in force" and that "the company will continue to support the French government wherever its solutions are needed".

The continued Palantir contract was "to avoid suffering a capacity gap in this sensitive field essential to our national security", Lecornu's office told AFP.

Read moreCombating foreign digital interference: ‘France is among the top three in Europe’

Growing wariness

The announcement followed multiple calls from across the spectrum of candidates in next year's French presidential election to reduce the country's dependence on US firms.

Palantir was co-founded by Peter Thiel, a right-wing Silicon Valley billionaire close to President Donald Trump, with support from America's CIA.

It has notably, and controversially, worked with the US government to identify undocumented immigrants or targets in the US-Israel war on Iran.

Campaign groups have warned that the company's products pose risks related to mass surveillance, infringements on individual freedoms and data protection.

Watch moreFacial recognition: ICE agents use app to scan US citizens and immigrants

But Palantir says it simply provides powerful data processing services that can help reveal nuggets of useful information in the flood of intelligence sources available to government agencies and big companies.

British lawmakers earlier this month called for the country's National Health Service (NHS) to end its contract with the American company.

The London mayor's office also blocked a bid by the British capital's Metropolitan Police to work with Palantir.

Read moreAI, Palantir & the pushback: The London mayor vs the Met Police

New funding

Lecornu said Tuesday that Paris would plough an additional €655 million of new public investment into developing the country's own AI.

The funds would flow to "infrastructure, computing capacity, research, companies and industrial sectors", he said.

But Patrick Martin, head of the heavyweight Medef business association, said the government's investments were "in line with France's very limited means, given the state of public finances".

Following an "experimentation phase", France is now moving to "the roll-out phase" of a government AI tool, Lecornu said, offering a chatbot to around a million of its 2.6 million civil servants.

Built on models from the French AI startup Mistral, the system is supposed to help in areas like speeding up certain legal cases or helping university researchers secure grants.

Ministers also want to crack down on the "off-book" use of commercial AI tools that could introduce security risks.

(FRANCE 24 with AFP)