France has decided to replace the services of the American data analytics firm Palantir, which has been used by its intelligence agencies, with products from a domestic company. This shift comes amid rising concerns about entrusting critical national information to American technology.
On June 16, French Prime Minister Sébastien Lecornu announced that the General Directorate for Internal Security (DGSI) will use the data analytics products of the French company ChapVision instead of Palantir. Lecornu stated, "We cannot accept new dependencies in the digital realm," emphasizing the need to establish true autonomy.
The transition is expected to occur gradually. Palantir had extended its contract with the DGSI for several years at the end of last year. The Prime Minister's office also noted that existing services will continue until ChapVision demonstrates sufficient performance. Lecornu indicated that the transition could take several years.
The French government is also accelerating efforts to foster its own AI and data industries. France plans to invest an additional €655 million (approximately $700 million) in AI development, research, and the implementation of AI in public services. The government has also decided to adopt products from the local AI startup Mistral AI.
Concerns surrounding Palantir are also growing in the UK and Germany. The UK's National Health Service (NHS) is reviewing its large-scale data processing contract with Palantir due to increased pressure from politicians and civil society organizations. The City of London has blocked a contract between the London police and Palantir, citing cost-effectiveness. Palantir has threatened legal action in response.
In Germany, there are moves to reduce reliance on Palantir, with military authorities reportedly stating they will no longer use the company's products.
Palantir maintains that concerns in Europe are exaggerated, asserting that its services are analytical tools designed to help government agencies and large corporations find necessary information within vast datasets.
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.