Digital sovereignty, a democratic emergency

The loss of our digital sovereignty does not result from the thoughtlessness of citizen-users, but from a form of political unconsciousness regarding the power issues carried by industrial tools, including and particularly digital tools. This dependency, built step by step over several decades, places us today in a situation of democratic vulnerability.

Public Administration, Unwitting Vector of Dependency

When administrations, ministries, cities, departments, regions, and major institutions entrust the processing of their data to a company like Microsoft, one of the most tenacious monopolies since the origins of personal computing, they are not simply making a technical choice. They impose on society as a whole the use of these same tools, which then become a de facto standard. This normalization from above transforms an administrative decision into a collective constraint.

David Monniaux, research director at CNRS and cybersecurity specialist, reminds us that criticism of this dependency is nothing new. As early as 1998, Roberto Di Cosmo published with Dominique Nora Le Hold-Up planétaire (Calmann-Lévy and Éditions 00h00), which denounced Microsoft’s practices related to the monopolistic sale of operating systems and office software with professional and personal computers. However, Monniaux notes, this criticism “did not take hold, notably because it was not perceived as political, but rather as a ’geek’ fantasy” (Le Monde, October 23, 2025).

Attempts to break free from this dependency have often been ridiculed. The replacement of Windows with Ubuntu Linux at the National Assembly in 2007 was quickly abandoned. Microsoft, long in a quasi-monopoly position and generating considerable margins, was able to extract rent without raising any real institutional resistance. We are facing a serious political thoughtlessness, which offers no easy solution but demands work—work that even major institutions most often prefer to avoid, choosing the ease of dependency.

A Real Legal and Geopolitical Vulnerability

This situation evokes a historical comparison for me, which I hope is not too inappropriate. Until 1965, with the reform of the matrimonial regime, a French woman who was of legal age and independent became legally a minor again through marriage. She then lost her capacity for autonomous action. Our relationship with American digital infrastructures reproduces this logic: by “marrying” actors like Microsoft, we believe we gain efficiency, but we lose our sovereignty over our political choices—that is, over our real capacities for freedom.

As David Monniaux emphasizes in his op-ed in Le Monde, this concern is not at all paranoid. During the summer of 2025, Microsoft confirmed that it would hand over its users’ data to U.S. authorities if they requested it, even if that data were stored in France. This is simply the application of the American Cloud Act of 2018, which compels any American company to obey first and foremost its country’s laws, regardless of European laws. Monniaux then poses this question: “What would happen if (Trump) ordered the GAFAMs (Google, Apple, Facebook, Amazon, Microsoft) to cease their cloud services with regard to our government, our public services, our armed forces?” (Le Monde, October 23, 2025).

Some will respond that there are contracts, legal guarantees, that this hypothesis amounts to alarmist speculation. But precisely, Monniaux insists: “These are speculations, that such acts would imply the breach of contracts between these companies and their users as well as violations of French and European law.” The American government, which has already expressed its desire to occupy Greenland and whose ships suspected of transporting drugs navigate international waters, “clearly doesn’t care about law, a fortiori international or foreign law.” Just because an event has not yet occurred does not mean it cannot happen. And this type of event is unfortunately all too predictable. The question is no longer if this will happen, but when it will happen, in what context, and with what consequences.

Recent history teaches us how quickly a political system can tip over. The repeated hackings of hospitals and universities in 2020, notably Paris Saclay and Grenoble INP, as well as the blocking of administrative and financial services for months, demonstrated our vulnerability to digital threats. The reflex of certain organizations was to give up managing their own IT and outsource it to major foreign operators, in order to rely on supposedly better-secured infrastructures. But this is solutionism, an absence of foundational work, which in my view opens onto a dramatic future. Researchers are now even advised not to use institutional tools, including their official email address, for their research activities. This abdication reveals our collective inability to think long-term.

The Time of Sovereignty as Political Act

How can we regain our sovereignty? The answer lies in a cooperative approach: discussing, drawing inspiration from successful experiences, informing ourselves, accepting to dedicate time to it. The time dedicated to digital sovereignty, to safeguarding our data, to awareness of their location, must be made official. This is not useless and tedious technical time, but political time in the noblest sense of the term.

If we view this time as political time, emancipatory and empowering time, we then radically change perspective. The example of the national gendarmerie, cited by Monniaux, is eloquent: it first replaced its Microsoft office suite with a free suite, then almost all of its Windows workstations with its own Linux distribution (GendBuntu). It is one of the rare public organizations to have developed a long-term strategy in the face of this dependency. This approach demonstrates that with political will and patience, solutions based on free software, managed locally, are absolutely possible.

Barbara Cassin, in Google-moi. La deuxième mission de l’Amérique (Albin Michel, 2007), already criticized access to media increasingly controlled by the algorithms of major platforms. She approached it then from a cultural angle, the “pillaging” of heritage libraries, the use of English. Since then, the cloud has taken over, and this has gone even further, into a dependency of businesses and administrations. It is a matter of thinking together the technical, cultural, and political dimensions of digital technology.

It is this political culture of tools that we must transmit. Through this path of personal politics of sovereignty, of thinking about oneself projected into the future and the long term, we can escape the dictatorship of the present, which itself constitutes a real political issue. Monniaux concludes aptly: “It is more than time for France and Europe to react to this dependency, moreover very costly, on foreign providers who can at any moment cut off their IT infrastructures. It is incomprehensible that we persist in choices which, while they may appear more economical or simpler in the short term, commit us to long-term vulnerabilities, when solutions based on free software, managed locally, would often be possible.”

It is thus, through this awareness and this progressive reappropriation, that we will regain our grip, our footing, and our power over our future, our democracy, and our personal freedom. Digital sovereignty is not a matter for technophiles: it is one of the conditions of our collective political autonomy.