AI is not just a technology: it is a choice of civilization. Europe must master its tools to better regulate them, strengthen human judgment and reconnect with the spirit of the Enlightenment.
The citizen consultation recently launched by the Artificial Intelligence and Digital Council is, in many ways, a welcome initiative. Finally, it recognizes that artificial intelligence and digital technology are not simple technical subjects reserved for engineers or businesses, but deeply political, social and human questions.
Employment, education, environment, transformation of work, protection of minors: these themes directly concern the daily lives of citizens. Opening up a space for expression is useful, both to highlight realities on the ground and to strengthen the democratic legitimacy of future public orientations.
However, this approach is only relevant on one condition: that the contributions collected do not remain a dead letter. The risk of a purely symbolic consultation always exists. Citizen participation only has value if it really sheds light on decisions, if it is synthesized, made public and translated into decisions. Without it, the democratic promise turns into frustration.
But the major interest of this consultation perhaps lies elsewhere: it opens the door to a more fundamental question, still too rarely asked publicly.
An issue that goes beyond usage: a civilizational shift
Artificial intelligence is not just about practical applications or efficiency gains. It is gradually inserting itself into the heart of our judgment mechanisms: search engines, cultural recommendations, medical diagnoses, administrative, legal or financial decisions.
As these systems become established, their statistical calculations cease to be simple aid tools. They become operational judgments: evaluations that guide, constrain or replace human discernment. What was initially only a probability then tends to acquire a normative value, sometimes incontestable.
AI thus influences not only what we do, but how we think, prioritize and decide. It shapes our representations of reality, risk, performance and normality. In this sense, the issue is not only technological or economic: it is fully civilizational.
A historic opportunity for Europe
Faced with this shift, Europe has a unique responsibility. It can choose to undergo models designed elsewhere, or to propose another path: that of artificial intelligence included in an explicit social project.
The challenge is not to claim an illusory neutrality, but to design an AI that reasons from an accepted frame of reference: historical heritage, democratic values, fundamental freedoms, equality, solidarity, protection of the most vulnerable, respect for environmental balances.
In other words, the tool must serve the vision, and not the other way around. A European AI should not only aim for optimization or prediction, but integrate principles of responsibility, justice and sustainability. It must strengthen the capacity for human judgment, develop critical thinking and deliberation, not dissolve them in automated decisions for which no one anymore fully assumes responsibility.
Technological mastery and chosen dependencies
To achieve this objective, Europe must have a clear strategy of technological mastery and controlled dependencies. It is neither a question of autarchy nor naivety, but of strategic lucidity.
This involves securing critical infrastructure, choosing partners compatible with European political and ethical objectives, and reducing the most sensitive dependencies in the value chain.
Europe has a major asset for this: a considerable digital heritage, made up of massive, qualitative and diversified data, coming from pluralistic, regulated societies and rich in their institutions. We still need to be able to promote them within a framework of trust.
This requires controlled platforms, maximum internalization of the value chain — from data to models, from infrastructure to uses — and coherent governance of the whole.
Finally, it must be said bluntly: seeking to regulate technologies that we do not control most often means suffering their effects rather than directing them. The standard does not replace infrastructure, skills or industrial capacity. Without a systemic approach to the subject, without a deep understanding of systems, models and their real uses, regulation risks remaining declarative, circumventable or dependent on frameworks defined elsewhere and inexorably a source of conflict with technology supplier countries. Normative sovereignty cannot exist sustainably without a minimal base of technological sovereignty.
It is at this price that European artificial intelligence will be able to strengthen thinking, inform decisions and preserve the sovereignty of Member States, rather than accelerating its dilution or even erasure.
Towards a strategic data and knowledge complex
This change of scale calls for a profound transformation of our collective organization. Like military-industrial complexes, which have historically enabled States to defend their security, sovereignty and decision-making capacity in an unstable world, Europe must today equip itself with a strategic data and knowledge complex.
It is neither a question of denying the military dimension of artificial intelligence – already fully integrated into contemporary conflicts, which have become hybrid – nor of reducing digital technology to a simple instrument of confrontation. It is about recognizing a broader strategic reality: the control of data, algorithms, models and cognitive infrastructures has become a priority issue. In a world where power is exercised as much through the capacity to produce, organize and interpret knowledge as through armed force, renouncing these essential levers would amount to accepting a lasting loss of political autonomy.
This strategic complex must articulate public research, industry, regulation, training and democratic institutions within the same coherent whole, explicitly designed to support, defend and project the European civilizational project. It must make it possible to secure critical infrastructures, structure European industrial capacities and guarantee that artificial intelligence remains a tool serving human judgment, collective deliberation and the European civilizational project — and not a substitute for political responsibility.
An open and exportable model
Finally, this project only makes sense if it is open. Artificial intelligence based on democratic, social and environmental values is intended to become an exportable model, offered to societies and states that share these principles.
The current citizen consultation is only a starting point. We still need to have the ambition to transform this collective speech into a true European civilizational project.
Because artificial intelligence can be much more than a disruptive or dominant factor. It constitutes a fantastic opportunity to reconnect with the spirit of the Enlightenment, that of the 18th century, which made reason, knowledge, emancipation through knowledge and critical debate the foundations of human progress. This spirit, born in Europe, exerted a lasting global influence by affirming that technology and science should serve the autonomy of the individual and the common good.
Provided it is thought out, governed and oriented, AI can become a contemporary extension of this heritage: not an automated reason which replaces humans, but an augmented reason, in the service of understanding, deliberation and collective responsibility. By making artificial intelligence a tool of emancipation rather than an instrument of alienation, Europe has the rare opportunity to offer the world a technological model faithful to its history — and focused on the future.
