The single market for data is not a technical detail, it is a strategic precondition for Europe’s digital sovereignty and AI competitiveness. Europe now faces a clear choice: move beyond convergence through dialogue toward convergence by design. The next step in Europe’s digital agenda is not more principles, but shared market rails that make trusted data exchange possible at scale. Data space interoperability is the missing link between Europe’s commitment to trust and its ambition for scale. Without it, Europe may succeed in creating multiple data spaces, yet still fail to build a genuine single market for data & AI.

“ Turning Data Space islands into a European Data & AI continent.”

Europe needs a ‘GSM moment’ for data & AI: a political and industrial commitment to a minimal, enforceable interoperability core that is common, measurable and reusable across sectors and Member States.

Europe will not win the next phase of the digital and AI economy by copying the raw-scale model of foreign platforms. Its comparative advantage lies in its regulation and partnership culture, and in governed access to high-value private & protected data. But that advantage becomes economic and geopolitical leverage only if identity, catalogues, contracts, semantics and compliance can work across ecosystems at low cost. The real objective is therefore not more data spaces as such, but a Data & AI Union built on coordination without centralisation: many actors, many implementations, and shared rails.

Europe already has the rules and the building blocks. The choice now is whether it remains  a power of regulation or becomes a power of implementation. If interoperability becomes a real market condition, Europe can turn trust into transactions, transactions into scale, and scale into strategic autonomy.

« Europe can shift from sovereignty by regulation, to strategic autonomy by shared standards »

Discover the thirteen proposals developed by the contributors to this collective report.

This report by Singapore-based expert Damien Kopp analyses 25 national AI strategies and reaches an unequivocal conclusion: today’s pursuit of autonomy is being built on borrowed technologies. We have entered the era of the “sovereignty paradox.” The more governments and companies invest in building their own AI, the more they reinforce their structural dependence on a small number of foreign suppliers for chips (GPUs), cloud infrastructure, and foundation models.

AI is no longer just a technology; it has become a geopolitical supply chain, comparable to energy, structured around a handful of critical chokepoints. Each link in this chain is concentrated, territorially anchored, legally constrained—and therefore potentially instrumentalised. The strategic question is no longer one of ownership, but of control over critical dependencies and the ability to continue operating in the event of disruption.

A comparative analysis of 25 countries shows that only the United States and China come close to “full-stack” sovereignty (hardware, cloud, models, data). For the rest of the world, the reality is one of managed dependency, expressed through four main archetypes:
– 1 “Full-stack / hybrid” sovereignty (e.g. United States, China, South Korea, Japan): these nations invest heavily in hard infrastructure—semiconductor fabs and national compute capacity—to reduce their dependence over the medium term.
– 2 Regulatory sovereignty (e.g. France, Canada, Germany): this is the European path. Control is exercised through law (AI Act, GDPR) over infrastructure that is not owned. It is a powerful trust-building lever, but one that does not resolve dependence on foreign chips and cloud infrastructure.
– 3 The “Open-Yet-Local” model (e.g. India, Brazil, Singapore): unable to compete on hardware, these countries focus on linguistic inclusion and control over local use cases, relying on foreign infrastructure while retaining control over the cultural and application layers.
– 4 The partnership model (e.g. United Arab Emirates, United Kingdom, Australia): these countries prioritise speed. They accept deep technological dependence on U.S. tech giants in exchange for immediate access to state-of-the-art capabilities.
None of these models fully escapes dependency. But some manage it strategically, while others simply endure it. The paradox is clear: the faster a country seeks to deploy AI, the more it may end up reinforcing the very dependencies it claims to reduce.

« AI is no longer just a technology: it has become a geopolitical supply chain »

KEY RECOMMENDATIONS
For both enterprises and public decision-makers :
1.  Treat AI as a strategic business continuity risk, on par with energy and critical supply chains.
2.  Make dependencies visible and measurable through dedicated tools such as the Digital Resilience Index, in order to prioritize investments where disruption would have the greatest impact.
3.  Build resilience by design through technology-agnostic architectures, supplier diversification, exit capabilities, and local control over data and models.
4.  Elevate AI governance to board level, fully integrating geopolitical, energy, and industrial dimensions.

 

KEY LEARNING FOR EUROPE
The report highlights a decisive and counter intuitive insight: Europe can genuinely regain control over its digital destiny. Our capabilities are fragmented, but if pooled at the European level, they can reach critical mass and rise to the level of the United States and China.
The Japanese and Korean trajectories show that a path toward strategic autonomy is possible, including in the “hard” layers (semiconductors and compute).
This requires an integrated European strategy, making the Digital Single Market a shared framework for investment and deployment, while pursuing an open policy toward global partners who share our values and face similar geopolitical vulnerabilities.

The study shows that Europe’s lag is not structural, but organizational:
– if Europe were to genuinely unify its forces, it would move from 28 to 34;
– and if it were to pool its investments in the “hard” layers (as Japan has done), Europe would reach parity with the United States and China (36 out of 40).

 

At a time when artificial intelligence is reshaping the global balance of power, Europe faces a strategic dilemma: how to make law a driver of innovation rather than a constraint? From the Letta Report to the Draghi Report, a common ambition is emerging: to turn Europe’s complexity into strength.

Echoing the warning issued in 1999 by Lawrence Lessig with his famous “Code is Law” later hijacked by libertarians to celebrate the supremacy of code over democratic rules Digital New Deal’s report highlights the urgent need to avoid a shift toward AI writing its own rules. From “Code is Law” to “AI is Law”, there is a line Europe cannot cross without repoliticizing the digital space and reasserting the primacy of law.

Faced with an accumulation of often-contested regulations, it becomes essential to repoliticize the digital realm. The thesis defended in this report is clear: we do not have to choose between regulation and innovation — it is their strategic interplay that will shape a truly European AI.

“CODE IS LAW by Lawrence Lessig was a warning.
AI IS LAW must be a wake-up call.”

The author, Simon Bernard, identifies three essential conditions to achieve this objective:

• The standard must support scalability;
• The standard must accelerate its acceptability;
• The standard must be actionable to become a strategic asset.

His report is structured in three parts:

1. It clarifies the rules specific to the digital realm — often misunderstood — and aligns them around a shared goal: building ethical and safe AI.
2. It broadens the reflection to general branches of law — civil, commercial, environmental — which can also contribute to a distinctive form of AI, balancing tradition and disruption.
3. It proposes a political and strategic architecture aligned with the ambitions expressed by Letta and Draghi, in order to transform this legal framework into an operational instrument of European sovereignty.

AI is Law thus argues that clear, codified law is not a brake on innovation, but a powerful tool for sovereignty, competitiveness, and scaling in the age of artificial intelligence.

RECOMMENDATION TO ADAPT THE LEGAL ARCHITECTURE TO THE NEEDS OF AI

Why?
Law must become a lever for the scalability of innovation: it protects, legitimizes, and secures innovation.

What?
The legal architecture must be adapted to support technologies that evolve faster than the norms, while preserving its protective role for society.
An architecture inspired by the layered model of computing systems can be proposed (when Kelsen meets Turing):

• Law as an Infrastructure (LaaI): with fundamental rights and freedoms guaranteed by nature and rendered inalterable.
• Law as a Platform (LaaP): legal texts that can be coded (an “API-based law”), starting with data regulations.
• Law as a Service (LaaS): an experimental perimeter within a flexible but secure framework, inspired by regulatory sandboxes.

How?

• Recommendation No. 1 – Make code the 25th official language of the European Union, so that any new legal text can be translated into machine-readable language.
• Recommendation No. 2 – Include a legal coding clause in the upcoming omnibus regulations, enabling, at the European level, the short-term codification of key digital regulations, in line with ongoing work.
• Recommendation No. 3 – Support the Legal Data Space initiative, which already brings together a large number of stakeholders in the legal sector. This would be a shared infrastructure for processing and sharing public (open data) and private (data spaces) legal data — with the aim of developing sovereign legal AIs, by and for the legal sector.

PROMOTE A READABLE, HARMONIZED, AND SOVEREIGN LEGAL FRAMEWORK

Why?
To be a lever for innovation, legal rules must be predictable, intelligible, and actionable.

What?
Adapt the existing framework to provide greater coherence and strategic ambition.

How?

• Recommendation No. 4 – Simplify, harmonize, coordinate and sovereignize, limiting future changes to the legal framework to these core objectives.
• Recommendation No. 5 – Assess the effectiveness of AI-specific rules using a three-part framework:
  • Standardization (the rule should support scalability),
  • Acceptability (the rule should help innovation gain public and market acceptance),
  • Actionability (the rule should empower actors to use it as a competitive advantage).

We cannot risk allowing generative AI to further amplify the centralization and capture of value. We must prevent what we call “squared lock-in” (GenAI = Web2²). If we do not act, our companies could find themselves caught in a vise between Big Tech on one side, with their Big Clouds (hyperscalers), and now their Big AI on the other.

« Faced with this revolution, European companies have a choice to make: unite or endure. We firmly believe that by sharing our companies’ data, we can co-create a trustworthy, high-performance, and sovereign generative AI. »

We need to be realistic and focus our efforts on achievable goals.
The race to accumulate data for LLMs is over (Large Language Models are already standard products, according to Yann LeCun). There’s no point in trying to scrape all the web’s data as the giants do. Not only does this raise ethical and legal issues, but it also leads us astray. Europe must focus on a specific category of data — the most valuable kind and the only one Big Tech doesn’t yet have: the data from our corporates.

Capitalizing on corporate Data
We propose to capitalize on this private data by leveraging the European Data Strategy. This strategy aims to encourage data pooling and sharing among companies to create a common data market comparable to those in the U.S. and China, thereby enabling us to finally benefit from network effects. In practice, this involves participating in Data Spaces—trusted ecosystems that offer the right scale and governance to create generative AIs tailored to each sector. This is the best solution to avoid dependence on “Powered by OpenAI” or the counterproductive temptation of a model developed by a single company.

Promoting Data Spaces:
American studies have noted the emergence of a need for harmonized data layers essential for developing generative AI and observed that Europe has gained a lead with its Data Spaces initiatives under the Data Governance Act (data-sharing infrastructures with sector-specific governance). The report proposes building on this second competitive advantage and making these trusted data ecosystems the foundation of a trustworthy generative AI (performant, sovereign, and ethical).

Taking the Industrial Bet on LAM (Large Action Models)
In response to the Draghi report, which encourages taking bold technological bets, we propose focusing on Large Action Models (LAM). Rather than concentrating on language models (LLM), we believe Data Spaces should be used to create generative AIs dedicated to the autonomous execution of actions (agentification). These innovative models provide Europe with a strategic advantage through access to business data—a domain for which Data Spaces are particularly well-suited.

Following the example of the EONA-X Transport-Tourism Data Space with Gen4Travel
EONA-X has launched the Gen4Travel consortium to develop a LAM dedicated to the travel sector. This model will allow “Conversational Travel Assistants” of member companies such as Accor or ADP to become autonomous intelligent agents capable of automating complex tasks, such as booking or canceling flights and hotels in case of issues (including payment management). This approach protects the tourism sector from excessive disintermediation by an “agentified Booking or Tripadvisor.”

« Generative artificial intelligences based on a network of urbanized AI agents fed by trusted private data—that is the European ideal we propose to you »

This collaborative report, coordinated by Laura Létourneau and published by the two think tanks Digital New Deal and Terra Nova, is both a call to action and the first draft of a practical guide for deploying public infrastructures for data sharing — essential for the effectiveness of public policies, particularly the ecological transition and the country’s economic competitiveness.

« Urbanizing Digital Infrastructures for Collective Resilience»

Accelerating the secure circulation of data to address the challenges of our century
Public digital infrastructures for data sharing are a fundamental pillar for access to healthcare, the management of natural hazards and health crises, for accelerating economic transitions, and more broadly for the successful implementation of all public policies that require close collaboration between public, private, and/or non-profit actors.
Although the management of the Covid crisis highlighted the proportionally much stronger role of data sharing compared to AI — and even though inspection reports frequently point out failures in data sharing as a brake on public policies — these infrastructures still suffer from a lack of political support and insufficient resources.

Thinking of digital infrastructures in terms of urban planning
Just as in a city where houses are not built by public authorities but must comply with zoning codes and connect to sewage, water, and electricity networks, digital technology too must benefit from public infrastructures that organize a smooth and ethical exchange of data (avoiding a sort of slum where each one manages their own waste and lays their own power lines).

Providing concrete solutions to complex problems
The report highlights seven concrete examples of infrastructures, from which a practical, collective, and ambitious field guide has been drawn for their proper deployment. These include platforms such as Mon espace santé, the French citizens’ health record system that collects health data to empower each individual in managing their own health; Agdatahub, which facilitates data exchange in the agricultural sector to promote sustainable practices; or EONA-X, which pursues the same ambition for the mobility and tourism sector.
These initiatives bring major direct and indirect benefits to the effectiveness of public authorities, to the environmental and economic efficiency of companies, and to the strengthening of French and European digital sovereignty.

Aiming for transversal political governance
As with any systemic issue, data-sharing infrastructures raise complex governance questions. The report underlines the central role of public agents, as well as private and non-profit actors, in the success of this digital transformation. It also stresses the need for a shared interministerial vision, steered by the Prime Minister and supported by a DINUM/DGE duo at the national, European, and international levels — and for a collective, humanistic, and determined implementation.

Ce rapport collaboratif, coordonné par Laura Létourneau et publié par les deux think-tanks Digital New Deal et Terra Nova, est à la fois un plaidoyer et la première version d’un guide pratique pour le déploiement d’infrastructures publiques de partage de données essentielles pour l’efficacité des politiques publiques, notamment la transition écologique et la compétitivité économique du pays.

« Urbanisons le numérique pour une transformation durable »

Accélérer la circulation sécurisée des données pour faire face aux défis du siècle
Les infrastructures numériques de partage de données publiques sont un pilier fondamental pour l’accès aux soins, la gestion des aléas naturels, des crises sanitaires, pour accélérer les transitions de notre économie et globalement pour la mise en œuvre réussie de toutes les politiques publiques qui impliquent un travail étroit entre différents acteurs publics, privés et/ou associatifs. Quand bien même la gestion de la crise du Covid a mis en exergue le rôle proportionnellement bien plus fort du partage de données par rapport à l’IA et que les rapports d’inspections pointent fréquemment les défaillances du partage de données comme un frein aux politiques publiques, les infrastructures souffrent d’un déficit de portage et de moyens délétère.

Penser les infrastructures numériques en termes d’urbanisation
Comme dans une ville où les maisons ne sont pas faites par les pouvoirs publics mais doivent respecter le code de l’urbanisme et se raccorder aux réseaux d’égout, d’eau, d’électricité…, le numérique doit bénéficier, lui aussi, d’infrastructures publiques qui urbanisent un échange fluide et éthique de données (évitant une sorte de bidonville où chacun gère ses déchets, tire ses propres lignes d’électricité…).

Offrir des solutions concrètes à des problématiques complexes
Le rapport présente 7 exemples concrets d’infrastructures dont est dégagée un guide pratique terrain, collectif et ambitieux pour le bon déploiement des infrastructures. Parmi ceux-ci, on retrouve des plateformes comme “Mon espace santé”, le carnet de santé des français qui collecte les données de santé pour que chaque citoyen puisse être acteur de sa santé, “Agdatahub” qui facilite l’échange de données dans le secteur agricole pour promouvoir notamment des pratiques durables, ou bien “EONA-X” qui poursuit la même ambition pour le secteur mobilité-tourisme.
Ces initiatives ont des bénéfices directs et indirects majeurs sur l’efficacité des pouvoirs publics, pour sur l’efficacité environnementale et économique des entreprises et pour le renforcement d’une souveraineté numérique française et européenne.

Ambitionner un pilotage politique gouvernemental et transversal
Comme pour tout sujet systémique, les infrastructures de partage de données posent des questions de gouvernance complexes. Le rapport insiste sur le rôle central des agents publics et des acteurs privés et associatifs dans la réussite de cette transformation numérique. Il insiste aussi sur la nécessité d’une vision interministérielle commune pilotée par le Premier Ministre et soutenue par un binôme DINUM/DGE, au niveau national, européen et international), et sur celle d’une implémentation collective, humaniste et ultra-déterminée.

The 2024 European elections coincide with unprecedented challenges posed by Artificial Intelligence (AI) and global election participation. Indeed, we are in the year of all dangers, with half of the world’s democracies voting in 2024 while a threat of massive manipulation looms. The danger of cyber-interference has never seemed so high as generative AI gives enemies of democracy unprecedented means of destabilization. In this context, the European elections are particularly crucial; they indeed represent a vast “democratic crash test” with potentially major geopolitical consequences.

« Mastering AI means mastering social media and therefore influencing elections »

This report, written by Axel Dauchez and Hendrik Nahr, highlights a series of concrete measures initiated in the European program “Democratic Shield” (an initiative by Make.org bringing together civil society players such as SciencesPo, Reporters Without Borders, France Digitale, Sopra Steria, Schuman Foundation, ACTE, Alliance4Europe, EU Disinfo Lab, DemocracyWatch, Mozilla, and Digital New Deal).

This paper explores the Shield’s measures on the basis of a workshop held on this topic at the European Economic and Social Committee in Brussels. It provides a deep dive into the measures of reinforcing the quiet period, creating innovative tools to protect the electoral period, promoting citizen consensus debates as well as stakeholders pledges. It underscores the urgency of proactive collaboration between civil society and institutions to safeguard democracy. As technology evolves, ongoing vigilance and adaptive strategies are essential. The paper offers insights into navigating AI’s impact on elections, urging every segment of society to unite in raising a Democratic Shield.

A closer look on measures :

• Promotion of the citizen consensus debate through technological tools.
• Code of conducts for European political parties and for media ensuring that fakenews is not relayed by these stakeholders
• Reinforcement of the quiet period to prevent deepfakes from toppling the elections in the last 48 hours (as happened in Slovakia).
• Classify advertising algorithms as “High-risk AI” (AI act) to impose transparency and certification on these AIs of social media platforms (which amplify the virality of fakenews to boost their revenues).

“This report addresses the question of the role of digital technology, and more specifically digital data, for a sustainable future. Among the data that allow us to measure progress towards the Sustainable Development Goals (SDGs), the ESG data (Environmental, Societal, Governance) produced by the private sector is the perfect example of capture by a monopolistic group of actors to the detriment of the rest of the stakeholders and a global effort. This configuration symbolizes the effect of ‘platformization’ and privatization of crucial information for the implementation of the SDGs by companies. If digital is at the origin of this asymmetry, it is also the bearer of virtuous systems such as digital commons.”

« Our house is burning, and we are leaving the thermometer in a few hands»

We are convinced that a ‘Green – Digital New Deal’ is possible if we create a digital common for ESG data, managed by the UN, to fully benefit from the potential of big data and open data.

This ambition aims to create the conditions for trust necessary to achieve the Sustainable Development Goals (SDGs) through the triptych:

1. SHARED INFRASTRUCTURE : replace the manufacturing of information asymmetry between providers with a non-proprietary technological solution that is reasoned and painless for all actors (coordinated by the World Intellectual Property Organization – WIPO).
   
2. COLLECTIVE GOVERNANCE : rely on the governance of a community of trustworthy operational actors at the global level (to ensure the involvement of stakeholders in this mission of general interest).
3. TRUSTED DATA : guarantee access to a global registry of “raw” extra-financial data (data non-intermediated/biased via an Open Source Software Forge).

« Either we share data amongst us, or it remains centralized by bigtech »

This white paper is the result of multiple discussions with private and public actors involved in the creation of a sovereign and competitive French and European digital ecosystem. It attempts to express as pragmatically and as usefully as possible tech actors’ expectations in this great data adventure, and trustworthy digital world in its entirety.

Stakeholders sometimes worry about our time gap, and are consequently vigilant about the technological, legal and commercial impacts that this may imply. Are we agreeing to see our strategic data, and even data of general interest (health today, education or energy tomorrow,…) entirely delegated to American or Chinese hyperscalers ? The current geopolitical turmoil invites us to avoid choosing the dangerous path of dependencies.

« For the startup nation not to become a mirage, France must also become an infrastructure nation »

The purpose of this position paper is to shed light on a complex issue. First, we are convinced that data sharing can offer a trustworthy digital for Europe’s technological and economic ambition. Secondly, we’ll see how data spaces can play a strategic role. Finally, our conviction is that only a strategist state can think and impulse a joint set of “Cloud, AI, and Data” programs. In France, what we need is a “data strategy” that echoes the European Data Strategy.

From our perspective, the states should harness the emergence of a data sharing InfraTech, driven by digital commons, to support the development of the data spaces. In doing so, to provide the technical and political foundation and impose trust as a doctrine of the digital third way.

The funding request that we are expressing here (see right side of the figure below) is ultimately to provide France, and Europe, with the means to have weight in the implementation of Gaia-X and thus seize the economic opportunities that it creates for our related industries and ecosystems.

For AI to be a guarantor and vector of these values, we must demonstrate high ethical and political standards towards it. The “AI des Lumières” as understood by the Digital New Deal think-tank, is an “AI of trust” that refuses to let ideological biases – libertarian or authoritarian – guide its definition today, only to be translated into technological biases with a wide range of impacts: economic, societal, or ecological. Hence the importance of translating our values into global standards through the European regulatory package that creates this space of trust for the benefit of citizens and companies

« Trustworthy AI, a true beacon of our digital sovereignty and industrial competitiveness »

For our excellence in AI research to be translated into industrial leadership capacity, regulation alone is not enough. We must succeed in creating industrial cooperation at the European level, in line with our ambitions, to guarantee this digital sovereignty and to develop our industrial competitiveness. Writing this new chapter by 2030 is entirely possible. France for instance has already launched several ambitious programs that this report proposes to complete to increase its continental dimension and guarantee its impact.

In the following report, we present a method for achieving this at the European level. On the one hand, we propose making trustworthy AI the arrowhead of an ecosystem of trust and guaranteeing our strategic autonomy, and, on the other hand, we propose capitalizing on our industrial culture (critical systems) to seize this historic opportunity of conquering the global AI market.