Yesterday afternoon, the European Commission released its long-awaited 'Tech Sovereignty Package'.[^1] Here at Leitmotiv, we will take our time to properly read and digest the many pages of text produced by the Commission further. We also look forward to a necessary period of serious democratic debate, deliberation and reflection on the proposed legislation and policies. This blog post is only a first general reflection on the package, focusing in particular on the Cloud and AI Development Act ("CADA"), on which we prepared a submission in the summer of last year.[^2]

Overall, what strikes us (once again) when reading through the Commission's package is that the Commission aims to go all-in on an AI future for Europe. While the Commission's Communication to the European Parliament and the Council ("Communication") accompanying the package is wrapped in – somewhat defensive – language around the amorphous concept of 'technological sovereignty', CADA is all about moving rapidly towards an AI-future. It aims to "accelerate the adoption of AI and cloud computing on a large scale, including at regional and local level" and to have "broad adoption of AI in private and public sectors".[^3]

Disconnect between technological sovereignty & AI adoption

While technological sovereignty and AI adoption may appear connected, they are not inherently so. Technological sovereignty implies that critical minerals, chip production, data centers, cloud services and AI/other digital services production should be controlled – at least to some degree – by European governments, European companies and other European organizations. The Commission rightly points out that Europe has become increasingly "reliant on non-EU providers for over 80% of its digital products, services, infrastructure and intellectual property".[^13] Reaching that conclusion, however, does not necessarily mean we should be steering Europe towards a future in which AI adoption is embraced across the board. By seemingly merging these two ideas of 'tech sovereignty' and 'AI embrace', it appears as if the Commission believes that the only way to respond to actual and perceived threats to Europe's technological sovereignty is to become an AI leader in the global context.

The danger of this ambition is that it increases the willingness to brush aside serious concerns. CADA appears almost unapologetically enthusiastic about artificial intelligence: "By enabling faster innovation, greater efficiency and smarter decision-making, AI contributes to substantial economic, environmental and societal gains and is a fundamental driver of competitiveness." Throughout CADA, the emphasis is on AI uptake and adoption rather than on managing the significant environmental, social and human rights risks associated with such a dive into an uncertain AI future.

We believe this attitude is also reflected in an apparent departure from concepts with clear legal reference points, choosing vague and undefined terms instead, such as 'human-centric approach'. There are repeated references to 'European values' in the Commission's Communication, yet it is good to remind ourselves that these values are explicitly anchored and listed in article 2 of the Treaty on European Union: "The Union is founded on the values of respect for human dignity, freedom, democracy, equality, the rule of law and respect for human rights, including the rights of persons belonging to minorities. These values are common to the Member States in a society in which pluralism, non-discrimination, tolerance, justice, solidarity and equality between women and men prevail."

A fair question to ask as EU legislators, member states and broader public when reading this package is: What values does the European Union still stand for in 2026, amid the actual and perceived threats to its 'tech sovereignty' and its rush towards an AI society?

Shifting European values

More particularly, one of the EU's values is that the Union is based on the rule of law, which, we may assume, includes the international rule of law. Among its many other international legal obligations, the EU is obliged to meet the Paris Agreement goal of keeping global temperatures within safe limits. Yet, the only role this climate goal (and legal obligation) appears to play in CADA is to justify that data center projects in so-called 'Data Center Acceleration Zones' should get approved rapidly because these data center projects supposedly contribute to this climate goal.[^4] Substantively, the Commission appears to equate 'sustainability' in digital infrastructure deployment with the efficient use of energy and other resources by data center operators.[^5] It thereby loses sight of the overall impact of accelerating the uptake of AI and AI infrastructure on climate change, among other things.

While individual data centers deployed as a result of CADA may meet high standards of energy efficiency and other criteria, the focus on tripling Europe's data center capacity in the next 5-7 years will virtually drive up the sector's overall greenhouse gas emissions. As far as we can see, CADA does not explain how pushing the digital infrastructure sector towards a steeply upward emissions trajectory would align with the EU's climate obligations under the Paris Agreement.

Ignoring the obvious impact of CADA on the emissions trajectory of the digital infrastructure sector appears deliberate. The starting point and strong belief of the Commission remains that the EU has "limited data centre capacity". The explicit goal of the Cloud and AI Development Act ("CADA") therefore remains unchanged: to triple EU data center capacity in the next 5-7 years. Last year, however, we have written to the European Commission that we have reason to believe Europe has more than enough data center capacity at present.

Recent research by Leitmotiv also confirms this for the Netherlands.[^6] We are currently missing sufficient transparent facts on the actual utilization rate of any part of digital infrastructure that could guide political decision-making, including the decision in CADA to increase overall capacity in the market. We therefore asked the Commission last year to conduct a full assessment of the utilization rate of existing data center capacity in Europe. How can we move forward, after all, if we don't know where we stand at present?

A step forward with room to improve

There appears some acknowledgment at least by the Commission that such measurement is necessary. Article 15 of the CADA proposal provides that the Commission will monitor "compute capacity available in the Union" and "volume of demand for data centre capacity". The document also mentions the following indicator for measuring progress on achieving CADA's objectives: "Utilisation rate of EU computing capacity".[^7] This is an improvement compared to the plans presented last year. However, we think it is not good enough yet.

First of all, the assessment of current capacity on the market should happen before the goal is set to triple data center capacity in Europe in the next 5-7 years, not in the text of CADA legislation already adopting that goal.

Second, we take issue with the fact that article 15 of CADA is framed as monitoring progress towards closing a supposed 'capacity gap', a questionable assumption as far as Leitmotiv is concerned.

Third, the Commission has not described any standard or methodology by which it will measure and monitor compute capacity, volume of demand or utilization rates.[^8]

Conclusion

To conclude, we want to briefly point out a few key elements of the package that we have questions about (although many more questions will emerge as we read and debate the plans further, we are sure!):

1. The European Commission claims that "public funding alone cannot close Europe's investment gap..." in digital infrastructure.[^9] Even if we assume there is indeed an investment gap, what is the right answer to it? According to the Commission, it is essential to "crowd in massive volumes of private capital".[^9]

   We at Leitmotiv would like to know: Why would private capital come to/stay in the EU for investment in digital infrastructure by the hundreds of billions of EUR? And what are the opportunity costs of hundreds of billions of EUR invested in EU digital infrastructure compared to other very worthwhile investments (such as in renewable energy)?
   
   

2. According to the Commission, "measures under CADA to accelerate the deployment of data centre infrastructure in the EU are expected to generate a demand-side effect."[^10]

   We at Leitmotiv question the somewhat circular logic underlying such 'demand-side measures'. Is the Commission claiming that by pumping a lot of money in Gigafactories and tripling data center capacity through – effectively – cutting red tape, the EU will be creating more demand for 'homegrown' AI chip production? Will Europe then be able to produce enough competitive AI chips in the next 5-7 years that can be installed in those new EU data centers and Gigafactories? Is that really the fastest and most efficient way to give European AI startups or cloud developers a chance to compete on existing markets?[^11]
   
   

3. The EU has experienced increased resistance against (especially hyperscale) data center development in recent years. As we see in the United States, there is resistance across the political spectrum among citizens against the rush towards building ever-more energy-hungry hyperscale data centers. Calls for moratoria are therefore on the rise in the EU and beyond in response (a movement we support at Leitmotiv).[^14]

   How does this political reality relate to the call by the Commission for "Data Centre Acceleration Zones" that aim to speed up data center deployment by giving operators "simplified procedures to access land and reliable energy infrastructure"?[^12] Is this proposal for acceleration zones is not dramatically out of touch with the concerns of many citizens across the EU?

These are some of our first thoughts on the Tech Sovereignty Package announced yesterday by the European Commission. Stay tuned as we continue our research, participation in democratic deliberation and advocacy for a more equal, democratic, rights-respecting and truly sustainable digital economy here at Leitmotiv.

References