Now Reading: Digital Sovereignty: Europe’s Real Problem Is the Exit Door
-
01
Digital Sovereignty: Europe’s Real Problem Is the Exit Door
Europe does not need to build every technology itself. Digital sovereignty begins with something less dramatic: retaining the practical ability to leave.
European digital sovereignty is often caricatured as an attempt to build a continental copy of every American or Asian technology. That makes the idea easy to dismiss as protectionism. The real question is more concrete: how autonomous can an economy be when cloud infrastructure, semiconductors, AI models and essential platforms depend on a small number of external suppliers?
On September 23, 2026, Bank of Italy deputy director general Gian Luca Trequattrini offered a useful definition in a speech on European digital sovereignty: autonomy does not require producing everything in Europe. It requires preserving a credible ability to choose and replace critical suppliers.
That definition is less spectacular than self-sufficiency. It is also far more practical.
Digital sovereignty is the ability to leave
A system is sovereign when it can continue to function even if commercial, geopolitical or regulatory conditions change. In digital infrastructure, that idea applies to at least four layers: chips, compute infrastructure, software and data.
A European company may use a non-European cloud provider, AI models developed elsewhere, GPUs made by a small group of vendors and proprietary software that is difficult to replace. None of those choices is necessarily a problem in itself. The risk begins when those dependencies become irreversible.
Sovereignty, then, is not about the passport of the supplier. It is about the ability to migrate, interoperate, negotiate and maintain operational continuity.
Big Tech platform power comes not only from economic scale, but from becoming infrastructure that is expensive or dangerous to replace.
Cloud and AI make dependency visible
The European Commission describes technological sovereignty as the capacity to act independently by developing and controlling key technologies, data and infrastructure while reducing dependency on third-country suppliers. In June 2026 it presented a package including Chips Act 2.0, the Cloud and AI Development Act and a European open-source strategy.
The important point is not the list of programmes. It is the attempt to intervene across the whole stack. A data center built in Europe can strengthen local compute capacity, but it does not automatically make the underlying technology European if it still depends on processors, orchestration software and models controlled elsewhere. An AI model developed in Europe offers limited autonomy if the infrastructure required to run it remains accessible only through a handful of hyperscalers.
Digital sovereignty is therefore a stack problem. Every layer can create a dependency on the layer below it.
Chips are the hardest bottleneck
Europe retains important industrial capabilities across the semiconductor supply chain, but it does not control the full chain for the frontier processors used in AI. The Commission itself acknowledges that dependencies remain significant and that Europe’s share of the global semiconductor market is still far from the targets set for 2030.
The problem is not solved by building one factory. Advanced chips depend on equipment, intellectual property, materials, packaging, design software and markets large enough to sustain enormous investment.
Resilience therefore comes from diversification and control over strategically important segments, not from attempting to duplicate every stage of the supply chain.
Open source and interoperability are sovereignty tools
The least glamorous part of Europe’s strategy may also be one of the most important. Open-source software does not eliminate dependency, but it can lower the cost of leaving a platform and make critical systems easier to inspect, modify and move.
The same is true of interoperable standards. A market in which switching cloud provider, AI model or infrastructure requires rebuilding everything from scratch may be formally competitive while remaining practically locked.
In that sense, digital sovereignty can be compatible with open markets. Its purpose is not to close them, but to prevent an open choice made today from becoming an irreversible dependency tomorrow.
Sovereignty is bargaining power, not isolation
Europe’s greatest risk would be turning digital sovereignty into an industrial slogan. Funding European infrastructure is not enough if companies and public institutions remain tied to the same proprietary architectures. Regulating foreign operators is not enough if credible alternatives do not exist.
The practical measure of autonomy is much simpler: can we change supplier without ceasing to function? Can we preserve data and services? Can we move between infrastructures? Do standards, skills and industrial capabilities exist in sufficient depth to prevent a single external decision from blocking an entire sector?
This is where digital sovereignty stops being an identity debate and becomes an economic one. It does not mean doing everything alone. It means making sure today’s technological choice does not become tomorrow’s inability to choose.
