EU digital sovereignty is a grand delusion in a sharper suit

The EU has unveiled its latest digital sovereignty strategy in its bid to loosen dependence on American Big Tech. On paper, it sounds bold, muscular and geopolitical, the moment Europe finally stops merely regulating Silicon Valley and starts building its own technological future. In reality, it is the same grand delusion in a sharper suit. It is, in fact, a grand delusion.

The phrase rests on three falsehoods. First, the EU is not sovereign. Sovereignty belongs to nation-states that can tax, coerce, legislate, and answer to a demos. The European Union can coordinate, regulate, and subsidise, but it cannot act as a sovereign state because it is not one. Second, no country is digitally sovereign across the entire technology stack – not America, not China. Third, the EU’s preferred route to sovereignty through regulation, targets and subsidy programmes has weakened the very conditions from which genuine technological power would have to emerge.

The digital world is not a territory to be enclosed. It is a layered, global, and interdependent system of platforms, chips, operating systems, data centres, standards, research ecosystems, and manufacturing networks. Power does not come from pretending to own the whole stack. It comes from occupying positions that the stack cannot function without.

France already demonstrated what digital sovereignty looks like in practice. It was called Minitel. In the 1980s, France built a nationally controlled online infrastructure years before the web. It was technologically impressive, commercially successful, and deeply embedded in everyday life.

Yet Minitel’s strength became its weakness. The system was closed, national and controlled by a state telecommunications monopoly. When the open internet arrived, France’s sovereign system became an obstacle. The very features that made Minitel successful made adaptation to a borderless, global network more difficult. France pioneered mass-market online services yet failed to produce a global internet platform.

The lesson is uncomfortable. Digital sovereignty can often be achieved only at the expense of openness, scale, and adaptability-the qualities that made the internet dominant.

This is why Europe’s current obsession is so dangerous. It encourages Brussels to fight yesterday’s battles. Europe does not have a Google, Amazon Web Services, Apple, Microsoft, Nvidia, Meta or OpenAI. It does not control the cloud layer, smartphone operating systems, browsers, frontier AI or global digital platforms.

Its distinctive power lies elsewhere: Regulation.

But regulatory power is not digital sovereignty. It is merely the power to regulate technologies built by others. The celebrated Brussels Effect depends on the continued presence of foreign firms in Europe. Without them, there would be little of significance to regulate.

Indeed, the compliance burdens generated by Europe’s regulatory model often fall hardest on European innovators. The result is a paradox at the heart of EU technology policy: Europe asserts authority over foreign technologies while making life more difficult for the domestic firms that might one day challenge them.

The results are visible everywhere. Gaia-X, the Franco-German cloud initiative, was supposed to create a European alternative to American hyperscalers. Instead, Europe gained standards frameworks while remaining dependent on foreign cloud providers. The EU Chips Act promised to double Europe’s share of global semiconductor manufacturing, yet its flagship Intel project in Germany was cancelled and the target now appears aspirational. Northvolt, once hailed as Europe’s battery champion, entered bankruptcy.

This is not bad luck. It is a pattern.

Brussels repeatedly confuses announcements with achievements, regulation with capacity and target-setting with strategy. It behaves as though technological power can be summoned through legislation, funding programmes, and press releases.

There is a better European model.

It is called ASML.

The Dutch lithography company was not created by a digital-sovereignty programme. It emerged over decades through engineering excellence, patient capital, supplier networks and a willingness to solve problems others considered impossible.

Today, every advanced semiconductor manufacturer depends on ASML’s extreme ultraviolet lithography machines. Every cutting-edge chip passes, in effect, through Veldhoven.

That is real power: Not sovereignty, but indispensability.

Recent conflicts have once again highlighted the geopolitical importance of chokepoints. The Strait of Hormuz matters because global trade depends on it. ASML occupies a similar position in the digital economy. It is a point in the global technology system through which the most advanced chips must pass.

The United States understands this perfectly. Washington has used ASML’s position to restrict China’s access to the tools needed to manufacture advanced AI chips. Europe’s most strategically important technology asset has become part of a broader exercise of American state power.

Meanwhile, Brussels continues to produce new strategies, acts and initiatives intended to manufacture sovereignty through administrative design.

The contrast is striking. Brussels talks about sovereignty. Washington exercises it.

Yet the lesson remains largely ignored. Europe’s future will not be secured by decreeing European clouds, search engines, social networks or foundation models into existence. The next strategic opportunities lie elsewhere: Photonics, quantum technologies, advanced materials, industrial software, semiconductor research infrastructure, AI evaluation and precision biotechnology.

Europe already possesses examples of this model. ASML in lithography. IMEC in semiconductor research. Airbus in civil aviation. Fraunhofer in applied research. SAP in enterprise software. These organisations differ in important ways, but they share a common characteristic: They became globally significant because they developed capabilities others needed, not because governments declared them sovereign.

The required shift is simple to describe, though difficult to execute: Stop performing sovereignty and start cultivating indispensability.

That means completing the Single Market in capital, services, insolvency and corporate law so European firms can scale. It means permitting telecoms consolidation instead of trapping the continent in fragmented markets unable to finance next-generation infrastructure. It means creating more IMEC-style research platforms in strategically important fields. It means reducing compliance burdens on start-ups and accepting that meaningful technological breakthroughs often require decades rather than electoral cycles.

Europe’s principal obstacle is not Silicon Valley nor Shenzhen. It is its own institutional culture: A technocratic preference for the appearance of power over the difficult, unglamorous and long-term work required to build it.

The language of digital sovereignty has become an evasion. It allows political leaders and Brussels officials to speak the vocabulary of power while avoiding the burdens that genuine power entails.

Europe does not need another sovereignty strategy.

It needs more chokepoints.

The future belongs to those who shape the system, not those who issue rules for systems built by others.

https://brussels.mcc.hu/uploads/default/0001/02/0838fe5c247ec346ca0b3b9cf36a616b74e20412.pdf

Dr Norman Lewis is Visiting Research Fellow at the think tank MCC Brussels and author of the new report “The Grand Delusion: The European Union and Digital Sovereignty”