Europe’s $100 Billion Bet: Can the Continent Break Free From American and Chinese Tech Dominance?

Europe has a dependency problem. Not on oil. Not on rare earth minerals. On software, cloud infrastructure, and the digital plumbing that runs virtually every government agency, hospital, bank, and factory from Lisbon to Helsinki.

The numbers are stark. European companies and governments spend tens of billions of euros annually on technology services provided overwhelmingly by American hyperscalers — Amazon Web Services, Microsoft Azure, and Google Cloud. China’s rising tech giants add another vector of concern. And as geopolitical tensions sharpen and data sovereignty becomes a matter of national security rather than regulatory housekeeping, European leaders are confronting an uncomfortable truth: the continent that invented the World Wide Web has almost no control over the infrastructure that powers it.

Now, a sprawling and ambitious effort is underway to change that. It spans EU-level regulation, national industrial policy, startup funding, and military procurement. Whether it will work — or whether it’s too late — is one of the most consequential technology questions of the decade.

The Scale of the Problem

As TechRadar reported in a detailed analysis, Europe’s digital dependency extends far beyond consumer apps and social media. Enterprise cloud computing, cybersecurity tools, AI platforms, semiconductor design software, and even the operating systems running critical government functions are dominated by non-European providers. The three major U.S. cloud providers — AWS, Microsoft, and Google — control an estimated 70% of the European cloud market. That’s not a gap. It’s a chasm.

This dependency creates multiple risk vectors. There’s the obvious data sovereignty concern: European citizen data, government records, and corporate intellectual property flowing through servers ultimately controlled by companies subject to U.S. law, including the CLOUD Act, which can compel American firms to hand over data stored abroad. There’s the supply chain risk, laid bare during the COVID-19 pandemic and the semiconductor shortage that followed. And there’s the strategic risk — the possibility that in a moment of geopolitical crisis, access to essential digital services could be throttled, restricted, or weaponized.

The war in Ukraine made this tangible. When Russia invaded, Western tech companies pulled services from Russia almost overnight. The move was widely applauded. But European policymakers noticed something else: the speed and unilateral nature of those decisions demonstrated just how much power a handful of American tech executives wield over entire national digital infrastructures. If it could happen to Russia, the logic goes, it could theoretically happen to anyone.

That realization has accelerated European efforts across several fronts simultaneously.

The EU’s approach is characteristically multilayered. The European Data Act, which entered into force in January 2024, establishes new rules on cloud switching and data portability designed to reduce vendor lock-in. The EU Chips Act has committed €43 billion in public and private investment to boost European semiconductor production, aiming to double Europe’s share of global chip manufacturing to 20% by 2030. And the AI Act, the world’s first comprehensive artificial intelligence regulation, is partly designed to ensure Europe doesn’t simply become a consumer market for AI systems built and controlled elsewhere.

But regulation alone won’t close the gap. Europe has tried regulatory approaches before. GDPR, enacted in 2018, was supposed to give Europe leverage over global tech companies. It did — to a point. But it didn’t spawn European competitors to Google or Facebook. It didn’t create a European cloud champion. Some critics argue it actually entrenched incumbents by raising compliance costs that only the largest firms could absorb.

So this time, Europe is pairing regulation with industrial policy on a scale not seen in decades.

France has been particularly aggressive. President Emmanuel Macron has positioned France as Europe’s AI hub, courting investment from both domestic startups and foreign players. Mistral AI, the Paris-based large language model developer, has raised over €1 billion and is frequently cited as Europe’s best hope for an indigenous AI champion. Germany, meanwhile, has focused on industrial digitalization, pouring resources into initiatives that connect its manufacturing base — still the continent’s strongest — with homegrown digital tools and platforms.

The Nordics are punching above their weight. Finland’s investments in quantum computing, Sweden’s thriving fintech sector, and Estonia’s pioneering digital government infrastructure all represent pockets of genuine European digital capability. But pockets are precisely the problem. Fragmentation — 27 member states with different procurement rules, different defense priorities, different startup cultures — remains Europe’s Achilles’ heel.

Cloud Sovereignty and the GAIA-X Question

No initiative better illustrates both Europe’s ambition and its challenges than GAIA-X. Launched in 2019 as a Franco-German project, GAIA-X was supposed to create a federated European cloud infrastructure — not a single monolithic cloud, but a set of common standards and interoperable services that would allow European providers to compete collectively against the American hyperscalers.

The results have been mixed. Charitably.

GAIA-X attracted hundreds of member organizations, including, controversially, the very American and Chinese tech companies it was ostensibly designed to counterbalance. Progress has been slow, bogged down by governance disputes and the sheer complexity of getting competitors to collaborate on technical standards. Several high-profile members have quietly stepped back. The project hasn’t failed — it’s produced useful frameworks for data exchange in sectors like healthcare and automotive — but it hasn’t delivered the transformative European cloud alternative its founders envisioned.

More targeted national efforts have shown greater traction. France’s “cloud de confiance” strategy, which certifies cloud providers meeting strict sovereignty requirements, has created a market for European providers like OVHcloud and 3DS Outscale. The catch: even France’s sovereign cloud framework allows American technology to be used, provided it’s operated under French legal jurisdiction by French-controlled entities. It’s sovereignty with an asterisk.

Germany’s Bundescloud initiative for federal IT, Italy’s national cloud strategy, and similar programs across Europe are all variations on the same theme: trying to create protected markets for European providers while not cutting off access to the superior technology that American firms currently offer. It’s a delicate balance, and one that satisfies almost nobody completely. Purists say it doesn’t go far enough. Pragmatists say going further would cripple European competitiveness by denying businesses access to the best tools available.

The defense sector is where sovereignty concerns are sharpest. NATO’s increasing reliance on cloud computing, AI-driven intelligence analysis, and networked warfare systems means that digital dependency has direct military implications. European defense ministries are scrambling to build classified cloud environments that don’t depend on American providers — or at least to ensure that when they do, the arrangements include ironclad guarantees about data access and operational continuity.

And then there’s AI. Europe’s position in artificial intelligence is a source of genuine anxiety among policymakers. The continent produces world-class AI researchers — many of whom promptly leave for better-funded labs in the U.S. The talent pipeline is strong. The commercialization pipeline is not. While Mistral AI and a handful of other European startups have made headlines, the overwhelming majority of frontier AI development is happening at American companies — OpenAI, Google DeepMind (technically a London subsidiary of an American parent), Anthropic, Meta — and increasingly at Chinese firms like DeepSeek and Alibaba Cloud.

Europe’s AI Act, for all its regulatory sophistication, doesn’t solve this. You can’t regulate your way to technological leadership. European Commission officials know this, which is why the InvestEU program and the European Innovation Council have been directing billions toward AI startups and research. But the funding gap with the U.S. remains enormous. American venture capital invested roughly $70 billion in AI companies in 2024 alone. Europe managed a fraction of that.

The semiconductor picture is similarly daunting. The EU Chips Act targets are ambitious — 20% of global production by 2030, up from roughly 8% today. Intel’s planned mega-fab in Magdeburg, Germany, was supposed to be the centerpiece, but the project has faced delays and uncertainty. TSMC’s new facility in Dresden, built as a joint venture with European partners, is more advanced but won’t fundamentally alter Europe’s dependence on Asian chip fabrication for the most advanced nodes.

What Europe does have, and what often gets overlooked in the doom-and-gloom narrative, is ASML. The Dutch company is the sole manufacturer of extreme ultraviolet lithography machines — the tools required to produce the most advanced semiconductors on earth. Without ASML, neither TSMC nor Samsung nor Intel can make cutting-edge chips. It’s arguably Europe’s single most important piece of strategic tech leverage. The Dutch government’s willingness to restrict ASML exports to China, under pressure from Washington, demonstrated both the company’s importance and the geopolitical entanglements that come with it.

There’s also the open-source angle. Europe has historically been a strong contributor to open-source software, and some policymakers see this as a path to reducing dependency without building proprietary European alternatives to every American product. The European Commission’s Open Source Software Strategy, updated in recent years, encourages public administrations to adopt open-source solutions. Linux, after all, was created by a Finnish university student. The philosophical roots are European even if the commercial exploitation has largely happened elsewhere.

But open source alone isn’t a strategy. It’s a tool. And tools need investment, maintenance, and institutional support to be effective at scale.

What Comes Next

The European Parliament elections in 2024 and the resulting new European Commission under Ursula von der Leyen’s second term have injected fresh urgency into these debates. The new Commission has signaled that digital sovereignty will be a top priority, with specific focus on reducing dependencies in cloud computing, AI, and telecommunications infrastructure. The proposed European Sovereignty Fund, while still taking shape, could channel significant new resources toward strategic tech investments.

Recent developments have added fuel. Concerns about TikTok’s Chinese ownership have spread from Washington to Brussels, with multiple EU institutions banning the app from official devices. The ongoing U.S.-China tech decoupling — export controls on advanced chips, restrictions on Chinese access to American AI models — has forced Europe to pick sides, or at least to hedge its bets more carefully.

And the rise of generative AI has created a new urgency. Every month that passes without a competitive European foundation model is a month in which European enterprises become more dependent on American AI infrastructure. The data generated by European businesses training on and fine-tuning American AI models flows back to American companies, creating a feedback loop that’s extraordinarily difficult to break.

Some observers are cautiously optimistic. Europe’s regulatory frameworks, while imperfect, create market conditions that can advantage European providers. The Digital Markets Act’s interoperability requirements, for instance, could open doors for European competitors in messaging, app distribution, and cloud services. Public procurement — the single largest technology purchasing force on the continent — is slowly being redirected toward European providers where viable alternatives exist.

Others are more pessimistic. They point to decades of failed European tech champions, from the Quaero search engine (meant to rival Google, dead on arrival) to various European social media platforms that never gained traction. They note that Europe’s venture capital markets, while growing, still can’t match American scale. They observe that Europe’s best tech companies — Spotify, Klarna, Adyen — tend to incorporate or list in the U.S. as they scale. Brain drain remains real.

The honest assessment lies somewhere in between. Europe is unlikely to replicate Silicon Valley or build a European AWS from scratch. That ship sailed years ago. But it doesn’t need to. What it needs is sufficient domestic capability in critical areas — cloud, AI, semiconductors, cybersecurity — to ensure that dependency doesn’t become vulnerability. Enough redundancy to maintain options. Enough market power to negotiate favorable terms.

That’s a less inspiring vision than digital independence. But it might be the achievable one.

The stakes are real. A Europe that controls none of its digital infrastructure is a Europe that has outsourced a significant portion of its sovereignty — economic, military, and political — to companies and governments whose interests may not always align with its own. The continent’s leaders understand this now in a way they didn’t a decade ago.

Whether understanding translates to execution is the question that will define European technology policy for the next decade. The money is being committed. The regulations are being written. The political will, for now, exists. What remains to be seen is whether Europe can overcome its chronic fragmentation, its risk-averse investment culture, and its habit of producing excellent strategy documents that gather dust in Brussels filing cabinets.

The dependency problem is real. So is the effort to solve it. The outcome is genuinely uncertain — which, given Europe’s track record in technology, might actually count as progress.