Can Europe go without American tech?

Maybe one day, but only at an incredible cost...

Can Europe be competitive without relying on American technology?¹

It’s an important question—and one that more European leaders are contemplating as the movement for transatlantic independence gains steam.

Discussions are rampant in Brussels about how to ramp up European economic growth and competitiveness in tech as the Americans and Chinese leap ahead, but there’s also a strong movement in favor of European tech sovereignty—in effect, centering European tech and businesses rather than American ones.

But that puts Europe stuck between a rock and a hard place—Europe is so reliant on American technology, that any efforts to pry apart the tech relationship would probably have deleterious short- to medium- term impacts on the European economy. It could take years to recover.

While there are huge upside for Europe in reducing dependency on American tech—from security considerations to reducing competition for European firms—the realities of American influence in Europe means that the bloc also risks shooting itself in the foot. American Big Tech—especially in data centers and cloud computing—is so pervasive that it would years and hundreds of billions for Europe to fill the gap.

Is that a tradeoff Europe would be willing to take?

Europe’s tech dependency dilemma

The vast majority of European companies are heavily reliant on American Big Tech. A study by the Swiss company Proton showed that Big Tech usage, as measured by use of American email clients, is largest in Scandinavia and Western Europe.

But the dependency goes far beyond just email. Data centers and cloud computing capacity in Europe are primarily American. There are also concerns that America could be surveilling European data, as so much European data is funneled through American companies, such as Microsoft.

That means much of the tech stack in Europe, stretching from infrastructure (chips), to intermediaries (payment processing, cloud computing), to applications (artificial intelligence, email, maps, etc.) are controlled by American firms. It’s not that European options don’t exist, but they are tiny in scale compared to the American giants. In terms of data centers and cloud capacity, it’s like a European bicycle lane besides an eight-wide American highway. 70 percent of cloud capacity in Europe is American.

At a conference in Brussels this week, Maria Farrell, an Irish writer on technology, said that “the US big tech monopoly is extractive on every level of the stack,” and that Europe had to assert sovereign principles to fight back. European values, she said, should drive Europe’s approach to tech—not money-grabbing American ones.

But others in the startup field said not so fast. Europe is so dependent on American technology, that founders would flee even more quickly to the US to make use of superior American tech. Cutting ties, according to European startup leader Robin Wauters, would just lead to isolationism and Europe running even further behind the US. Instead, he said that Europe needed to focus more on integrating the single-market, to make it easier for European companies to scale within Europe. The problem isn’t the Americans, he argued, but structural issues within Europe itself.

Europe’s American-free options

Does Europe truly have alternatives to American platforms? That’s where Eurostack comes in. It’s an initiative to develop European built infrastructure along all aspects of the tech stack. That includes European-built, owned, and operated cloud storage and database management, video conferencing tools, messaging software, and content management systems.

The problem is, however, that few European companies and consumers are actually using these technologies. On all levels of the tech stack, American companies are significantly more popular, even if European options exist.

Here are some examples, taken from the three slices of the tech stack: infrastructure, intermediaries, and applications.

---

Infrastructure

Chips

American firms like Nvidia, AMD and Intel dominate. While ASML in the Netherlands has the lithography technology to mass produce chips, Europe itself cannot build semiconductor chips at the cutting-edge, nor at scale. Most of Europe’s chips capacity is for automobiles, not for use in AI. The EU is trying to scale to 20 percent of global demand by 2030, but that’s still a sliver of what the EU would actually need if it were to go completely sovereign.

Data centers

On data centers, Europe is even further behind. Amazon, Google, and Microsoft provide the vast majority of physical data center capability in Europe. While there are European options, their capacity is just a fraction of the Americans. For AI in particular, the situation is particularly grim, with the cutting-edge AI facilities in Europe owned by the Americans or Chinese.

Intermediaries

Cloud computing

This is the biggest problem for Europe. American companies like Amazon Web Services, Microsoft Azure and Google Cloud are still the most important players here. While groups like Gaia-X are trying to create more European options, the recent failure of AWS in Europe further reflected the dependence of many applications running Europe on the American cloud. European firms have a fraction of the capacity of American ones, and it would take years to build up enough cloud capacity to meet European needs.

Applications

Artificial Intelligence

OpenAI, Anthropic, Google, and Meta dominate the AI space. Mistral AI is far behind the cutting-edge in the field. While European leaders talk a big game about Mistral, the size of the company and its ambitions pale in comparison to the American firms. Mistral wants to open a 100 MW data center in Paris. In one project in Texas, OpenAI wants to build 10 GW. That’s 100 times more capacity from one project in the US.

Messaging

The entire European Union runs on WhatsApp (owned by Meta) and Microsoft Teams. While European options certainly exist, the network effect means that changing to another messaging app would be a significant change. There are platforms like Threema and Skred, but they’re not widely used at all.

---

The worst-case scenario

Let’s say the worst-case tech scenario happened for Europe²—Donald Trump demands the European Union back down on its regulations governing American technology companies. The Commission (but probably not under Ursula von der Leyen) finally decides to use the trade bazooka, effectively blasting American firms out of the European market. WhatsApp, Google Maps, Gmail, and many, many more apps stop working.

Yes, the scenario is highly implausible—it’s much more likely that if things got this bad, Europe would target specific companies with fines and sanctions rather than restricting market access wholesale—but it’s still an interesting thought experiment.

If that happened, could Europe still be competitive in technology?

The short answer—maybe one day, but at an incredible cost.

On the application level of the tech stack, Europe would survive and perhaps even thrive. There are open source and European alternatives to ChatGPT, WhatsApp, Gmail, Google Docs, and many others. Those applications would grow remarkably in the wake of a transatlantic tech split. Numerous European founders are in the tech space, showing there’s no lack of talent and ideas. The problem is too many of them take their companies to the American market to grow. Perhaps with additional attention and funding, Mistral AI could even catch up to its American and Chinese competitors.

But Europe is far behind on the critical intermediary and hard infrastructure necessary to compete and grow technologically. If Amazon Web Services, Microsoft Azure, or Google Cloud could no longer operated in Europe, the entire tech ecosystem in Europe would fail overnight.

Recent data shows that US firms hold a whopping 70 percent of the cloud computing capacity in Europe. If somehow that capacity were turned off, much of the European internet would go along with it—and it would take years and hundreds of billions of dollars in infrastructure investment to build up that capacity again. That said, maybe that’s the kind of shock those calling for European sovereignty want—a fundamental reshaping of the tech stack that would lead to true European independence.

At this point, contemplating a split of that kind is simply not realistic for Europe, but the bloc certainly should be doing more to prepare for the scenario. While the EU is doing some work in this direction, the investment gap in cloud computing and data centers between the EU and US is so large, and European budgets already so strained, that it would take a fundamental re-ordering of European budgets and investment priorities. Here’s a study published in the European Centre for International Political Economy.

The investment gap between the EU and the US in ICT and cloud-related sectors, totalling some USD 1.36 trillion, presents a significant challenge for native European companies. To catch up by 2030, 2040, and 2050, European technology companies would need to substantially increase annual investments, ranging from approximately USD 157 billion to USD 1.2 trillion annually, representing 0.8 percent to 6.4 percent of the EU’s GDP. This would – hypothetically – require a significant commitment of resources.

For context, investments of between 0.8 percent to 6.4 percent of the EU’s GDP make the bloc’s American-enforced increase in NATO defense spending look small. And this study came out in 2024, before the biggest AI data center booms. That means Europe is even further behind than these numbers show.

So, for now, it’ll be more WhatsApp, AWS, Gmail, and OpenAI. That is, unless Europe’s leaders decide now to make the huge investments in infrastructure and cloud needed for a truly tech sovereign Europe. However, that would require spending like a superpower, and even if the world may be demanding Europe become one, the bloc is not ready for prime time yet.

1. This question was debated at a conference in Brussels I attended this week put on by the Center for Future Generations on Europe, technology, and competitiveness. Quotes in the piece are from speakers there.

2. This idea was entertained at the Center for Future Generations conference.