Where the next big companies in Europe could come from, according to US investor

We have spent the last decade watching Europe play a game of catch-up with Silicon Valley. For a long time, the narrative was simple: the US builds the platforms, and Europe regulates them. But the recent shift toward deep tech and artificial intelligence has changed the math. The question is no longer whether Europe can produce a billion-dollar company, but whether it can produce a company that actually anchors an entire sovereign ecosystem.

The DeepMind Legacy and the Talent Loop

When you look at the landscape of European AI, you have to talk about Google DeepMind. It is the definitive case study for both the region's potential and its greatest frustration. DeepMind proved that world-class research could happen in London, away from the Bay Area. However, it also served as a reminder that without local capital and massive infrastructure, the biggest breakthroughs eventually get swallowed by American incumbents.

For founders today, the DeepMind story is a double-edged sword. On one hand, the "DeepMind Mafia" is real. Former employees are spinning out to start their own ventures, bringing with them a level of technical rigor that you typically only find in places like Stanford or MIT. This talent loop is critical. In the past, European talent tended to stay in academia or migrate to the US. Now, we are seeing a generation of builders who want to stay and build locally.

But building a company in the shadow of a giant like Google is difficult. The talent is there, but the commercial infrastructure is lagging. If you are a builder in this space, you aren't just competing for engineers; you are competing for the sheer computational resources that the US hyperscalers control by default.

The Shift to Physical Intelligence

One of the most interesting trends emerging from the European scene is the pivot toward where AI meets the physical world. While the US is currently obsessed with Large Language Models and chatbots, there is a distinct movement in Europe toward robotics, biotech, and specialized hardware. This is a smart play. Europe has a long history of industrial excellence, and applying AI to those existing strengths is a much more viable path than trying to build another generic search engine or social media platform.

Investors are starting to realize that the next "Big Tech" might not be a software-only play. It might be a company that revolutionizes how we manufacture goods or how we manage energy grids. For a founder, this means looking at the boring sectors. The less flashy the industry, the more room there is for AI to provide actual, measurable value rather than just hype.

Capital is Getting Serious, But It Is Still Cautious

We are seeing more US venture capital firms setting up permanent shops in London and Paris. They see the valuation arbitrage. A team of PhDs in Berlin is often significantly cheaper to fund than a similar team in Palo Alto, yet the output is frequently comparable. This inflow of US capital is a vote of confidence, but it comes with strings attached. US investors expect US-style growth, which can be a culture shock for European founders who are used to more conservative scaling models.

The challenge for builders is maintaining control. When a US firm leads your Series A, the gravitational pull toward relocating to the States becomes very strong. If Europe wants to keep its winners, it needs more than just early-stage seed funds; it needs the kind of late-stage growth capital that can sustain a company through the expensive process of scaling to a global market.

Regulatory Moats and Friction

You cannot talk about building in Europe without talking about the AI Act. From a founder's perspective, regulation is usually seen as a hurdle. It creates friction. It slows down deployment. But there is a counter-argument to be made: if you can build a company that is fully compliant with the world's strictest AI regulations from day one, you have a built-in moat when you decide to expand globally.

However, I remain skeptical of the idea that regulation alone can foster innovation. You don't regulate your way to a trillion-dollar company. You build your way there. The builders who will win are those who treat the regulatory environment as a fixed constraint, like physics, and find ways to innovate within those boundaries without letting it dampen their ambition.

The Infrastructure Gap

The biggest bottleneck for the next European giant isn't talent or ideas; it is compute. As long as European AI startups are dependent on AWS, Google Cloud, and Azure, they are essentially paying a tax to their competitors. We are seeing some movement toward sovereign cloud initiatives, but these are often bogged down by political bureaucracy.

For a founder, this means your stack matters more than ever. Efficiency is a competitive advantage. If you can build models that are smaller, faster, and less energy-intensive, you reduce your dependence on the American hardware monopoly. This is where the "builder-first" mentality pays off. The goal shouldn't be to build the biggest model, but the most efficient one for a specific, high-value problem.

The Founder Perspective

If you are starting a company in Europe right now, you have to be honest about the trade-offs. You have access to incredible talent and a supportive (if sometimes slow) public sector. But you are also fighting against a fragmented market and a venture ecosystem that is still learning how to take massive risks. The next big European company won't be a copy of a Silicon Valley success story. It will be something that could only have been built in Europe—something that combines deep scientific research with the continent's industrial backbone.

The era of the "copycat" startup is over. The next generation of builders is focused on hard problems that require more than just a clever UI. They are looking at the foundational layers of how the world works. Whether they can stay independent long enough to become the new establishment is the real question.

The goal is not to build a European version of an American company. The goal is to build a company that the rest of the world has to react to.

We are currently in a transition period. The pieces are on the board—the talent from DeepMind, the interest from US VCs, and a renewed focus on industrial AI. The missing piece is the conviction to stay the course through the difficult middle years of scaling. For builders, the takeaway is clear: stop looking at what the US is doing and start looking at the unique problems in your own backyard that AI is finally capable of solving.

Read the original at Sifted →