Europe has joined the AI race. The real gap is scale

The familiar account of European technology is one of arriving late. The United States has the hyperscalers, the frontier laboratories and the deep capital markets; Europe has talent, regulation and promising companies that adopt more slowly. The European Investment Bank's March 2026 Investment Report complicates that story, and the contributed piece below argues that it has been read the wrong way round.

Its case is that adoption is no longer Europe's problem. On the EIB's numbers, European firms use big data analytics and AI at least as widely as their American counterparts. What Europe lacks is the capital, market structure and exit routes to turn adoption into companies that lead, and that gap is measured in tens of billions of euros a year.

Oleg Khusaenov is chief executive and founder of Zubr Capital, a Cyprus-based growth investor, and a founder of the Atlant-M holding. The article that follows sets out his opinion.

For most of Europe, the story of AI integration and adoption was one of delay, delay, delay. While the United States had hyperscalers, frontier labs, mega-round funding and deeper capital markets, Europe had talent, regulation and promising companies, but not the same speed of adoption.

At Zubr Capital, we analysed the March 2026 EIB Investment Report, and its findings challenge this familiar narrative. It showed that in 2025, 76 per cent of EU companies used digital technologies in their operations. That is almost level with the 78 per cent in the US. However, 46 per cent of EU firms used big data analytics and AI compared to only 40 per cent in the US. That shows Europe is not simply playing catch-up any more. It is actually slightly ahead. But adoption is not the same as AI leadership.

Good news: AI is entering mainstream European business

European AI use is now showing up in more places than among early adopters. This adoption wave is beginning to show up in economic performance. From our perspective at Zubr Capital, this is one of the most important signals in the EIB report. The EIB's analysis of EIBIS data shows AI adoption accounts for about 12 per cent of the total productivity increase since 2019.

Capital is moving in the same direction. In Q1 2026, European startups raised $17.6 billion in venture funding, according to Crunchbase data published in April. AI companies alone attracted $9.2 billion of that total, more than half of all European venture funding, and it was the first time on record.

AI in Europe is not just in research labs. It is entering the mainstream, shaping where productivity and capital move next. In our view, that changes the terms of the debate. But that makes the next question unavoidable: what exactly are we measuring when we say Europe has caught up?

Parity depends on which AI metric is being measured

Even within the EIB data, there are nuances when the question shifts from whether firms use AI to how deeply they use it. Only 55 per cent of EU companies using big data analytics and AI invest in generative AI across at least two business areas. That is significantly different from 81 per cent of US peers.

Measuring at the worker level also tells a different story. Bick et al. found that 43 per cent of US workers used generative AI at work in early 2026, higher than the 32 per cent average across six European countries. These measurements are not contradictions of the EIB reporting, but rather a reminder that a company can report using AI even if the technology has not yet spread widely across employees' daily workflows.

The conclusion is less flattering for Europe, but more useful. European firm-level AI adoption has a strong case for catching up with the US, but on the depth of integration, the US lead is still visible.

The real gap is in scale-up capital

The more AI moves into mainstream European business, the more the question shifts from adoption to scale. This is where the story becomes less comfortable. According to the EIB, Europe faces an annual funding gap of around €120 billion for startups and scaleups, including €80 billion at the scale-up stage alone. US funds provide 42 per cent of venture investment in AI and data software in the European Union.

Even when sources use different methodologies, they point in the same direction. A European Commission working document, drawing on PitchBook data, shows EU companies raised $21.3 billion in late-stage venture rounds in 2024. That is far below the $133 billion from the US. Writing for CEPR/VoxEU, Josh Lerner put total EU VC investment at €66.2 billion in 2025, roughly 22 per cent of the US total, despite broadly comparable economic size.

The harsh reality is that when promising European tech companies reach a point where they need larger cheques, deeper networks and patient growth capital, more than half of late-stage investment comes from outside the European Union. Europe could help create a new AI wave, then be forced to watch as its most promising companies scale on someone else's capital.

Why the US advantage persists: capital markets, exits, fragmentation

The scale-up gap is part of a wider market structure. Europe's capital markets are shallower than those in the US across almost every layer, with 2025 stock market capitalisation at 69.5 per cent of GDP in the EU compared to 184.5 per cent in the US. In the same 2025 research, follow-on equity issuance reached $92 billion in the EU and $294 billion in the US, with corporate bond issuance at $780 billion and $1.68 trillion, respectively, with the US bond market's growth driven in part by the technology sector, particularly AI.

There is a practical difference in funding, especially for high-growth companies. In the EIF Equity Survey 2025, cited in the EIB report, fund managers reported fragmented markets, with access to capital and lower risk appetite as core obstacles to scaling European firms. These findings pointed to the absence of a single EU stock exchange or harmonised equity landscape. Regulatory fragmentation often makes it easier to seek funding and expand into the US than to expand across EU borders.

The single market is Europe's great asset, but it is still not fully behaving like one. Around 62 per cent of EU firms report fragmented rules that make exporting within the bloc far too complicated. Removing such barriers could lift investment intensity by 10 per cent, which is why policy responses now matter.

How European policy change could help scale-up finance

Europe's policy debate is starting to move in the same direction: from supporting innovation to financing scale. For example, in May 2026, the European Union selected EQT to manage the Scaleup Europe Fund, with a fund size of around €5 billion, for strategic deep-tech sectors including AI, quantum, clean energy and space, with first investments expected later in the year.

Initiatives like these build on existing public finance foundations. The EIB Group accounts for 24 per cent of Europe's VC market and 30 per cent of its venture debt. The European Tech Champions Initiative 2.0 is designed to expand late-stage financing capacity. That public structure will not be enough to solve a market design issue, but its real test is whether it brings in private institutional capital, and whether growth investors can turn the gap into investable scale.

Investor angle: where there is opportunity

Where investors can see real gains is in the shift from AI exposure in European businesses to AI-enabled scale. The question is no longer whether a company can add AI to its product story, but whether AI helps it improve retention, lower operating friction, get to market faster, offer clearer unit economics, and develop products that travel across borders.

We believe frontier models will remain important, but much of the good AI investment in Europe may be in vertical applications, where AI-powered software improves specific workflows in real estate, retail, finance, logistics, or industrial operations. These companies will need distribution, capital discipline, and the ability to scale across international markets.

Recent deals point to where these opportunities may lie. Our investments in Zing Coach and Placy illustrate how AI can be applied to concrete consumer and business problems, with the investment case tied to product expansion rather than AI branding alone. The scale-up gap is not only a weakness. For selective growth investors looking at EU markets, it is also where value can be created.

Zing Coach shows how vertical AI can expand an existing market rather than merely automate an existing service. Personal training remains expensive and inaccessible to many consumers, particularly in the US. By using generative AI to tailor programmes to users' goals, fitness levels and progress, the company turns personalised coaching into a scalable digital product. Its partnership with Paris Saint-Germain demonstrates how this model can reach large audiences through established sports ecosystems. The investment case is not simply that AI makes coaching more efficient. It is that AI can make a traditionally premium service accessible to a much broader market.

Placy takes on a different problem: efficiency in the real estate market. Instead of growing headcount, agencies get AI tools that help them process client enquiries faster, qualify leads, and support deals through to close. We backed the Cyprus-based startup's pre-seed round in 2024, drawn by a founding team that had already built and scaled consumer products before.

Adoption gets Europe into the AI race. Capital decides the winners

We believe the EIB report matters because it changes the starting point of Europe's AI story. It is no longer a matter of EU companies lagging behind in the AI transition. Firm-level adoption of big data analytics and AI shows European companies are already in the race.

From our perspective, the question for the next stage is how to turn such adoption into lasting advantages. That will require larger late-stage rounds, deeper capital markets, stronger exit routes, and a single market that behaves more like one.

The same distinction between adoption and effective integration applies to investors themselves. Zubr Capital uses AI tools across the investment process, screening incoming deal flow, processing investment materials, structuring information, drafting analytical materials, and supporting decision-making. The result is more time for our team to make judgement calls grounded in human expertise.

In one internal exercise, a working group examined how our results compared with those of other teams. It found that the AI-assisted group's conclusions were broadly consistent with those of the non-AI group. We have had to ask questions about how to use AI not only when evaluating portfolio companies, but also in how we run our own shop. Just like Europe, we are still learning how best to integrate emerging technologies across business operations while keeping human judgement at the centre.

Europe may have shown it can adopt AI, but the harder test is whether it can finance, scale and retain the companies that turn adoption into global strength.