Prosus just published The State of AI in Europe — and the findings challenge everything you think you know about Europe’s position in the AI race.
The mainstream narrative: America dominates, China competes, Europe regulates. The data tells a different story.
Europe Is the World’s Largest AI Consumer
133 million monthly LLM users in Europe versus 61 million in the US. Europe is not sitting on the sidelines — it is the world’s leading adopter of AI.
The problem is not adoption. The problem is what Europeans are using: ChatGPT, Gemini, Copilot, DeepSeek — all built outside Europe. As the Prosus report puts it: “Europeans consume AI brilliantly, but we train the algorithms owned by others.”
This is not just a sovereignty concern. As AI gets embedded deeper into apps, the company that owns the app trains the model on user behavior. Distribution becomes the competitive moat, and Europe is handing that moat to US and Chinese platforms.
The Talent Is Here
Europe and the United States have roughly the same number of AI professionals — approximately 325,000 each. Three of the ten most-cited AI scholars are European. DeepMind was founded in London. Meta’s LLaMA model was invented in Paris.
But the talent is structurally misallocated:
- 48% of European AI professionals work in old economy roles (vs. 33% in the US)
- 35% work in digital-native tech companies (vs. 54% in the US)
- 6 of Europe’s top 15 AI employers are US Big Tech — not a single European tech company makes the list
Europe is caught in what Prosus calls a “retrofitting loop” — elite talent is used to enhance legacy sectors rather than build the next generation of global platforms. This is the same dynamic I see in enterprise consulting: organizations apply AI to optimize existing processes rather than rethinking what is possible.
The Capital Gap Is the Real Problem
Europe creates AI startups at the same rate as the US — roughly 900 new VC-backed companies per year. European founders account for 25% of all global AI unicorn founders. At early-stage funding, Europe and the US are comparable: $4 billion versus $5 billion.
Then comes the chasm:
| Stage | Europe | United States |
|---|---|---|
| Early-stage VC | $4B | $5B |
| Late-stage VC | $12B | $141B |
Late-stage funding in Europe is nearly 12x smaller than in the US. And by the time European startups break through, 73% of lead investors in large rounds are American.
The numbers that matter most: European VC funds direct only 12% of their capital into AI, versus 76% for American counterparts. Europe does not have a talent problem. It has a capital allocation problem.
Where Europe Can Win
The Prosus report is honest about what is lost and what is still in play.
The GenAI Race Is Over
Only Mistral has produced notable European foundation models. Pretending otherwise is not a strategy. But losing one race does not mean losing the war.
Trust as a Competitive Advantage
US labs are going closed-source. China has embraced open-source for global reach. Europe’s path is a third way: trusted, open-source AI built to European values. Think “Made in Germany” but for AI.
This aligns directly with the digital sovereignty conversation. Organizations increasingly want to know where their data goes, who trains on it, and what governance frameworks apply. Europe’s regulatory posture — often criticized as a constraint — could become a differentiator.
Vertical AI Is the Opportunity
Over 75% of European AI investment already targets specialist applications. In AI for energy, Europe leads the world with 50% of global VC in that segment. Health, energy, defense, and financial services are precisely where European domain expertise creates durable advantage.
This is where I see the strongest opportunity from a platform engineering perspective. Vertical AI requires deep domain knowledge combined with production-grade infrastructure — Kubernetes, GPU orchestration, observability, and automated deployment pipelines. Europe has the domain expertise. It needs the platforms to scale it.
World Models: The Next Frontier
Systems that understand the physical world — robotics, autonomous systems, manufacturing, logistics — represent a frontier where no one has established dominance yet. Europe is at parity with the US in the talent required. This is the window to move aggressively, before it closes like the GenAI window did.
What Needs to Change
The Prosus report outlines five structural shifts:
Unlock pension capital. Trillions of euros sit in pension funds earning cautious returns. Raising VC allocation from 0.12% to just 3% would generate approximately 100 billion euros for European growth capital and close the late-stage funding gap almost overnight.
Focus on strongest hubs. London, Paris, Munich, Amsterdam, Stockholm — rather than diluting impact through fragmented funding across every member state. Exponential technologies thrive on density. This is why events like the recent AI House Amsterdam session on agentic commerce matter — they build the concentrated ecosystem that scales.
Build EU-Inc. One corporate structure recognized across the bloc, with a full operating framework covering stock options, capital markets, and employment law. Borderless company building.
Simplify the rulebook. The AI Act, GDPR, Data Act, NIS2 — each with its own obligations and timelines — pile cumulative burden on the startups Europe most needs. The EU Cyber Resilience Act adds another layer. The Digital Omnibus should be fast-tracked and adopted by mid-2026.
Own the distribution. The most urgent threat to European AI sovereignty is not the models Europeans use — it is the apps and agents that control distribution. A continent that relies on foreign apps for its digital life has surrendered its AI future.
The Bottom Line
Europe has world-class talent, the world’s largest AI user base, record investment of $21.8 billion in 2025, and a growing trust advantage as AI geopolitics shift. What it lacks is urgency.
As someone based in Amsterdam working across AI strategy, cloud infrastructure, and open-source communities, I see this tension daily. The talent is here. The demand is here. The capital and structural frameworks need to catch up.
Read the full report at prosus.com/state-of-ai-europe.
For more on AI strategy and platform engineering in Europe, connect with me on LinkedIn or follow @TheLucaBerton.
