Europe’s AI moment: bold plan, real stakes

The European Union wants to become the world’s first AI continent. That’s the headline behind the newly unveiled AI Continent Action Plan—a sweeping strategy to turn Europe from a regulatory heavyweight into a genuine innovation hub. It’s ambitious. And it’s necessary.

A commentary by our AI-Expert Arian Okhovat.

---

Infrastructure: no AI without compute

One thing the Commission gets right: there’s no digital sovereignty without compute power. The plan promises to expand Europe’s capacity dramatically—through AI Factories built around public supercomputers, and a handful of massive “Gigafactories” with tens of thousands of high-performance chips.

If these sites become accessible—not just to big players, but also to startups and public interest projects—they could break the cycle of dependence on U.S. cloud providers. But this hinges on execution. Infrastructure on paper is easy. Infrastructure people can actually use is something else. At PANTA, we frequently face the same challenges when developing scalable solutions. While there are a few niche providers—often US-based but hosted in Europe—they can be expensive and inflexible. Building our own cloud infrastructure is usually only an option for our larger enterprise customers. The new “AI Continent Plan” offers hope of closing this gap and establishing sustainable, European alternatives in the long run.

Data: from silos to scale

The same applies to data. AI doesn’t just need data—it needs volume, structure, quality, and legal clarity. Right now, Europe has some of the strongest data protection rules in the world, but still lacks a functioning data economy.

The new plan wants to change that through “Data Labs” and a forthcoming European Data Union. The vision: a continent-wide framework where data can be shared responsibly and used to train AI models that serve health, mobility, climate, and more.

If that works, Europe won’t just have trustworthy AI. It’ll have useful AI.

Talent: make or break

No AI ecosystem can thrive without people. The EU finally seems to understand that. With its AI Skills Academy, new study programs, fellowships, and returnship schemes, the plan lays out how Europe could grow—and retain—its own AI talent.

This matters. Because so far, too many European researchers have ended up in labs in California or Beijing. And too many companies struggle to find the expertise they need to innovate at speed.

Talent is Europe’s most underused resource. This plan could change that—if the programs go beyond pilot scale and actually meet demand on the ground.

Regulation: from ‘rules-first’ to ‘ready-to-implement’

Europe made headlines with the AI Act. And rightly so: it’s the first serious attempt to set guardrails for high-risk AI. But regulation alone doesn’t create progress. It creates questions, compliance costs, and sometimes confusion.

The Action Plan acknowledges that. It introduces support structures: a dedicated AI Service Desk, practical guidance, and more agile mechanisms to help companies navigate requirements.

Crucially, the EU is also actively involving AI practitioners in shaping the regulatory landscape. A good example is the recently introduced “Code of Practice,” developed collaboratively with industry experts—I’ve personally had the opportunity to contribute to some of these working groups. Incorporating real-world perspectives helps ensure regulations stay both practical and relevant.

The plan also hints—carefully—that some rules might evolve. That’s a welcome sign of flexibility. Especially for startups, who need clarity and speed, not red tape.

Europe’s bet: trust as a competitive advantage

At its core, the plan makes a strategic bet: that openness, transparency and ethics aren’t just constraints, but competitive assets. That companies will want to build on infrastructure they can trust. And that societies will adopt AI faster if they understand how it works and why it matters.

Projects like OpenEuroLLM—open-source, multilingual, European-built—are more than symbolic. They show what a different path could look like: one where AI is not just “safe” by design, but owned, understood, and shaped by European actors.

The ambition is right. Now prove it.

The AI Continent Action Plan sets the tone for a different kind of AI future—European, values-based, yet globally relevant. But it’s just a plan. What matters now is what happens off the page.

Will compute and data access actually reach the people building the next generation of applications? Will the balance between innovation and regulation hold? Will Europe move from potential to presence? This is the moment to find out.

You can find the publication by the EU parlament and the Action Plan available for download here .