The CEPS Task Force on the Apply AI Strategy: Disentangling What Europe Actually Disagrees About

In Brief

The Centre for European Policy Studies (CEPS) is convening a Task Force to address implementation challenges in the European Commission's Apply AI Strategy. An information session scheduled for 15 April 2026 invited organisations to participate across three sector tracks: Healthcare & Pharma, Automotive & Mobility, and Government & Public Sector. The initiative arrives at a critical juncture: Europe has shifted from regulating AI to deploying it, but the strategy comes with ambitious goals and no dedicated budget. The real question isn't whether Europe should apply AI at scale, but whether the current approach matches the actual disagreements that need resolving.

This analysis scratches the surface of a debate that deserves sustained attention. For those ready to engage with these questions directly, Human x AI Europe convenes today in Vienna, where the conversation moves from paper to practice.

The Pivot That Matters

Something significant happened in European AI policy, and it wasn't the AI Act. After years of regulatory architecture, the European Commission has pivoted toward deployment. The Apply AI Strategy targets eleven sectors with the explicit goal of making Europe an "AI Continent." It promotes an "AI first" policy orientation and a "buy European" approach, particularly for public sector procurement.

The framing is ambitious. The execution is another matter.

When the CEPS Task Force on the Apply AI Strategy launched, it did so with a specific diagnosis: the strategy "comes with a heavy implementation burden, no dedicated budget, and hard questions that remain unanswered." This is not criticism for its own sake. It's an invitation to disaggregate what "insufficient" actually means.

Three Disagreements Wearing One Coat

When critics say the Apply AI Strategy won't work, they might mean several different things. The failure to distinguish between these positions produces debates that generate heat without light.

The resources disagreement is the most straightforward. Either the European Competitiveness Fund will be linked to Apply AI priorities, or it won't. Either AI Factories and Gigafactories will receive sufficient investment, or they won't. As CEPS researchers observed, the strategy "comes with a lot to do but no budget." This is a factual claim about implementation capacity. It can be resolved with numbers, political will, and budget lines.

The infrastructure disagreement is more complex. The strategy assumes that compute infrastructure will serve sectoral needs. But the same CEPS analysis notes that "simply relying on large-scale data centres and powerful GPUs isn't going to be enough. Every sector will require a different solution." Healthcare AI may need edge computing for privacy-sensitive diagnostics. Automotive AI may need real-time inference capabilities. Government AI may need sovereign cloud architectures. The question isn't whether Europe needs infrastructure. It's whether the infrastructure being built matches the use cases being prioritised.

The data governance disagreement is the most stubborn. The EU's data strategy has struggled to produce results. From the European Health Data Space to Catena-X in automotive, industry hasn't found effective mechanisms for data sharing that support both primary and secondary uses. The Apply AI Strategy assumes these mechanisms will materialise. Sceptics ask a pointed question: why would it be different this time?

These are three distinct problems requiring three distinct solutions. Conflating them produces policy debates where participants talk past each other, each convinced the other is missing the point.

What the Task Force Actually Does

The CEPS information session scheduled for 15 April 2026 outlined an approach designed to disaggregate these questions systematically. The Task Force operates across three sector tracks, each applying a consistent analytical structure across five issue areas: key use cases, infrastructure requirements, data governance, flagship applications, and sovereignty and implementation roadmaps.

This structure matters. It forces participants to specify which problem they're actually discussing. A healthcare executive worried about compute latency for diagnostic AI is having a different conversation than a public sector technologist concerned about data sovereignty. Both may be sceptical of the Apply AI Strategy, but their scepticism points in different directions.

The Task Force is led by Andrea Renda, CEPS Director of Research, and brings together senior executives, policymakers, technologists, and researchers. Organisations interested in participating were invited to contact Katja Spanz or Artur Bogucki through the registration process.

The Governance Question Nobody Wants to Answer

The Apply AI Strategy creates new governance structures: the Apply AI Alliance as a coordination forum, and an AI Observatory to track sectoral impact. These are sensible additions. But they raise a question that deserves direct engagement.

Who decides what counts as success?

The strategy's eleven sectors have different metrics, different timelines, and different stakeholders. Healthcare success might mean reduced diagnostic errors. Automotive success might mean competitive European autonomous vehicle platforms. Public sector success might mean reduced administrative burden. These are not the same thing, and they may not be achievable with the same resources, infrastructure, or governance mechanisms.

The CEPS Task Force approach, by separating sectors while maintaining a common analytical framework, at least makes these trade-offs visible. Whether European policymakers will engage with them honestly is a separate question.

The Sovereignty Dimension

The Apply AI Strategy promotes a "buy European" approach, particularly for public sector procurement. This is a values position, not merely a technical one. It assumes that European AI sovereignty is worth pursuing even if it comes with costs in terms of capability, speed, or price.

Reasonable people disagree about this. Some argue that European sovereignty in AI is essential for democratic accountability and strategic autonomy. Others argue that it risks locking Europe into inferior solutions while the rest of the world moves ahead. Still others argue that the framing itself is wrong: that sovereignty should be defined in terms of governance capacity rather than geographic origin of technology.

The Task Force doesn't resolve this disagreement. But by making it explicit, it creates space for participants to argue about what they actually disagree about, rather than talking past each other with different definitions of the same words.

What Would Have to Be True

For the Apply AI Strategy to succeed, several things would have to be true simultaneously:

The European Competitiveness Fund would need to allocate meaningful resources to AI deployment, not just AI research. The AI Factories and Gigafactories initiative would need to produce infrastructure that matches sectoral needs, not just generic compute capacity. Data governance mechanisms that have failed to materialise for years would need to suddenly work. And European AI providers would need to offer solutions competitive enough that "buy European" doesn't mean "accept inferior."

Each of these is possible. All of them together is ambitious. The CEPS Task Force exists to identify which of these conditions are most likely to fail, and what can be done about it.

The Question Worth Asking

The Apply AI Strategy represents a genuine shift in European AI policy. The question is whether Europe's institutions are capable of the implementation work that follows.

The CEPS Task Force is one attempt to provide the structured analysis and stakeholder engagement the strategy requires. Whether it succeeds depends on whether participants are willing to disaggregate their disagreements, specify their assumptions, and engage with trade-offs honestly.

That's not a technical challenge. It's a political one.