Summary Under the proposed Cloud and AI Development Act (CADA), Centres for AI (Experience and Acceleration Centres for AI) enjoy "substantial overall autonomy" regarding their internal organisation, composition, and working methods. This autonomy is conditional on their compliance with the Regulation's objectives. While Member States are responsible for establishing them and national strategies must designate them as key entry points to the European AI ecosystem, the Centres are not micromanaged in their daily operations. This structure allows them to act as agile, locally adapted accelerators for AI adoption across the Union.

Detail

The Cloud and AI Development Act (CADA), as proposed in COM(2026) 502 final, establishes a network of Experience and Acceleration Centres for AI (hereinafter "Centres for AI") to bridge the gap between EU-level research and local market adoption. A defining characteristic of this framework is the deliberate balance between maintaining a coherent Union-wide strategy and granting local entities the flexibility to respond to regional needs.

The Legal Basis for Autonomy

The core of the Centres' operational independence is explicitly codified in Article 5(5) of the proposal. This provision states:

"Centres for AI shall have substantial overall autonomy as regards their organisation, composition and working methods, in compliance with the objectives set out in this Regulation."

This clause is pivotal for understanding the governance model. It confirms that while the Centres are established under a Union Regulation, they are not direct administrative extensions of the European Commission or rigidly controlled by centralised bureaucracy. Instead, they are empowered to determine:

  • Organisation: How they structure their internal teams, management hierarchies, and operational workflows.
  • Composition: Who sits on their boards, which experts they engage, and how they partner with local stakeholders.
  • Working Methods: The specific methodologies they employ to deliver training, testing, and innovation support.

However, this autonomy is not unlimited. The phrase "in compliance with the objectives set out in this Regulation" acts as a binding constraint. As detailed in Article 5(2), these objectives include:

  • Supporting the integration and scaling-up of AI use cases in strategic industrial and public sectors.
  • Accelerating the broad adoption of cloud and AI technologies at regional and local levels, notably for SMEs, small mid-caps (SMCs), and public sector bodies.
  • Leveraging relevant infrastructure to accelerate the development and fine-tuning of AI models.

Consequently, a Centre for AI cannot exercise its autonomy in a manner that contradicts these missions. For instance, a Centre could not refuse to support public sector bodies or ignore the "AI first" principle mandated by the broader framework.

Integration into National Strategies

The autonomy of the Centres is further contextualised by their strategic role within Member States. Article 7(2)(b) mandates that Member States must include specific measures in their national cloud and AI strategies to:

"accelerate the development and adoption of cloud and AI at national, regional and local level... including by supporting the Centres for AI referred to in Article 5 as entry points to the European AI innovation ecosystem."

This requirement creates a dual layer of accountability. On one hand, the Centres are the practical interface where national strategies meet local industry needs. On the other, Member States are legally obliged to support them, ensuring they have the necessary resources and political backing to function effectively. This linkage ensures that while the Centres are autonomous in how they operate, they are strategically anchored in national policy goals. They serve as the designated "entry points" for organisations seeking to navigate the European AI landscape.

Governance, Oversight, and Network Effects

While the Centres possess substantial autonomy, they do not operate in isolation. The proposal establishes a robust governance and oversight framework to ensure consistency and quality across the Union:

  1. Commission Oversight: Under Article 5(4), the Commission is empowered to adopt implementing acts detailing the procedure for establishing Centres for AI. These acts will cover participant organisation profiles, selection criteria, and further arrangements concerning tasks and functions. This ensures a baseline standard for establishment across all Member States.
  2. Network Collaboration: Article 5(6) establishes a network of Centres for AI to support collaboration and the exchange of best practices. This network allows Centres to provide specialised services across regions where local skills or compute capacity are insufficient, ensuring that autonomy does not lead to fragmentation.
  3. Alignment with Existing Hubs: Article 5(1) requires that Centres for AI build on the existing network of European Digital Innovation Hubs (EDIHs), ensuring continuity and leveraging established expertise rather than creating entirely new administrative silos.

For public-sector bodies and industry stakeholders, this structure implies that a Centre for AI is a trusted, vetted entity. Its autonomy allows it to be responsive to local market dynamics, while its inclusion in the CADA framework and national strategies ensures it adheres to high standards of transparency and alignment with EU values, such as technological sovereignty and open-source promotion.

What this means for you

For public-sector procurement officers, legal teams, and business leaders, the autonomy of the Centres for AI has several practical implications:

  1. Flexible and Tailored Engagement: Because Centres for AI determine their own working methods, they can tailor their support to your specific challenges. You are not bound by a rigid, one-size-fits-all service model. Whether you need advice on evaluating cloud sovereignty levels (Union assurance levels) or structuring procurement tenders to include EU added-value criteria, the Centre can adapt its approach to your needs.
  2. Local Relevance and Expertise: The autonomy in composition means Centres are likely to include local experts, industry representatives, and academic partners who understand your specific regional context. This makes them invaluable partners for identifying local or European providers that might otherwise be overlooked in favour of global hyperscalers.
  3. Strategic Alignment: When engaging with a Centre for AI, you can expect their advice to be aligned with your national cloud and AI strategy. As mandated by Article 7(2)(b), they act as entry points to the European AI innovation ecosystem. Their recommendations will support broader goals like reducing dependency on third-country providers and promoting open-source solutions.
  4. Direct Procurement Support: As part of their operational objectives under Article 5(2), Centres for AI help organisations accelerate digital transformation. For procurement officers, this translates into direct support for preparing tender documents, assessing technical requirements for AI systems, and understanding the new EU added-value criteria that may be applied in public procurement.

Common misconceptions

  • Misconception: Centres for AI are direct extensions of the European Commission.
    • Reality: While they are part of an EU-wide network and must comply with Regulation objectives, Article 5(5) grants them substantial autonomy. They are established by Member States and operate with independent governance, composition, and working methods. They are local/regional accelerators, not EU bureaucrats.
  • Misconception: Centres for AI can set their own legal requirements.
    • Reality: Their autonomy is operational, not legislative. They must act "in compliance with the objectives set out in this Regulation." They cannot override national procurement laws or EU data protection regulations. Their role is to facilitate compliance and adoption, not to create new legal mandates.
  • Misconception: Centres for AI are only for private companies.
    • Reality: Article 5(2)(b) explicitly states that their objectives include accelerating adoption "notably for SMEs, SMCs and public sector bodies." They are a critical resource for public-sector entities looking to modernise their IT infrastructure and adopt AI in compliance with CADA's sovereignty frameworks.
  • Misconception: Autonomy means a lack of oversight.
    • Reality: The Centres operate within a strict framework. The Commission can adopt implementing acts on their establishment (Article 5(4)), and they must report to national strategies. Their autonomy is designed to foster agility, not to bypass accountability.

Related

This is general information about a draft EU regulation, not legal advice.