Summary As proposed, the Experience and Acceleration Centres for AI (β€˜Centres for AI’) under the Cloud and AI Development Act (CADA) would not be funded by a single, dedicated new budget line. Instead, Article 6(3) explicitly states that the Cloud and AI Leadership Initiativesβ€”which the Centres supportβ€”would be supported by funding from existing Union programmes, primarily Horizon Europe and the Digital Europe Programme. Recital 25 clarifies that these Centres would build upon the existing network of European Digital Innovation Hubs (EDIHs), repurposing and expanding them rather than creating entirely new entities. Furthermore, Recital 29 confirms that Member States may support these initiatives through national research, development, and innovation measures, provided they comply with applicable State aid rules, alongside private-sector investments.

Detail

The funding architecture for the Centres for AI represents a strategic integration of existing EU instruments rather than the creation of a standalone financial silo. To understand the financial mechanics, one must examine the interplay between the establishment mandate in Article 5, the implementation mechanisms in Article 6, and the financial context provided in the recitals of the proposal COM(2026) 502 final.

The Foundation: Building on Existing Hubs (Article 5 & Recital 25)

The Centres for AI are not designed to emerge from a vacuum. Article 5(1) mandates that each Member State shall establish these Centres, but it crucially specifies that they "shall build on the European digital innovation hubs established under Article 16 of Regulation (EU) 2021/694."

Recital 25 reinforces this continuity, noting that the network will "build on the network of European digital innovation hubs." This has significant financial implications:

  • Infrastructure Leverage: A substantial portion of the physical and digital infrastructure required for the Centres already exists within the EDIH network, which has been funded under the Digital Europe Programme.
  • Operational Continuity: The Centres would likely inherit existing operational frameworks, staffing structures, and management systems, reducing the need for new capital expenditure on setup.
  • Targeted Expansion: The funding challenge shifts from "building new" to "expanding capabilities." The new CADA mandate requires these hubs to take on specific AI-focused tasks, such as supporting frontier AI, physical AI, and industrial AI, which may require additional resources beyond their original EDIH scope.

The Primary Funding Vehicle: Article 6(3)

The core mechanism for Union-level financial support is detailed in Article 6, which governs the implementation of the Cloud and AI Leadership Initiatives. These initiatives define the operational objectives that the Centres for AI are tasked to achieve (e.g., accelerating AI adoption, supporting SMEs, and fostering skills).

Article 6(3) provides the definitive funding route:

"The Cloud and AI Leadership Initiatives may be supported by funding from Union programmes, including Horizon Europe and the Digital Europe Programme, in accordance with Regulation (EU) 2021/694 and Regulation (EU) 2021/695."

This provision establishes a dual-stream funding model:

  1. Horizon Europe (Regulation (EU) 2021/695): As the EU's primary instrument for research and innovation, Horizon Europe would likely fund the upstream R&D components associated with the Centres. This includes the development of new AI models, testing methodologies for frontier and physical AI, and the research necessary to address the "grand challenges" outlined in Annex I of CADA. The Centres would act as implementation nodes for these research outcomes.
  2. Digital Europe Programme (Regulation (EU) 2021/694): This programme, which already funds the EDIHs, would support the deployment and uptake aspects. Funding would likely cover the operational costs of providing access to AI factories, testing facilities, and the practical support services for SMEs, SMCs, and public sector bodies. It would also support the development of the common cloud and AI curriculum mentioned in Article 5(3)(b).

The proposal entrusts the implementation of these initiatives to the Commission and Member States, potentially through joint undertakings or other structures (Article 6(1)). This implies that while the legal framework is EU-wide, the actual disbursement of funds would flow through the established management structures of these two major programmes.

National Contributions and State Aid Compliance (Recital 29)

While Union programmes provide the backbone, CADA explicitly envisions a co-financing model involving national budgets. Recital 29 clarifies the role of Member States:

"In addition to receiving funding under Union programmes, the Cloud and AI Leadership Initiatives may be supported by Member States through research, development and innovation measures, in line with the applicable State aid rules, ensuring that national policies and Union policy are mutually consistent, as well as through private-sector investments."

This clause introduces two critical constraints and opportunities:

  • State Aid Rules: Any national public funding directed toward the Centres for AI must comply with EU State aid regulations. This ensures that public subsidies do not distort competition within the single market. For example, if a Member State provides direct grants to a Centre that then offers services to private companies, those grants must be notified to the Commission or fall under a block exemption regulation.
  • Policy Consistency: National measures must be "mutually consistent" with Union policy. This means national funding strategies for the Centres must align with the objectives of CADA, such as strengthening technological sovereignty and reducing dependencies on third-country providers.

Private Sector Investment

Recital 29 also highlights the role of the private sector, stating that "private-sector stakeholders should be encouraged to take into consideration the Cloud and AI Leadership Initiatives when developing their investment strategies." This suggests a public-private partnership model where private capital complements public funding, particularly for scaling up technologies developed within the Centres or for co-investing in specific "grand challenge" projects.

What this means for you

For public-sector procurement officers, IT strategists, and SME leaders, understanding this funding structure is critical for planning and resource allocation.

1. Access to Subsidised Services

Since the Centres for AI would be supported by a mix of Union and national funds, the services they offerβ€”such as AI testing, skills training, and access to compute resourcesβ€”may be available to public bodies and SMEs at reduced costs or free of charge.

  • Action: Monitor your national Centre for AI (likely evolving from your local EDIH) for calls for participation in pilot projects or funding opportunities. Do not assume these services will be fully commercial; they are designed to be accelerators.

2. Strategic Alignment for Funding

The Centres are designed to align with national cloud and AI strategies (Article 7).

  • Action: Ensure that your local Centre's activities are mapped to your national strategy. If your national strategy prioritises specific sectors (e.g., healthcare or manufacturing), aligning your Centre's focus with these priorities will maximise the likelihood of securing continued national and EU funding.

3. Compliance with State Aid

If your public body plans to fund the establishment or operation of a Centre for AI using national budget lines, you must ensure compliance with State aid rules.

  • Action: Consult with your national competition authority or legal team. Determine if specific subsidies (e.g., grants for equipment or operational costs) require Commission notification or if they fall under existing block exemptions. Failure to comply could result in the recovery of funds.

4. Leveraging the "Grand Challenges"

The Centres are tasked with supporting the "grand challenges" in Annex I (e.g., environmental sustainability, frontier AI, physical AI).

  • Action: If your organisation is working on projects in these areas, engage with the Centres early. They are the designated nodes for accessing Horizon Europe and Digital Europe funding streams related to these specific challenges.

Common misconceptions

"CADA creates a new, standalone budget for AI Centres."

  • Reality: CADA does not create a new EU budget line specifically for the Centres for AI. Instead, Article 6(3) directs existing funding streamsβ€”Horizon Europe and the Digital Europe Programmeβ€”towards these initiatives. The funding is integrated into broader, pre-existing EU programmes.

"The EU funds 100% of the Centres for AI."

  • Reality: Funding is shared. Recital 29 explicitly states that Member States may support these initiatives through national RDI measures. The EU provides the framework and co-funding, but national budgets play a crucial role in sustainability and localisation. Private investment is also encouraged.

"Only private companies can access these centres."

  • Reality: Article 5(2)(b) explicitly states that the centres shall accelerate adoption "notably for SMEs, SMCs and public sector bodies." Public authorities are key target beneficiaries, particularly for the adoption of sovereign AI solutions and for simplifying administrative procedures.

"The Centres are entirely new entities replacing the EDIHs."

  • Reality: Recital 25 and Article 5(1) make it clear that the Centres "shall build on" the existing European Digital Innovation Hubs. The funding model relies on repurposing and expanding the existing network rather than building from scratch.

Related

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