Summary Under the proposed Cloud and AI Development Act (CADA), Estonia would be required to establish a network of Experience and Acceleration Centres for AI (Centres for AI). These centres would not be built from scratch but would instead build upon Estonia's existing European Digital Innovation Hubs (EDIHs). Their primary role would be to act as local entry points for SMEs, small mid-caps (SMCs), and public sector bodies to adopt cloud and AI technologies, connect with European providers, and access upskilling schemes. These centres would serve as the operational arm of Estonia's national cloud and AI strategy, ensuring alignment with the EU-wide network and the "AI first" principle.

Detail

The proposed Cloud and AI Development Act (CADA), COM(2026) 502 final, introduces a comprehensive framework to strengthen the EU's cloud and AI ecosystem. A critical component of this framework is the decentralised support structure designed to drive adoption at the regional and local levels. For Estonia, as for all Member States, this translates into a specific legislative obligation to establish and maintain a network of Experience and Acceleration Centres for AI.

The Legal Basis: Article 5 of the Proposal

The mandate for these centres is explicitly set out in Article 5 of the CADA proposal. Article 5(1) states: "Each Member State shall establish Experience and Acceleration Centres for AI ('Centres for AI')."

Crucially, the proposal avoids duplicating existing infrastructure. Instead, it mandates continuity and efficiency. Article 5(1) specifies that these Centres for AI "shall build on the European digital innovation hubs established under Article 16 of Regulation (EU) 2021/694 and, where applicable, any successor entities established under Union law."

For Estonia, this means the proposed Centres for AI would likely evolve from the country's current network of EDIHs. This approach ensures that the new centres leverage existing expertise, physical premises, and established relationships with local businesses and public bodies, rather than creating parallel administrative structures.

Objectives: Bridging the Adoption Gap

The CADA proposal defines clear, actionable objectives for these centres, aimed at closing the gap between advanced AI research and practical deployment in the real economy. According to Article 5(2), the objectives of the Centres for AI are to:

  • Support integration and scaling: Assist in the integration and scaling-up of AI use cases in strategic industrial and public sectors.
  • Accelerate adoption: Accelerate the broad adoption of cloud and AI technologies at regional and local levels, notably for SMEs, SMCs and public sector bodies, in line with the 'AI first' principle.
  • Leverage infrastructure: Leverage relevant infrastructure to accelerate the development and fine-tuning of AI models and systems.

The explicit mention of SMEs, SMCs, and public sector bodies highlights that these centres are designed to support entities that may lack the in-house resources to navigate the complex AI and cloud landscape independently.

Specific Tasks: Connecting, Upskilling, and Scaling

Article 5(3) details the specific tasks these centres must undertake. For Estonian public bodies, SMEs, and SMCs, the following tasks are particularly significant:

  1. Connecting with European Providers: The centres are tasked with "helping organisations accelerate their digital transformation through access to and use of AI technologies, including by connecting organisations with European providers of cloud and AI technologies." This task directly supports CADA's broader sovereignty goals by steering public and private procurement toward EU-based solutions, reducing reliance on non-European incumbents.
  2. Upskilling and Reskilling: The centres must "ensure or provide access to relevant upskilling and reskilling schemes, in close collaboration with the AI Skills Academy." This is vital for public bodies needing staff capable of managing AI systems and for SMEs needing to upskill their workforce to compete in an AI-driven market.
  3. Facilitating Expertise Transfer: They are tasked with "facilitating the transfer of expertise across regions." This ensures that best practices, technical knowledge, and successful use cases developed in one part of Estonia or the EU can be shared widely, preventing silos.
  4. Supporting Start-ups and Spin-offs: The centres must "support the scaling-up of spin-offs and start-ups emerging from universities, incubators and other accelerators by facilitating access to clients, companies and organisations seeking specialised AI services." This fosters a vibrant local innovation ecosystem.

The EU-Wide Network: Collaboration Beyond Borders

The CADA proposal envisions these centres not as isolated national entities but as nodes in a cohesive European network. Article 5(6) states: "A network of Centres for AI shall be established to support collaboration and the exchange of best practices among Centres for AI, and to provide specialised services across regions where the required skills or compute capacity are not available locally."

This network mechanism is critical for Estonia. If a specific niche expertise or high-performance compute capacity is not available locally, the Estonian Centre for AI can tap into the broader EU network to provide these services to local organisations. This promotes cross-border cooperation, prevents duplication of effort, and ensures that even smaller Member States have access to the full spectrum of EU capabilities.

Link to the National Strategy: Article 7

The establishment of Centres for AI is not an isolated measure; it is intrinsically linked to Estonia's national cloud and AI strategy. Article 7 of the CADA proposal requires Member States to establish national cloud and AI strategies within one year of the regulation's entry into force.

Article 7(2)(b) explicitly mandates that these national strategies must include "measures to accelerate the development and adoption of cloud and AI at national, regional and local level, particularly among public sector bodies, SMEs and SMCs, including by supporting the Centres for AI referred to in Article 5 as entry points to the European AI innovation ecosystem."

Therefore, Estonia's national strategy would be required to explicitly detail how it will support, fund, and integrate these Centres for AI. The centres would act as the operational arm of the national strategy at the local level, translating high-level policy goals into tangible support for businesses and public bodies.

Governance and Autonomy

While the CADA sets the framework, it allows for flexibility in implementation to suit local contexts. Article 5(5) notes that 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 means that while the objectives and core tasks are fixed by EU law, the specific operational model, staffing structures, and local partnerships in Estonia can be tailored to fit the national and regional context. Furthermore, Article 5(4) empowers the Commission to adopt implementing acts detailing the procedure for establishing Centres for AI, including selection criteria and participant organisation profiles, ensuring quality and operational standards across the Union.

What this means for you

For public-sector procurement officers, administrators, and business leaders in Estonia, the proposed establishment of Experience and Acceleration Centres for AI has several practical implications:

  1. A Single Point of Contact: These centres would serve as your primary local resource for understanding and implementing AI and cloud solutions. Instead of navigating the complex EU landscape alone, you could turn to your local Centre for AI for guidance, training, and connection to providers.
  2. Support for Sovereign Procurement: As CADA introduces new sovereignty requirements and assurance levels for cloud services, the Centres for AI could help you understand which European providers meet the necessary criteria. They could facilitate matchmaking between public buyers and innovative European SMEs and start-ups, as encouraged by Article 33 of the proposal.
  3. Skills Development: The centres would offer upskilling and reskilling opportunities in collaboration with the AI Skills Academy. Public bodies should actively engage with these programmes to ensure their staff are equipped to manage AI projects, conduct risk assessments, and oversee human-in-the-loop processes as required by the AI Act and CADA.
  4. Strategic Alignment: Engaging with these centres would demonstrate compliance with the 'AI first' principle and support the broader goal of technological sovereignty. Procurement officers should be aware that the use of these centres is a key part of Estonia's national strategy under Article 7.
  5. Access to EU Resources: If local expertise is lacking, the network nature of these centres means you could access specialised services from other parts of the EU. This is particularly useful for niche AI applications or advanced cloud architectures that may not be available domestically.

Common misconceptions

Misconception 1: These centres are only for large tech companies. Reality: Article 5(2)(b) explicitly states that the centres aim to accelerate adoption "notably for SMEs, SMCs and public sector bodies." They are specifically designed to be accessible to smaller entities and public administrations that may lack in-house AI expertise.

Misconception 2: Estonia must build entirely new institutions from scratch. Reality: Article 5(1) clearly states that the Centres for AI "shall build on the European digital innovation hubs." This means Estonia can upgrade and expand its existing EDIH network rather than creating parallel structures, ensuring efficiency and continuity.

Misconception 3: The centres dictate which specific cloud providers you must use. Reality: The centres help connect organisations with European providers, but they do not mandate specific commercial choices. Their role is to facilitate access, provide expertise, and ensure that providers meet the necessary sovereignty and security standards set out in CADA. The final procurement decision remains with the contracting authority, guided by the sovereignty framework.

Misconception 4: These centres replace national regulatory authorities. Reality: The Centres for AI are support and acceleration bodies. They do not have regulatory or enforcement powers. Regulatory oversight remains with the national competent authorities designated under Article 25 of the CADA proposal.

Official sources

Related

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