Summary No. Under the proposed Cloud and AI Development Act (CADA), an acceleration centre and a data centre acceleration zone are different things despite the shared word. An Experience and Acceleration Centre for AI (a "Centre for AI") is a support hub that would help organisations adopt cloud and AI — built on the existing European Digital Innovation Hubs and established under Article 5 as proposed. A data centre acceleration zone is a geographic area a Member State would designate under Article 10 to fast-track the physical deployment of data centres. One accelerates adoption and skills; the other accelerates construction and permitting. CADA is a proposal, not yet in force.

Detail

CADA, as proposed, uses the word "acceleration" for two unrelated mechanisms that sit at different layers of the cloud and AI value chain. For public-sector buyers and digital leaders, telling them apart matters for who you engage, what you can expect, and how you reference them in strategy and tender documents.

1. Experience and Acceleration Centres for AI (the "soft" layer)

The "Experience and Acceleration Centres for AI" ("Centres for AI") are established under Article 5 as part of the Cloud and AI Leadership Initiatives (Title II). They are not data centres. As proposed, each Member State would establish these centres, building on the European Digital Innovation Hubs set up under Article 16 of Regulation (EU) 2021/694 and any successor entities (Article 5(1)).

What Centres for AI would do:

  • Objectives (Article 5(2)): support the integration and scaling-up of AI use cases in strategic industrial and public sectors; accelerate broad adoption of cloud and AI at regional and local level, notably for SMEs, small mid-caps (SMCs) and public sector bodies, in line with the "AI first" principle; and leverage infrastructure to help develop and fine-tune AI models.
  • Tasks (Article 5(3)): help organisations access and use AI technologies — including by connecting them with European providers of cloud and AI; provide access to upskilling and reskilling schemes in collaboration with the AI Skills Academy; facilitate the transfer of expertise across regions; and support the scaling-up of spin-offs and start-ups.
  • Network (Article 5(6)): a network of Centres for AI would be established to share best practice and provide specialised services across regions where the required skills or compute are not available locally.

In short, a Centre for AI is a service hub for knowledge, skills and matchmaking — the "soft" infrastructure of the ecosystem. (The broader policy aim of promoting adoption through this network also appears among the operational objectives of the Leadership Initiatives in Article 4.)

2. Data centre acceleration zones (the "hard" layer)

By contrast, data centre acceleration zones sit in Title III ("Data centre capacities") and are governed mainly by Article 10. These are physical, geographic areas within a Member State.

What acceleration zones would do:

  • Designation (Article 10(1)): where data centre capacity is being deployed in its territory, a Member State would designate at least one acceleration zone — as proposed, within six months of entry into force. In designating zones, Member States would consider factors such as site location and dimensions, current and future power-grid capacity and on-site clean energy generation, network connectivity, potential to reuse waste heat, measures to speed up permitting, a preference for brownfield over greenfield sites, and the ability to function sustainably.
  • Conditions (Article 11): Member States would set sustainability requirements using the key performance indicators in Delegated Regulation (EU) 2024/1364, and ensure access to zone resources on fair, reasonable and non-discriminatory terms.
  • Faster permitting (Article 13): the permit-granting procedure for projects in a zone "shall not exceed 12 months, from the moment a comprehensive application has been submitted" (Article 13(5)), supported by an aggregated baseline permit issued for the zone (Article 13(2)).
  • Single information point (Article 12): operators would have the right, on request, to assistance from a single information point across the project's lifecycle.

In short, a zone is a "hard" infrastructure designation: a place where regulatory hurdles are lowered to encourage construction.

Why the names confuse people

Both mechanisms aim to "accelerate" the digital transition, but at different layers:

  • Centres for AI accelerate adoption, skills and matchmaking (human and organisational capacity).
  • Acceleration zones accelerate construction and deployment (physical infrastructure capacity).

They tackle different bottlenecks: a shortage of compute capacity (zones) and a shortage of AI adoption and skills (centres).

What this means for you

For public-sector procurement officers and digital leaders, the distinction is practical.

1. Who to engage. If you want to upskill staff, pilot AI, or find European cloud and AI providers, your entry point is a Centre for AI (Article 5(3)). If you are planning or procuring large-scale data centre capacity, look to providers building in acceleration zones (Article 10), which would offer faster permitting and better grid alignment.

2. National strategies. As proposed, Article 7 requires Member States to adopt national cloud and AI strategies, which must include measures to support both broad AI deployment (with Centres for AI as entry points, Article 7(2)(b)) and the deployment of data centre capacity (Article 7(2)(d)). You may need to reference these national measures in tenders to show alignment.

3. Funding and support. The Leadership Initiatives may draw funding from Union programmes such as Horizon Europe and the Digital Europe Programme (Article 6(3)). Infrastructure projects, meanwhile, may pursue separate "data centre strategic project" status designated by the Commission under Article 14, which can unlock Union and Member State support.

4. Sovereignty in procurement. When procuring cloud services, your required sovereignty tier comes from a risk assessment (Article 29) and is met by services recognised at the relevant Union assurance level (Article 17), with criteria set out in Annex II. The two "acceleration" mechanisms do not themselves confer assurance levels — but Centres for AI can help you find recognised European providers, and the physical location of data centres can bear on the data-localisation criteria for higher levels.

Common misconceptions

Misconception 1: "Acceleration zones are where AI models are trained." A zone is a regulatory and geographic designation for infrastructure deployment. Data centres located there may host hardware used for AI, but the zone does not itself imply any specific AI activity. Skills and adoption support sits with Centres for AI.

Misconception 2: "Any data centre can be in an acceleration zone." No. Member States designate specific zones against the factors in Article 10(1) — grid capacity, connectivity, sustainability and more. Not every site or project is in a zone.

Misconception 3: "Centres for AI are physical labs for testing hardware." As proposed they are primarily advisory and support hubs that build on existing innovation-hub networks (Article 5(1)). They may facilitate access to infrastructure and help fine-tune models (Article 5(2)(c)), but they are not standalone hardware-testing facilities.

Misconception 4: "The terms are interchangeable in EU law." They are not. "Experience and Acceleration Centres for AI" (Article 5) and "data centre acceleration zones" (Article 10) are distinct concepts with different obligations. Using the wrong term in a tender or strategy paper risks misalignment.

Official sources

Related

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