Skip to content

Regulation Tracker / EU AI Act / Chapter VI

Article 60a
Testing of high-risk AI systems covered by Union harmonisation legislation listed in Section B of Annex I in real-world conditions outside AI regulatory sandboxes

Chapter VI — AI regulatory sandboxes. 7 distinct duties, powers or definitions are coded from this article, applying from 2026-08-02.

Inserted by Regulation (EU) 2026/1744 (Digital Omnibus on AI), in force 2026-07-27. The text below is the consolidated version of 27 July 2026; the amending regulation are on EUR-Lex.

official text: EUR-Lex (consolidated 27 Jul 2026) · text © European Union (Decision 2011/833/EU) · coding CC BY 4.0 · data: obligations.json · Hugging Face · GitHub

The text

1. Member States may allow, in accordance with this Article, the testing of high-risk AI systems in real world conditions outside AI regulatory sandboxes by providers or prospective providers of AI enabled products covered by the Union harmonisation legislation listed in Section B of Annex I, with a view to assessing and verifying the conformity of those systems with the requirements laid down in Articles 8 to 15.

2. Member States that choose to allow testing as referred to in paragraph 1 shall, individually or jointly, adopt frameworks for real-world testing.

3. Each Member State shall notify the Commission of any real-world testing framework it adopts before implementing it. This shall not affect the competences of the Commission under the Union harmonisation legislation listed in Section B of Annex I.

4. Member States that have adopted real-world testing frameworks shall ensure that the relevant national competent authorities, relevant authorities and public authorities responsible for the management and operation of infrastructure and products covered by Union harmonisation legislation listed in Section B of Annex I cooperate closely with each other in good faith and remove any practical obstacles, including on procedural rules providing access to physical public infrastructure, where this is necessary, to successfully implement those real-world testing frameworks and test AI-enabled products covered by Union harmonisation legislation listed in Section B of Annex I.

5. The frameworks for real-world testing shall lay down the requirements under which testing in real-world conditions shall occur. Those frameworks shall:

  1. (a)include the provision of a mandatory real-world testing plan to be agreed between the provider or prospective provider and the national competent authority or relevant authority in accordance with the Union harmonisation legislation listed in Section B of Annex I;
  2. (b)ensure compliance with the requirements laid down in Article 60(2), (3), (4)(d)-(j) and (5)-(9), where any reference to market surveillance authorities in those provisions shall be read as a reference to the national competent authority or relevant authority, as appropriate in accordance with the Union harmonisation legislation listed in Section B of Annex I;
  3. (c)include effective governance and accountability arrangements;
  4. (d)ensure a high level of protection of health safety and fundamental rights.

6. The real-world testing shall comply with the applicable provisions laid down in the Union harmonisation legislation listed in Section B of Annex I. Any requirements laid down in those provisions shall not affect the application of this Article to the extent necessary to enable the testing referred to in paragraph 1.

What it requires, coded

SafeLegalAI's reading of each duty in this article: who, what, from when, under which fine tier, and whether it reaches a firm, chambers, court or legal-AI vendor. Descriptive, not advice; the quoted words are the Regulation's.

  1. Art. 60a(1)Support measureHigh-risk

    Member States may allow real-world testing outside sandboxes by providers or prospective providers of AI-enabled products covered by Annex I Section B harmonisation legislation to assess and verify conformity with Articles 8 to 15.

    "Member States may allow, in accordance with this Article, the testing of high-risk AI systems in real world conditions outside AI regulatory sandboxes by providers or prospective providers of AI enabled products"
    Who
    Member State
    From
    (Art. 113 second subparagraph)
    Legal practice
    Institutional — Mostly indirect for legal practice unless a legal-AI system is part of an Annex I Section B product.
    See
    Annex I Section B · Art. 8 · Art. 15

  2. Art. 60a(2)Governance institutionalHigh-risk

    Member States that choose to allow Article 60a testing must, individually or jointly, adopt frameworks for real-world testing.

    "Member States that choose to allow testing as referred to in paragraph 1 shall, individually or jointly, adopt frameworks for real-world testing"
    Who
    Member State
    From
    (Art. 113 second subparagraph)
    Legal practice
    Institutional — Framework duty has indirect relevance to legal-AI providers in regulated products.
    See
    Art. 60a(1)

  3. Art. 60a(3)Cooperation with authoritiesHigh-risk

    Each Member State must notify the Commission of any real-world testing framework it adopts before implementing the framework, without affecting Commission competences under Annex I Section B harmonisation legislation.

    "Each Member State shall notify the Commission of any real-world testing framework it adopts before implementing it"
    Who
    Member State
    From
    (Art. 113 second subparagraph)
    Legal practice
    Institutional — Institutional notification with indirect legal-practice relevance.
    See
    Commission · Annex I Section B

  4. Art. 60a(4)Governance institutionalHigh-risk

    Member States with real-world testing frameworks must ensure relevant competent, relevant and public infrastructure authorities cooperate closely in good faith and remove practical obstacles, including procedural barriers to physical public infrastructure access when necessary.

    "shall ensure that the relevant national competent authorities, relevant authorities and public authorities responsible for the management and operation of infrastructure and products covered by Union harmonisation legislation listed in Section B of Annex I cooperate closely with each other in good faith"
    Who
    Member State, National competent authority
    From
    (Art. 113 second subparagraph)
    Legal practice
    Institutional — Mostly indirect for legal practice unless legal-AI products depend on regulated public infrastructure.
    See
    Annex I Section B

  5. Art. 60a(5)RequirementHigh-risk

    Article 60a frameworks must ensure compliance with listed Article 60 testing safeguards, include effective governance and accountability arrangements, and ensure a high level of health, safety and fundamental-rights protection.

    "ensure compliance with the requirements laid down in Article 60(2), (3), (4)(d)-(j) and (5)-(9)"
    Who
    Member State
    From
    (Art. 113 second subparagraph)
    Fine tier
    Art. 99(4)
    Legal practice
    Institutional — Indirect for legal-AI providers unless they test AI-enabled regulated products under Annex I Section B.
    See
    Art. 60(2) · Art. 60(3) · Art. 60(4)(d)-(j) · Art. 60(5)-(9)

  6. Art. 60a(5) (a)DocumentationHigh-risk

    Article 60a real-world testing frameworks must require a mandatory testing plan agreed between the provider or prospective provider and the national competent or relevant authority under Annex I Section B harmonisation legislation.

    "include the provision of a mandatory real-world testing plan to be agreed between the provider or prospective provider and the national competent authority or relevant authority"
    Who
    Member State
    From
    (Art. 113 second subparagraph)
    Fine tier
    Art. 99(4)
    Legal practice
    Institutional — Indirect for legal-AI providers unless they test AI-enabled regulated products.
    See
    Annex I Section B

  7. Art. 60a(6)RequirementHigh-risk

    Real-world testing under Article 60a must comply with applicable provisions of Annex I Section B Union harmonisation legislation, without those provisions preventing Article 60a testing to the extent necessary.

    "The real-world testing shall comply with the applicable provisions laid down in the Union harmonisation legislation listed in Section B of Annex I"
    Who
    Provider
    From
    (Art. 113 second subparagraph)
    Fine tier
    Art. 99(4)
    Legal practice
    Institutional — Indirect for legal-AI providers unless covered by Annex I Section B harmonisation legislation.
    See
    Annex I Section B

Cite

Regulation (EU) 2024/1689, Article 60a (Testing of high-risk AI systems covered by Union harmonisation legislation listed in Section B of Annex I in real-world conditions outside AI regulatory sandboxes), as amended by Regulation (EU) 2026/1744, consolidated text of 27 July 2026, https://eur-lex.europa.eu/eli/reg/2024/1689/2026-07-27/eng#art_60a — text © European Union; only the Official Journal is authentic. Coding: SafeLegalAI (published by Cognesio LLP), "EU AI Act, structured", safelegalai.com/regulation/eu-ai-act/article-60a, accessed 2026-09-08, CC BY 4.0.