Skip to main content
    Pritect

    Regulation hub

    Run EU AI Act on one record, across every suite

    The EU Artificial Intelligence Act is the first horizontal law governing artificial intelligence. It sorts AI systems by the risk they pose, bans a short list of practices outright, and puts detailed duties on the organisations that build high-risk systems and on the organisations that deploy them.

    Book a demo
    CELEX
    02024R1689-20260727
    Official Journal
    OJ L, 2024/1689, 12.7.2024
    In force
    1 Aug 2024
    RegulationEU AI Act added to your scopeDemo data

    EU AI Act

    02024R1689-20260727

    Draft

    Ready

    1. Third-party governance

      Draft

      1 obligation mapped

      Processor due diligence, contract terms and monitoring

      • Art. 25
    2. AI governance

      Draft

      14 obligations mapped

      AI system register, roles and assessments

      • Art. 10
      • Art. 11
      • Art. 12
      • Art. 13
      • Art. 14
      • +9
    3. Enterprise risk

      Draft

      1 obligation mapped

      Risk register, assessments and treatment

      • Art. 9

    EU AI Act is now mapped across 3 suites. Each mapping waits for your AI governance lead to confirm it.

    01 · Article to module

    Every obligation lands in a suite.

    The regulation runs to 113 articles. Pritect decomposes 16 of them clause by clause into obligations you can adopt, own and evidence: the requirements that attach to a high-risk AI system, the duties that fall on the provider that builds one and on the organisation that deploys it, and the transparency duties that apply whether or not a system is high risk. The rest are deliberately not decomposed, because they are not duties you discharge. Articles 1 to 8 set the subject matter, scope, definitions, prohibited practices and the high-risk classification rules, Articles 22 to 24 address authorised representatives, importers and distributors, Articles 28 to 49 govern notifying authorities, notified bodies, standards and conformity assessment, Articles 51 to 56 cover general-purpose AI models, Articles 57 to 72 cover innovation measures, governance, the EU database and post-market monitoring, and Articles 74 to 113 govern market surveillance, remedies, penalties and the final provisions. Article 15 and Article 73 are catalogued at clause level but sit outside the article list the AI governance review confirmed for this hub, so they are not listed here. The set grows as the catalogue is extended.

    Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence and amending certain Union legislative acts (Artificial Intelligence Act)

    Article headings are reproduced verbatim from the Official Journal and stay in English in every language, because a citation has to remain quotable. They are transcribed from the published text; the subject-matter review that makes them authoritative inside the product is still pending. Read the text on EUR-Lex

    Requirements for high-risk systems

    Chapter III, Section 2, Art. 9 to 14

    Art. 9Risk management systemEnterprise risk
    Art. 10Data and data governanceAI governance
    Art. 11Technical documentationAI governance
    Art. 12Record-keepingAI governance
    Art. 13Transparency and provision of information to deployersAI governance
    Art. 14Human oversightAI governance

    Obligations of providers and deployers

    Chapter III, Section 3, Art. 16 to 27

    Art. 16Obligations of providers of high-risk AI systemsAI governance
    Art. 17Quality management systemAI governance
    Art. 18Documentation keepingAI governance
    Art. 19Automatically generated logsAI governance
    Art. 20Corrective actions and duty of informationAI governance
    Art. 21Cooperation with competent authoritiesAI governance
    Art. 25Responsibilities along the AI value chainThird-party governance
    Art. 26Obligations of deployers of high-risk AI systemsAI governance
    Art. 27Fundamental rights impact assessment for high-risk AI systemsAI governance

    Transparency obligations

    Chapter IV, Art. 50

    Art. 50Transparency obligations for providers and deployers of certain AI systemsAI governance

    02 · Timeline

    When it applies.

    1. Adopted by the Parliament and the Council

      13 Jun 2024

    2. Published in the Official Journal

      12 Jul 2024

    3. Entered into force

      1 Aug 2024

      Art. 113

    4. General application begins

      2 Aug 2026

      Art. 113

    03 · Questions

    What teams ask first.

    Which AI systems count as high risk?

    Article 6 sets the classification rules. A system is high risk where it is a safety component of a product covered by other Union law, or where it falls in one of the areas listed in Annex III, such as employment, education, essential services, law enforcement and migration. Pritect records the classification and the reasoning behind it against each system.

    Are we a provider or a deployer?

    The Act puts different duties on each. A provider develops a system, or has one developed, and places it on the market under its own name. A deployer uses a system under its own authority. Putting your own name on a system you bought can make you a provider under Article 25, which is why Pritect asks the question per system rather than per organisation.

    Do we need a fundamental rights impact assessment?

    Article 27 requires one from deployers that are public bodies or that provide public services, and from deployers of certain creditworthiness and insurance pricing systems, before a high-risk system is put into use. Pritect runs the screening and keeps the assessment attached to the system it belongs to.

    When do the obligations start to apply?

    Article 113 staggers commencement. The general date of application is the one shown in the timeline above, with the prohibited practices and the general provisions applying earlier, and the rules for high-risk systems that are safety components of regulated products applying later. Pritect tracks which date governs each system you have recorded.

    What has to be disclosed to people?

    Article 50 requires that people are told when they are interacting with an AI system, that synthetic audio, image, video and text is marked as artificially generated, and that emotion recognition and biometric categorisation are disclosed to the people exposed to them. These duties apply whether or not the system is high risk.

    This page summarises publicly available legal text so you can orient yourself. It is not legal advice.

    See EU AI Act mapped to your own programme.

    Book a demo