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.
- CELEX
- 02024R1689-20260727
- Official Journal
- OJ L, 2024/1689, 12.7.2024
- In force
- 1 Aug 2024
EU AI Act
02024R1689-20260727
Ready
Third-party governance
Draft1 obligation mapped
Processor due diligence, contract terms and monitoring
- Art. 25
AI governance
Draft14 obligations mapped
AI system register, roles and assessments
- Art. 10
- Art. 11
- Art. 12
- Art. 13
- Art. 14
- +9
Enterprise risk
Draft1 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
| Article | Heading | Pritect suite |
|---|---|---|
| Art. 9 | Risk management system | Enterprise risk |
| Art. 10 | Data and data governance | AI governance |
| Art. 11 | Technical documentation | AI governance |
| Art. 12 | Record-keeping | AI governance |
| Art. 13 | Transparency and provision of information to deployers | AI governance |
| Art. 14 | Human oversight | AI governance |
Obligations of providers and deployers
Chapter III, Section 3, Art. 16 to 27
| Article | Heading | Pritect suite |
|---|---|---|
| Art. 16 | Obligations of providers of high-risk AI systems | AI governance |
| Art. 17 | Quality management system | AI governance |
| Art. 18 | Documentation keeping | AI governance |
| Art. 19 | Automatically generated logs | AI governance |
| Art. 20 | Corrective actions and duty of information | AI governance |
| Art. 21 | Cooperation with competent authorities | AI governance |
| Art. 25 | Responsibilities along the AI value chain | Third-party governance |
| Art. 26 | Obligations of deployers of high-risk AI systems | AI governance |
| Art. 27 | Fundamental rights impact assessment for high-risk AI systems | AI governance |
Transparency obligations
Chapter IV, Art. 50
| Article | Heading | Pritect suite |
|---|---|---|
| Art. 50 | Transparency obligations for providers and deployers of certain AI systems | AI governance |
02 · Timeline
When it applies.
Adopted by the Parliament and the Council
13 Jun 2024
Published in the Official Journal
12 Jul 2024
Entered into force
1 Aug 2024
Art. 113
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.
