← current version of this provision

AI Act Article 20 — held text

This page is a held version, frozen at the capture below. It is a source for review. It is not advice, and it does not determine whether this provision applies to you.

Article 20 Corrective actions and duty of information 1. Providers of high-risk AI systems which consider or have reason to consider that a high-risk AI system that they have placed on the market or put into service is not in conformity with this Regulation shall immediately take the necessary corrective actions to bring that system into conformity, to withdraw it, to disable it, or to recall it, as appropriate. They shall inform the distributors of the high-risk AI system concerned and, where applicable, the deployers, the authorised representative and importers accordingly. 2. Where the high-risk AI system presents a risk within the meaning of Article 79(1) and the provider becomes aware of that risk, it shall immediately investigate the causes, in collaboration with the reporting deployer, where applicable, and inform the market surveillance authorities competent for the high-risk AI system concerned and, where applicable, the notified body that issued a certificate for that high-risk AI system in accordance with Article 44, in particular, of the nature of the non-compliance and of any relevant corrective action taken.
CELEX: 32024R1689 · provision: 20
Locator: Article 20; PDF page 64; derived-text line 1673
Held artifact: docs_evidence/eu-legislation/eurlex-full-text/20260823T093000Z/32024R1689.pdf
Artifact SHA-256: bba630444b3278e881066774002a1d7824308934f49ccfa203e65be43692f55e
Captured: 20260823T093000Z
Extracted with: pdftotext version 4.00
Official source: EUR-Lex
This version: /celex/32024R1689/ART_20/20260823-bba63044/
Corpus Merkle root: 581628a5b36518981241d2b51f048aa46384f30a74bf648ef542c652028cdeff
Membership proof: 10 hashes — see corpus-proof.json
Modal verbs, counted verbatim (not a legal characterisation): 'shall not': 0 · 'shall': 3 · 'must': 0 · 'may not': 0 · 'may': 0 · 'should': 0

To verify independently: fetch the official source above, extract it with the named tool, and confirm the artifact SHA-256 matches. Then check this provision's leaf against the published Merkle root using the proof in corpus-proof.json.