European UnionAI Act, Articles 88 to 94 (enforcement of the obligations of providers of general-purpose AI models)
Regulation (EU) 2024/1689, Articles 88 to 94
In force since .
An AI governance rule binding public and private bodies.
- How the hook was established
- express
- What makes it apply
- Market targeting
- Obligation class
- Reporting, Governance
- Audit expectation
- on_request
- Who audits it
- Regulator
As of .
What it requires
- If the Commission requests, under Article 91(1), the documentation you drew up in accordance with Articles 53 and 55 or any additional information necessary to assess your compliance, supply the information requested within the period stated in the request, or have your representative supply it. A lawyer duly authorised to act may supply it on your behalf, and you remain fully responsible if it is incomplete, incorrect or misleading.
- If the Commission requests, for an evaluation under Article 92, access to your model through APIs or further appropriate technical means and tools, including source code, provide the access within the period stated in the request, or have your representative or the persons authorised to represent you provide it on your behalf.
- If the Commission requests under Article 93(1) that you take appropriate measures to comply with Articles 53 and 54, implement mitigation measures where an evaluation has given rise to serious and substantiated concern of a systemic risk at Union level, or restrict the making available on the market, withdraw or recall your model, comply with the measure requested.
- If you provide a model with systemic risk, you may offer commitments to implement mitigation measures to address a systemic risk at Union level during the structured dialogue the AI Office may initiate before a measure is requested. The Commission may, by decision, make them binding and declare that there are no further grounds for action.
- You have the procedural rights that Article 18 of Regulation (EU) 2019/1020 gives economic operators, applied mutatis mutandis. Before the Commission fines you under Article 101 it must communicate its preliminary findings to you and give you an opportunity to be heard.
These lines are LexLint's reading of the law's text, as of .
If you get it wrong
Private right of actionNo
Penalty structure
Article 101(1): the Commission may fine a provider of a general-purpose AI model up to 3% of its annual total worldwide turnover in the preceding financial year or EUR 15,000,000, whichever is higher, where it finds that the provider intentionally or negligently infringed the relevant provisions of the Regulation (point (a)), failed to comply with a request for a document or for information under Article 91 or supplied incorrect, incomplete or misleading information (point (b)), failed to comply with a measure requested under Article 93 (point (c)), or failed to make access to the model available for an evaluation under Article 92 (point (d)).
- Rule
- Higher of
- As of
- Currency
- EUR
- Fixed cap
- 15,000,000
- Turnover percentage cap
- 3
Who enforces it
Enforcement body
The European Commission, which holds exclusive powers to supervise and enforce Chapter V of the Regulation and entrusts the implementation of those tasks to the AI Office.
Settledness
- As of
- Open questions
- Does the Commission's power under Article 91(1) to request the documentation drawn up in accordance with Articles 53 and 55 reach a provider whose model was placed on the market before , for which Article 111(3) allows until to take the necessary steps to comply?
What it makes you log
Who may demand the log
Regulator
Logging duty
Article 91(1) lets the Commission request the documentation the provider drew up in accordance with Articles 53 and 55, which cannot be produced without a record of it. Articles 88 to 94 name no log, record or retention period of their own, and the content and keeping of the technical documentation are set by Article 53(1), point (a), and Annex XI.
- Kind
- Implicit
- As of
- Provision
- Article 91(1)
What this law does
The Commission has exclusive powers to supervise and enforce Chapter V, taking into account the procedural guarantees under Article 94, and entrusts the implementation of these tasks to the AI Office.
Without prejudice to Article 75(3), market surveillance authorities may request the Commission to exercise the powers laid down in the Section on supervision of providers of general-purpose AI models, where that is necessary and proportionate to assist with the fulfilment of their tasks under the Regulation.
The AI Office may take the necessary actions to monitor the effective implementation of and compliance with the Regulation by providers of general-purpose AI models, including their adherence to approved codes of practice.
Downstream providers have the right to lodge a complaint alleging an infringement of the Regulation, and the complaint must be duly reasoned and indicate at least the point of contact of the provider of the model concerned, a description of the relevant facts and provisions, and any other information the downstream provider considers relevant.
The scientific panel may provide a qualified alert to the AI Office where it has reason to suspect that a general-purpose AI model poses concrete identifiable risk at Union level or meets the conditions referred to in Article 51. Upon a qualified alert from the scientific panel, the Commission, through the AI Office and after having informed the Board, may exercise the powers laid down in that Section for the purpose of assessing the matter.
The Commission may request the provider of the general-purpose AI model concerned to provide the documentation drawn up by the provider in accordance with Articles 53 and 55, or any additional information that is necessary for the purpose of assessing compliance of the provider with the Regulation. Before sending a request for information, the AI Office may initiate a structured dialogue with the provider of the general-purpose AI model.
Upon a duly substantiated request from the scientific panel, the Commission may issue a request for information to a provider of a general-purpose AI model, where access to the information is necessary and proportionate for the fulfilment of the tasks of the scientific panel under Article 68(2).
A Commission request for information to a provider of a general-purpose AI model must state the legal basis and the purpose of the request, specify what information is required, set a period within which the information is to be provided, and indicate the fines provided for in Article 101 for supplying incorrect, incomplete or misleading information.
The provider of a general-purpose AI model, or its representative, must supply the information the Commission requests under Article 91, and lawyers duly authorised to act may supply information on behalf of their clients, who remain fully responsible if the information supplied is incomplete, incorrect or misleading.
The AI Office, after consulting the Board, may conduct evaluations of the general-purpose AI model concerned to assess the provider's compliance with obligations under the Regulation where the information gathered under Article 91 is insufficient, or to investigate systemic risks at Union level of models with systemic risk, in particular following a qualified alert from the scientific panel.
The Commission may decide to appoint independent experts to carry out evaluations on its behalf, including from the scientific panel established pursuant to Article 68. For the purposes of an evaluation, the Commission may request access to the general-purpose AI model concerned through APIs or further appropriate technical means and tools, including source code.
A Commission request for access to a general-purpose AI model must state the legal basis, the purpose and reasons of the request and set the period within which the access is to be provided, and must indicate the fines provided for in Article 101 for failure to provide access.
A provider of a general-purpose AI model, or its representative, must supply the access the Commission requests for an evaluation under Article 92, and where the provider is a legal person the persons authorised to represent it by law or by its statutes provide the access on its behalf.
The Commission adopts implementing acts setting out the detailed arrangements and the conditions for the evaluations, including the arrangements for involving independent experts and the procedure for selecting them.
Prior to requesting access to a general-purpose AI model, the AI Office may initiate a structured dialogue with the provider to gather more information on the internal testing of the model, internal safeguards for preventing systemic risks, and other internal procedures and measures the provider has taken to mitigate such risks.
Where necessary and appropriate, the Commission may request providers to take appropriate measures to comply with the obligations set out in Articles 53 and 54, to implement mitigation measures where an evaluation has given rise to serious and substantiated concern of a systemic risk at Union level, or to restrict the making available on the market, withdraw or recall the model.
Before the Commission requests a measure from a provider of a general-purpose AI model, the AI Office may initiate a structured dialogue with the provider. If, during the structured dialogue before a measure is requested, the provider of a model with systemic risk offers commitments to implement mitigation measures to address a systemic risk at Union level, the Commission may, by decision, make those commitments binding and declare that there are no further grounds for action.
Article 18 of Regulation (EU) 2019/1020 applies mutatis mutandis to the providers of the general-purpose AI model, without prejudice to more specific procedural rights provided for in the Regulation. The Commission may fine a provider of a general-purpose AI model up to 3% of its annual total worldwide turnover in the preceding financial year or EUR 15,000,000, whichever is higher, where it finds that the provider intentionally or negligently infringed the relevant provisions of the Regulation.
Article 101(1), points (b) to (d), applies the same ceiling of 3% of annual total worldwide turnover or EUR 15,000,000, whichever is higher, where the Commission finds that the provider intentionally or negligently failed to comply with a request for a document or for information under Article 91, or supplied incorrect, incomplete or misleading information, failed to comply with a measure requested under Article 93, or failed to make available access to the model for an evaluation under Article 92.
Before fining a provider of a general-purpose AI model under Article 101, the Commission communicates its preliminary findings to the provider of the general-purpose AI model and gives it an opportunity to be heard. The Court of Justice of the European Union has unlimited jurisdiction to review decisions of the Commission fixing a fine under Article 101, and may cancel, reduce or increase the fine imposed.
Article 101 applies from , the date from which the Regulation applies, because Article 113 excepts it from the earlier date given to Chapter V.
When LexLint raises it
When your app profile says your app trains models.