Article 10 of 1139%
Article 10: Data and data governance
Praxikon tracks Article 10 (Data and data governance) under the EU AI Act, citing the source for every statement.
EU Official:
UpcomingAnnex III from 2 Dec 2027; Annex I from 2 Aug 2028
Title III: High-Risk AI Systems
Application dates
- : Requirements for high-risk systems listed in Annex III
- : Requirements for high-risk systems linked to Annex I
Article 10 requires training, validation and test data to be relevant, representative, error-free and complete. Datasets must take into account specific characteristics of the intended use context.
Official text
||
Source: EUR-Lex, Regulation (EU) 2024/1689. Text reproduced verbatim.
Download AI Act (PDF) →Official guidance on this article
2- Commission Implementing Decision C(2025) 3871 of 23.6.2025 on a standardisation request to CEN and CENELEC as regards high-risk AI systems in support of Regulation (EU) 2024/1689 and repealing Implementing Decision C(2023) 3215Requests standards for data and data governance so that the quality requirements for training, validation and test data from Article 10 become measurablePublishedImplementing actEuropean Commission23 Jun 2025
- prEN 18283 Managing bias in AI systems (draft European standard)Makes the requirement of Article 10 that datasets be examined for possible bias concrete by providing measurable criteria for identifying and addressing bias.DraftStandardCEN-CENELEC JTC 2127 May 2026
Compliance checklist
- Training data is relevant and representative
- Datasets checked for errors and bias
- Data governance procedures established
- Validation and test sets separated from training data
- Data privacy protection ensured
Want to save your progress? Create an account
Related tools
Related Recitals
Related enforcement
No enforcement actions for this article yet. Follow developments via the Enforcement Tracker.
Related blog posts
Cross-references
Annexes
Frequently asked questions
What requirements does the AI Act set for training data?+
Article 10 requires training, validation and test data to be relevant, representative, error-free and complete. Datasets must take into account specific characteristics of the intended use context.
What documentation does Article 10 of the AI Act require?+
Article 10 of the AI Act requires that relevant documentation is maintained as part of the compliance process. This may include technical documentation, instructions for use, logs or declarations of conformity, depending on the classification of the AI system.
How do I document compliance with Article 10 of the AI Act?+
You document compliance with Article 10 of the AI Act by establishing a risk management system, maintaining technical documentation, and conducting internal audits. Keep all relevant documents for the period prescribed by the AI Act.
May I use special category personal data for bias detection in training data?+
Yes, Article 10(5) explicitly allows processing of special categories of personal data (such as ethnicity or gender) for monitoring, detecting and correcting bias, provided this is strictly necessary, appropriate safeguards are in place, and it is not possible with synthetic or anonymised data.
What requirements does the AI Act set for the quality of training data?+
Article 10 requires that training, validation and test data are relevant, representative, as error-free as possible, and complete for the intended purpose. Datasets must have appropriate statistical properties, consider the specific context, and be free from bias that could lead to discrimination.
Do I need to document the provenance of my training data?+
Yes, Article 10 in combination with Annex IV requires you to document the provenance of training data, including how data was obtained, selected and preprocessed. This is essential for traceability and demonstrating compliance.