EU Code of Practice for AI-Generated Content
What the Commission’s Code of Practice on Transparency of AI-generated Content is for, why adherence is voluntary, and how it relates to Article 50 legal obligations.
- Published
- Review
- Reviewed against primary sources
Short answer
The European Commission facilitates a Code of Practice on Transparency of AI-generated Content to help organisations implement Article 50 marking and labelling duties in practice. Adherence to the Code is voluntary. Article 50 obligations remain legal obligations whether or not an organisation signs or follows the Code. Commission materials describe the Code as supporting Article 50(2), (4) and (5), with a structure that addresses providers (marking and detection) and deployers (deepfake and public-interest text labelling). Guidelines on transparency obligations sit alongside the Code; following either does not erase the Regulation text.
Voluntary code versus legal Article 50
Article 50 is binding EU law with an application date of 2 August 2026 for the transparency obligations described in Commission FAQ materials. The Code is a practical cooperation instrument. Signing up may help demonstrate how an organisation approaches marking and labelling, as reflected in Commission communications about the relationship between guidelines, the Code, and compliance — but voluntary adherence is not the same as the legal duty disappearing.
Section 1 providers and Section 2 deployers
Commission documentation for the Code organises commitments around provider-facing marking and detection work (often described as Section 1) and deployer-facing labelling for deepfakes and certain public-interest text (Section 2). That split mirrors the Act’s provider/deployer distinction: technical detectability of synthetic outputs versus clear disclosure when content is presented to people.
Relationship to Commission guidelines
Separate Commission guidelines on transparency obligations clarify scope, definitions, and exemptions under Article 50. The Code operationalises practices; the guidelines interpret how obligations apply. Readers should treat EUR-Lex, the guidelines, the FAQ, and the Code as a stack — not as interchangeable substitutes.
Limitations
- Does not list every Code commitment or annex.
- Does not score organisations for Code adherence.
- Does not map any AI Watermark Center tool to a Code requirement.
Related tools
Local C2PA inspection can help understand embedded provenance metadata. It is not a Code-of-Practice conformity assessment.
Sources
Code of Practice on Transparency of AI-generated Content — European Commission (Shaping Europe’s digital future)
Accessed August 29, 2026.
Voluntary code supporting Article 50(2), (4) and (5). Adherence is voluntary; Article 50 obligations remain legal obligations. Section 1 providers (marking/detection); Section 2 deployers (deepfake and public-interest text labelling).
Transparency obligations under Article 50 of the AI Act — European Commission (Shaping Europe’s digital future)
Accessed August 29, 2026.
Commission FAQ clarifying provider/deployer duties, machine-readable marking, and relationship to the Code of Practice. Article 50 obligations apply from 2 August 2026.
Guidelines on transparency obligations for providers and deployers of certain AI systems — European Commission (Shaping Europe’s digital future)
Accessed August 29, 2026.
Commission guidelines on Article 50 scope, definitions, exemptions, and how adherence to the Code of Practice may demonstrate compliance for marking/labelling duties.
Regulation (EU) 2024/1689 (Artificial Intelligence Act) — EUR-Lex / Official Journal of the European Union
Published July 12, 2024. Accessed August 29, 2026.
Primary legislative text. Article 50 sets transparency obligations for providers and deployers of certain AI systems. Application timing follows Article 113 as amended.