EU AI Office — Final Code of Practice on marking and labelling of AI-generated content (Article 50) published
Tag: S-2026-06-10-eu-ai-office-content-marking-code-final Type: press release (official European Commission news item) Author(s): European Commission — AI Office / DG CONNECT Date of source: 2026-06-10 (publication date stated on the official news page) Date ingested: 2026-06-12 Authority weight: high — official Commission news item fetched directly; cross-referenced to Commission press release IP/26/1328. Raw file: S-2026-06-10-eu-ai-office-content-marking-code-final.md. External URL: https://digital-strategy.ec.europa.eu/en/news/commission-publishes-code-practice-marking-and-labelling-ai-generated-content
What it claims
On 10 June 2026 the European Commission published the final Code of Practice on marking and labelling of AI-generated content (also referred to as the Code of Practice on transparency of AI-generated content), accompanied by press release IP/26/1328. The Code is voluntary and “sets out practical steps to help providers and deployers of generative artificial intelligence (AI) systems meet the AI Act transparency obligations that will apply from 2 August 2026.” This confirms the publication that the previous scan (S-2026-06-11-eu-ai-office-content-marking-code) flagged as scheduled/unconfirmed against the then-stale official policy page.
The news item restates the substantive Article 50 obligations the Code supports: from 2 August 2026 the AI Act will require clear labelling in key cases — deepfakes and AI-generated or AI-manipulated text published on matters of public interest must be clearly labelled, and users must be informed when they are interacting with an interactive AI system such as a chatbot. The Commission frames these transparency requirements as helping people recognise when content has been generated or altered by AI, “reducing the risk of deception and manipulation.”
The publication is accompanied by a supporting package: a set of EU Icons for labelling AI-generated content (signatories commit to visually disclose AI-generated markings using a standard set of generalised icons provided by the AI Office, together with information on whether content was created or manipulated using AI), guidance on how to sign the Code, Q&A documents on the Code and on the signature process, and a scheduled info session on the Code and signature process. The Article 50 applicability date of 2 August 2026 is unchanged.
Notable quotes
“The European Commission published the final Code of Practice on marking and labelling of AI-generated content.” — Commission news item, Publication 10 June 2026
“The Code is voluntary and sets out practical steps to help providers and deployers of generative artificial intelligence (AI) systems meet the AI Act transparency obligations that will apply from 2 August 2026.” — Commission news item, 10 June 2026
“Deepfakes and AI-generated or AI-manipulated text published on matters of public interest must be clearly labelled. Users must also be informed when they are interacting with an interactive AI system, such as a chatbot.” — Commission news item, 10 June 2026
What’s speculative vs. asserted
- Asserted (official Commission news item): the final Code was published on 10 June 2026; its voluntary status; that it provides practical steps for providers and deployers to meet Article 50 obligations; the substantive labelling/disclosure obligations (deepfakes, public-interest text, chatbot disclosure); the EU-icons-based visual disclosure for signatories; the 2 August 2026 Article 50 applicability date.
- Resolved from prior scan: the S-2026-06-11-eu-ai-office-content-marking-code capture recorded a secondary claim of a “~10 June 2026” publication as unconfirmed because the official policy page was then last updated 22 May 2026. This source confirms the publication and supersedes that open item.
- Speculative / not stated by source: whether and how many financial-services firms will sign; whether national competent authorities will treat Code adherence as a supervisory expectation (the Code is voluntary, so this is a downstream inference, not a Commission claim); the technical detail of the icons and marking methods (in the linked policy/icons pages, not the news item itself).
Topics this feeds
Open questions raised
- How does the final Code text differ from the 8 May 2026 draft guidelines and the second working-group draft once consultation feedback was incorporated? (Requires sight of the Code document itself, not just the news item.)
- Will financial-services regulators (EBA, FCA, national competent authorities) reference Code adherence as evidence of Article 50 compliance, or issue sector-specific guidance?
- How will the EU icons interoperate with industry technical standards for provenance and watermarking (e.g. C2PA)?