AI Watermark Center
Regulation

EU AI Act & AI-Generated Content

How the EU AI Act’s Article 50 transparency rules address AI-generated content, machine-readable marking, deepfake labelling, and the Code of Practice — informational, not legal advice.

Published
Last verified
Review
Reviewed against primary sources

Under the EU Artificial Intelligence Act, Article 50 sets transparency obligations for certain AI systems that interact with people or generate synthetic content. Those duties apply from 2 August 2026. This hub summarises the rules that matter for AI watermarks, Content Credentials, deepfake labelling, and related transparency measures — using official EU sources.

What Article 50 covers

Article 50 sits in the AI Act’s transparency chapter. It addresses several situations, including informing people when they interact with an AI system, machine-readable marking of synthetic audio, image, video or text, information duties for certain emotion-recognition and biometric categorisation systems, deepfake disclosure, and disclosure for certain AI-generated text published on matters of public interest.

Providers versus deployers

The Act distinguishes providers (who place AI systems on the market or put them into service) from deployers (who use an AI system under their authority, outside purely personal non-professional use). Machine-readable marking of synthetic outputs is primarily framed as a provider duty. Deepfake labelling and certain public-interest text disclosures are deployer duties. Mixing those roles is a common source of confusion.

Machine-readable marking and detection

Providers of systems that generate synthetic audio, image, video or text must ensure outputs are marked in a machine-readable format and detectable as artificially generated or manipulated, as far as technically feasible. That legal concept is not the same as any single product brand such as C2PA or SynthID. Technical standards and voluntary codes can inform how organisations implement marking; they do not, by themselves, rewrite the Regulation.

Deepfakes and public-interest text

Deployers of systems that generate or manipulate deepfake image, audio or video content must disclose that the content was artificially generated or manipulated, subject to described exceptions (including certain artistic contexts and law-enforcement authorisations). Separate deployer rules apply to AI-generated or manipulated text published to inform the public on matters of public interest, again with stated exceptions such as human editorial responsibility.

Code of Practice

The Commission facilitates a Code of Practice on Transparency of AI-generated Content to support practical implementation of marking and labelling duties. Adherence to the Code is voluntary. Article 50 obligations remain legal obligations whether or not an organisation signs the Code.

How this relates to AI Watermark Center

AI Watermark Center inspects characters and file signals that are present in a browser session. It does not certify EU AI Act compliance, detect SynthID, or run a video inspector. Use the tools below for technical inspection; use official EU materials for legal requirements.

Inspect embedded Content Credentials

Cluster pages

Sources

  1. 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.

  2. 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.

  3. 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.

  4. 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).

Related