Providers of generative AI must mark synthetic output as artificially generated in a machine-readable way

Practical action

Implement watermarking or content credentials on generated output and keep test evidence that the mark survives common transformations.

Who does it apply to?

Providers of systems that generate synthetic media or text, whatever the risk tier; codes of practice under Article 50(7) will detail acceptable techniques.

Actors

Provider / developerGeneral-purpose AI model provider

Sectors

Cross-sector / all sectorsMedia, platforms and informationConsumer products and retail

Use cases

Generative AI and foundation modelsContent moderation and platforms

Applies from: 2 August 2026

Which controls meet this duty?

Satisfies: the control, operated properly, does the work the duty asks for. Supports: it contributes but the duty needs more. Each control page lists every other duty it serves, so work done once can be counted once.

• satisfiesTechnical measureEngineering lead · continuous
  Synthetic content labelling and provenance marking

  Serves 5 recorded duties · evidence: Content labelling and provenance standard, Watermark and provenance robustness test, Visible AI-generated content label

  Machine-readable marking with robustness evidence.

• supportsTechnical measureQuality or testing lead · at launch and on material change
  Accuracy, robustness, fairness and security testing

  Serves 15 recorded duties · evidence: Pre-release test report, Test plan and acceptance criteria, Release test sign-off

  Tests that the mark is robust and reliable.

Templates that cover this duty

Generated from the records. Free · no account · links emailed to your work address. This duty is cited in each file with its source reference and a link back here.

• EU AI Act Role and Risk Classifier

  Checklist · XLSX · v4

• AI Agent Registry and Permission Matrix

  Register · XLSX and DOCX · v6

• EU AI Act Article 50 Transparency Kit

  Kit · DOCX and XLSX · v44

• ISO/IEC 42001 Gap Assessment and Statement of Applicability

  Assessment · XLSX · v3

All 10 templates that cover it →

What evidence would a reviewer expect?

Framework mappings

Original editorial crosswalks. They cite clause numbers only and reproduce no standard text; confidence reflects how direct the mapping is.

See every European Union duty mapped this way →

Frequently asked questions

Cite this record

AIPolicyTracker (2026). “Providers of generative AI must mark synthetic output as artificially generated in a machine-readable way (EU AI Act)” [AIPT-OBL-0039, verified 26 September 2026]. https://aipolicytracker.org/id/AIPT-OBL-0039 (accessed 8 November 2026). Data licensed CC BY 4.0.

Permanent id AIPT-OBL-0039: it never changes or gets reused, and its address keeps working if this page moves.

Cite the official text alongside it: Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence, Official Journal of the European Union, https://eur-lex.europa.eu/eli/reg/2024/1689/oj.

@misc{aipolicytracker2026_providersofgenerativeaim,
  author       = {{AIPolicyTracker}},
  title        = {{Providers of generative AI must mark synthetic output as artificially generated in a machine-readable way (EU AI Act)}},
  year         = {2026},
  howpublished = {\url{https://aipolicytracker.org/obligations/eu-ai-act-art-50-2-synthetic-content-marking}},
  url          = {https://aipolicytracker.org/obligations/eu-ai-act-art-50-2-synthetic-content-marking},
  urldate      = {2026-11-08},
  note         = {Data licensed CC BY 4.0. Accessed 8 November 2026}
}

Change history

Similar obligations in other instruments

• Employers must inform workers and their representatives before using high-risk AI at work — EU AI Act, European Union
• Disclose AI interaction and label synthetic content — EU AI Act, European Union
• Deployers of emotion recognition or biometric categorisation must inform exposed persons — EU AI Act, European Union
• Deployers must disclose deepfakes and AI-generated text published on matters of public interest — EU AI Act, European Union
• Deployers must tell natural persons that a high-risk AI system is used in decisions about them — EU AI Act, European Union
• Deployers must explain individual decisions taken with high-risk AI on request — EU AI Act, European Union
• Frontier developers must publish a transparency report before deploying a new frontier model — California SB 53, California (United States)
• Provide deployers with clear instructions for use — EU AI Act, European Union