[co-author: Virginie Liebermann]
On November 5 2025, the European Commission commenced a seven-month long, multi-stakeholder process to develop a voluntary code of practice (the Code) facilitating compliance with transparency obligations under Article 50 of the European Union Artificial Intelligence Act (the AI Act).
A first draft of the Code is anticipated in December 2025, with a second draft scheduled for around March 2026. The final publication of the Code is expected between May and June 2026. The Code will set voluntary standards for the labelling, disclosure and traceability of AI-generated outputs. Article 50 of the AI Act already requires providers and deployers of generative AI systems to clearly mark and label AI-generated and manipulated content (e.g. deepfakes and other synthetic material) so that persons know when they are interacting with an AI system.
The Code is expected to be effective from August 2 2026, prior to transparency obligations under the AI Act becoming applicable to AI systems such as chatbots, emotion-recognition tools, voice assistants, image/video synthesis tools and content-generation systems.
The press releases are available here and here, the public consultation (which closed on October 9 2025) is here and further information on the Code is here.
[View source.]
