EU AI Act: New Commission draft on AI transparency
On 8 May 2026, the European Commission published draft guidelines on the transparency obligations in Article 50 AI Act and opened a targeted stakeholder consultation. These guidelines are highly relevant because Article 50 applies from 2 August 2026 and affects many everyday AI use cases far beyond “high‑risk” systems.
The draft explains when and how transparency is required, and who is responsible:
1) Informing users when they are interacting with an AI system (e.g. chatbots, AI agents, voice assistants – Article 50(1))
2) Machine‑readable marking and detectability of AI‑generated or AI‑manipulated content (text, audio, image or video – provider obligation under Article 50(2))
3) Disclosure of the use of emotion recognition and biometric categorisation systems (Article 50(3))
4) Visible labelling of deep fakes and certain AI‑generated or AI‑manipulated texts published with the purpose of informing the public on matters of public interest (Article 50(4))
The draft guidelines also signal how authorities are likely to assess compliance in practice.
A voluntary Code of Practice on marking and labelling AI‑generated content is expected in June 2026 and will complement the guidelines.
If your organisation uses AI to interact with people or to generate content in the EU, Article 50 is likely already relevant. The draft guidelines are essential reading to understand what “transparency” will concretely mean under the AI Act.