Article 50 EU AI Act — Transparency duties
Art. 50 is the practical section for marketing, e-commerce, communications, and media: when must AI content and AI interactions be disclosed?
The four duty areas at a glance
- Art. 50(1) — Interaction: disclose chatbots and direct AI communication.
- Art. 50(2) — Machine-readable marking: duty for providers of generative systems (technical implementation).
- Art. 50(3) — Biometrics/emotion: information duty for corresponding systems.
- Art. 50(4) — Deepfakes & AI texts: visible disclosure by deployers upon publication.
Official sources
Check specific content under Art. 50
Web, social, print, phone, or internal — scenario or decision tree.
Start checkArt. 50 — FAQ
- What does Article 50 of the EU AI Act regulate?
- Art. 50 sets transparency duties: users should recognize when they interact with AI or see, hear, or read AI-generated or manipulated content.
- Which four areas matter most for deployers?
- Disclose chatbots and AI interactions, disclose deepfakes, label AI texts on public interest (without editorial responsibility), and — for providers — machine-readable marking of synthetic content.
- When does Art. 50 apply?
- Core transparency duties apply from 2 August 2026. Machine-readable marking by providers of legacy generative systems has a transition until 2 December 2026.
Orientation based on the EU AI Act (Art. 50). Not legal advice. For borderline cases consult a qualified lawyer. Effective from 02.08.2026 · ai-act-checker.com