In forceEU-wideEU AI Act

AI Act: chatbot, deepfake and AI-content transparency rules (Art. 50) apply; national enforcement starts

36 days ago (In force)

Date note: Kept at 2026-08-02 by the Digital Omnibus on AI; only the marking/detection duty for systems already on the market before this date is deferred to 2026-12-02

What changes

Users must be told when they interact with an AI system (chatbots, voice bots). AI-generated or manipulated images, audio, video and text must be marked in a machine-readable way, and deepfakes must be labelled. Since the same date national authorities can fine companies for most AI Act breaches.

Who is affected

Any company that offers a customer-facing chatbot, generates synthetic media or text for the public, or uses emotion recognition or biometric categorisation.

Sizes: micro, small, medium, large · Sectors: All sectors · Applies if: We use or build AI systems

What to do

Add a clear notice to chatbots and voice assistants that the user is talking to AI. Make sure generative tools you deploy embed watermarks or metadata in outputs and label deepfakes visibly. Update your privacy and terms pages and train customer-facing staff on the disclosure rules.

Penalty

Up to €15M or 3% of worldwide turnover (SMEs: lower of the two)

Sources

Last verified 7 September 2026. Informational only, not legal advice.

More EU AI Act deadlines

86 days
ConfirmedEUEU AI Act

AI Act: end of grace period for marking AI-generated content in pre-existing systems; ban on non-consensual intimate deepfake generators

Grace period introduced by the Digital Omnibus on AI (Reg. 2026/1744) for systems placed on the market before 2026-08-02; the new prohibition was also added by the Omnibus

Providers of generative AI products released before August 2026 that have not yet implemented watermarking; any business deploying image or video generators.