Glossary

AI Transparency Obligation

Requirement under Article 50 of the EU AI Act to clearly inform users when they are interacting with an AI system, including chatbots and AI-generated content.

A

Article 50 of the EU AI Act requires that users are clearly informed when they interact with an AI system. This applies to chatbots, AI-generated content (text, images, audio, video) and emotion recognition systems. The goal is to prevent deception — people have the right to know when they are talking to a machine.

How to implement this:

  • Chatbots and virtual assistants: Add a clear label such as "You are chatting with an AI assistant" at the start of each conversation. Keep it visible, not hidden in terms of service.
  • AI-generated content: Label synthetic text, images or audio with a visible marker. For deepfakes or synthetic media, machine-readable metadata is also required.
  • In-product notices: Add an "AI-powered" badge next to features that use AI to generate recommendations, summaries or decisions.
  • Documentation: Record where in your product AI is used and how users are informed. This evidence is needed for compliance audits.

Transparency builds trust — users who know they are interacting with AI tend to set appropriate expectations and report issues more constructively. Tidal Control helps you map all AI touchpoints in your product and track transparency measures as auditable controls.

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

T

U

V

W

Z

Frequently asked questions

What is the AI transparency obligation?
It is the requirement under Article 50 of the EU AI Act to clearly inform people when they are interacting with an AI system or with AI-generated content.
Who does the AI transparency obligation apply to?
It applies to providers and deployers of AI systems that interact with people, such as chatbots; to deployers of emotion-recognition and biometric-categorisation systems; and to systems that generate synthetic audio, image, video or text.
What does the AI transparency obligation require in practice?
Users must be told they are dealing with AI unless it is obvious; providers must mark AI-generated content in a machine-readable way; and deployers must disclose deepfakes and AI-generated text published to inform the public on matters of public interest. These obligations apply from 2 August 2026.
← Back to all terms