EU AI Act Article 50 Is Live: You Must Tell Users "You're Talking to an AI"

The transparency obligations under Article 50 of the EU AI Act took effect on August 2, 2026. The rule is simple but far-reaching: when a person interacts with an AI system, they must clearly be made aware of it.

What does the obligation cover?

• Chatbots and voice assistants: users must be told they're talking to an AI, not a human.
• AI-generated content: machine-generated text, image, audio and video should be marked as such.
• Additional labeling requirements for deepfakes and synthetic media.

What should enterprise teams do?

Every customer-facing AI flow now needs an 'this is an AI' notice, a traceable record for generated content, and a path to hand off to a human when needed. Rather than embedding these into every application one by one, enforcing them centrally in a single orchestration layer makes both compliance and auditing far easier.

The remot angle

When governance rules like transparency, masking and audit logging are scattered across application code, proving compliance is hard. Defining these rules in the orchestration layer gives you consistent, auditable enforcement across every model and every flow.