Get ready for the EU AI Act Transparency Obligations for AI Systems

Imagine your company wants to roll out an AI-powered tool across Europe. The technology performs well, the business case is clear, and the launch is ready, until a fundamental question is raised:

Do users know they are interacting with AI?

This question is about to become critical for many organisations. From 2 August 2026, (most of) the transparency obligations under Article 50 of the EU AI Act will apply. These rules extend beyond the EU and will also affect Swiss companies operating in or targeting the EU market.

They introduce specific and practical requirements, including informing users when they interact with AI, clearly identifying AI-generated content, and addressing high-risk use cases such as deepfakes, emotion recognition, and biometric categorisation systems. For many organisations, the key challenge lies in implementation rather than interpretation: identifying where AI is used, determining what must be disclosed, and understanding which exceptions apply.

If these questions have not yet been addressed, now is the time to act.

In their latest newsletter, Dr. Rehana Harasgama and Lara Burkhalter provide an overview of the key transparency obligations under Article 50 AI Act, explain the most important exceptions, and highlight the practical measures Swiss companies should take to ensure compliance when operating in or targeting the EU market.