From today, August 2, the transparency obligations of Article 50 of the European AI Act apply. They come down to two simple ideas: when a person interacts with an AI system, they have to know it, and artificially generated or manipulated content must be marked so that its synthetic origin is detectable.
They apply immediately to systems placed on the market from today. Systems already in operation have a transitional period for content marking until December 2, 2026.
For most companies, which use AI rather than build it, the translation is concrete: if a chatbot talks to your customers, it must be clear that it is a machine, and if you publish AI-generated content, it has to be marked. High-risk system obligations do not arrive until December 2027, so there is no need to chase them yet.
What starts today, what does not, and what a managing director should do this week: we covered it in AI Act, August 2: what takes effect and what doesn't.
