The AI Act's transparency obligations started to apply on 2 August 2026, with the European Commission's AI Office and national authorities enforcing from that date. Deepfakes, meaning images, video or audio generated or edited with AI, have to be labelled; AI generated or altered content has to carry machine readable marks so it can be detected; and chatbots and other interactive systems have to tell users they are not talking to a person. The Commission published a voluntary Code of Practice on marking and labelling on 10 June that providers and deployers can sign as their route to compliance.