transparency obligations
In plain wordsEU rules requiring that people be told when they are dealing with AI: chatbots must say they are bots, AI-made content must be marked, and deepfakes must be labeled.
The duties in Article 50 of the EU AI ActEU AI Act, prop. n.: The European Union's law on artificial intelligence, the first broad AI law anywhere: it bans some uses, sets strict checks for risky ones and puts duties on the makers of the biggest AI models. for systems whose main risk is deception, applying from 2 August 2026 whether or not a system is high-risk. Providers must ensure that people talking to an AI system can tell it is one, unless that is plain, and that generated or altered audio, images, video and text carry a machine-readable mark of artificial origin, such as a watermarkAI watermarking, n.: Hiding an invisible signal in AI-generated text, images, audio or video so that software can later tell the content was made by AI — like a banknote's watermark, but undetectable to the eye. or metadata. Deployers must tell people when an emotion recognition systememotion recognition system, n.: AI that tries to work out how people feel or what they intend from their faces, voices or bodies. The EU bans it at work and in schools and tightly regulates it elsewhere. or biometric categorisationbiometric categorisation, n.: Using AI to sort people into groups — by age, sex, ethnicity and the like — from their faces, voices or other body features. The EU bans using it to guess sensitive traits such as religion or sexuality. system is used on them, label deep fakesdeepfake, n.: A fake but realistic video, photo or voice recording made with AI that puts words in a real person's mouth or shows them in a scene that never happened. (with lighter rules for clearly artistic, satirical or fictional work) and flag AI-written text published to inform the public on matters of public interest, unless a human has reviewed it and someone takes editorial responsibility. Generative systems on the market before August 2026 have until 2 December 2026 to add machine-readable marking. A voluntary Code of Practice on marking and labelling AI-generated content, published by the Commission on 10 June 2026 with common EU label icons, supports compliance.
To meet the transparency obligations, the customer-service bot now opens by saying it is an AI.
Origin
The 'limited risk' tier of the Commission's April 2021 proposal; marking duties for generative AI were strengthened in the 2023 negotiations after the release of ChatGPT and popular image generators. The grace period for marking was added by the 2026 Digital OmnibusDigital Omnibus, prop. n.: A European Union package, proposed in 2025, that simplified and delayed parts of its digital rules — above all pushing back the strictest requirements of the EU AI Act by more than a year..
Usage note
Distinct from the documentation duties of general-purpose AI modelgeneral-purpose AI model, n.: The EU's legal name for the large, adaptable AI models behind chatbots and similar tools: models that can turn their hand to many different jobs and be built into many other products. providers and from the broad principle of AI transparencyAI transparency, n.: Openness about AI: telling people when they are dealing with one, how a system was built and tested, and why it reached a particular decision.. The Act defines 'deep fake' in Article 3(60). Commission guidelines interpreting Article 50 were published on 20 July 2026.
Related
- AI transparencyAI transparency, n.: Openness about AI: telling people when they are dealing with one, how a system was built and tested, and why it reached a particular decision.
- AI watermarkingAI watermarking, n.: Hiding an invisible signal in AI-generated text, images, audio or video so that software can later tell the content was made by AI — like a banknote's watermark, but undetectable to the eye.
- AI-generated content labelingAI-generated content labeling, n.: Marking text, images, audio or video as made by AI — with a visible notice, hidden data in the file, or both — so that people and platforms can tell synthetic content from real.
- deepfakedeepfake, n.: A fake but realistic video, photo or voice recording made with AI that puts words in a real person's mouth or shows them in a scene that never happened.
- Content CredentialsContent Credentials, prop. n.: A tamper-evident 'nutrition label' attached to a photo, video or document recording where it came from — which camera or AI tool made it and how it was edited — that anyone can inspect.
- EU AI ActEU AI Act, prop. n.: The European Union's law on artificial intelligence, the first broad AI law anywhere: it bans some uses, sets strict checks for risky ones and puts duties on the makers of the biggest AI models.
- risk-based approachrisk-based approach, n.: A way of regulating AI in which the strictness of the rules depends on how much harm a use could cause: bans for the worst uses, heavy checks for risky ones, little or nothing for the rest.
Sources
- European Commission (2026). Commission publishes Code of Practice on marking and labelling AI-generated content. 10 June 2026. digital-strategy.ec.europa.eu/en/news/commission-publishes-code-practice-marking-and-labelling-ai-generated-content
Defined in law or standards
EU AI Act (Reg. (EU) 2024/1689), Art. 50; 'deep fake' Art. 3(60); transitional period as amended by Reg. (EU) 2026/1744
Mentioned in
EU AI ActEU AI Act, prop. n.: The European Union's law on artificial intelligence, the first broad AI law anywhere: it bans some uses, sets strict checks for risky ones and puts duties on the makers of the biggest AI models., text-to-speechtext-to-speech, n.: Technology that reads written text aloud in a computer-generated voice, used in screen readers, voice assistants, audiobooks and navigation apps., AI transparencyAI transparency, n.: Openness about AI: telling people when they are dealing with one, how a system was built and tested, and why it reached a particular decision., AI-generated content labelingAI-generated content labeling, n.: Marking text, images, audio or video as made by AI — with a visible notice, hidden data in the file, or both — so that people and platforms can tell synthetic content from real., black boxblack box, n.: A system whose inner workings are hidden or too complex to understand: we can see what goes in and what comes out, but not why it gave a particular answer., General-Purpose AI Code of PracticeGeneral-Purpose AI Code of Practice, prop. n.: A voluntary rulebook, published by the EU in 2025, that makers of large AI models can sign to show they follow the EU AI Act's rules on transparency, copyright and safety., risk-based approachrisk-based approach, n.: A way of regulating AI in which the strictness of the rules depends on how much harm a use could cause: bans for the worst uses, heavy checks for risky ones, little or nothing for the rest., AI Basic ActAI Basic Act, prop. n.: South Korea's main AI law, in force since January 2026, which sets safety and transparency duties for AI used in sensitive areas and for AI-generated content., biometric categorisationbiometric categorisation, n.: Using AI to sort people into groups — by age, sex, ethnicity and the like — from their faces, voices or other body features. The EU bans using it to guess sensitive traits such as religion or sexuality., emotion recognition systememotion recognition system, n.: AI that tries to work out how people feel or what they intend from their faces, voices or bodies. The EU bans it at work and in schools and tightly regulates it elsewhere., market surveillance authoritymarket surveillance authority, n.: The government body in each EU country that checks AI systems on the market, investigates complaints, and can order unsafe or unlawful AI to be fixed or withdrawn.