What transparency duties does the AI Act set for chatbots and AI-generated content?

A few, and they apply whatever the risk class. People must be told when they are interacting with an AI system, such as a customer service chatbot, unless it is obvious. Providers of systems that generate synthetic audio, images, video or text must mark the output in a machine-readable way so that it can be detected as artificially generated. Deployers who publish deepfakes must disclose that the content is artificial, and those who publish AI-generated text on matters of public interest must disclose it too, unless a person has reviewed it and taken editorial responsibility. Deployers of emotion recognition or biometric categorisation systems must inform the people exposed. For a company, the practical steps are an inventory of where AI talks to customers or produces published content, a clear notice in those places, and a check that your suppliers mark their output. These duties apply on the date set in the Act’s final provisions. NetSenX publishes its own AI transparency page for the AI features in its service.

General information, not legal advice. National laws can add to the EU texts quoted here, for example deadlines, forms or reporting channels: check your own country's rules.

← All 20 questions on EU law: GDPR, DORA, CRA and AI ActAll topics