EU Requires AI Chatbots to Disclose Their Nature

Why this matters

For developers and founders building on AI in Europe, this is a compliance deadline with real product implications, not a background policy shift. If your chatbot, complaint hotline, or emotion-detection tool touches an EU customer, you now need disclosure language built into the interaction itself, at the start of the call or the first message, not buried in a terms-of-service page nobody reads. That means engineering work: flagging AI-generated content, scripting disclosure prompts, auditing which features count as "AI" under the Act's definition.

Companies that treated transparency as a legal afterthought will find it's now a UX requirement with an August 2 hard date attached. We're also watching how strictly this gets enforced against smaller firms versus the usual targets like Meta or Google, since a rule this broad only matters if regulators actually chase down the call centers and ad-tech vendors quietly running sentiment analysis. For researchers, it's a useful natural experiment: once disclosure becomes mandatory, we'll finally get a sense of just how much everyday software, from Spotify recommendations to customer service lines, was already leaning on AI without saying so.