Starting Sunday, August 2, European Union citizens must be notified whenever they interact with an artificial intelligence system or encounter content that the technology has created or modified, according to a report from Wired. The new transparency requirements arrive as part of the bloc's Artificial Intelligence Act, which takes effect this weekend. The legislation aims to cut down on deception and manipulation while helping people make informed decisions about AI use.
The disclosure mandate spans a wide range of applications, according to the report. Advertisements or corporate social media content employing deepfakes will carry labels identifying them as AI-generated. Chatbots and complaint hotlines must clearly state if they're AI-powered. Call centers that use machine learning to track caller emotions and identify frustration will need to announce this practice at the call's beginning. Even business-to-business uses—including appointment scheduling, correspondence management, or contract negotiation—will require disclosure. Companies that fail to comply face penalties of up to €15 million ($17 million) or 3 percent of global annual revenue, whichever amount is larger. The law also covers model developers, who will face oversight from the European Commission's newly created AI Office, including firms like Spotify for its AI-driven recommendations and Adobe for AI editing tools in Photoshop.
"If a provider's really compliant with the transparency obligations, it will become very visible how much AI is used," technology lawyer Frederiek Fernhout told Wired, adding "especially in marketing." A colleague at the same firm noted that once you start examining daily interactions, AI appears everywhere, and all such content will need labeling going forward. However, critics warn the regulation's broad scope could create information overload. The AI policy lead at the Computer & Communications Industry Association cautioned in December that excessive labeling might trigger "banner blindness" with constant notifications, and if everything from spell-checked emails to filtered photos requires labels, AI content disclosure will become meaningless.
The new law echoes the EU's 2018 General Data Protection Regulation, which transformed how people experience the web by introducing widespread cookie-consent pop-ups and sparking a compliance scramble among businesses, the report notes. Similar adoption and enforcement challenges are likely here, according to Fernhout, who says many terms remain unclear and will need specific guidance from authorities, while the technical nature of requirements means they must be integrated into underlying systems. The EU has granted a transition period until December for AI model providers to label synthetic audio, images, video, or text in machine-readable formats, acknowledging the technical difficulty. Other compliance deadlines have been pushed back as the European AI Office has taken longer than anticipated to produce key guidelines. The bloc's 27 member countries are at different stages of establishing supervisory frameworks, meaning enforcement "might not be as day one immediate or uniform as we think," according to data and AI regulatory lawyer Rosie Nance.
Despite uneven early enforcement, the regulation's sweeping reach will ultimately prove to be a "gamechanger" for companies, according to the report. Businesses will face a choice: accept the transparency of labeling their AI use, revert to non-AI systems, return to manual processes, or risk taking a chance on disclosure while continuing to use AI extensively. The decision will reveal just how deeply artificial intelligence has become embedded in daily operations—and whether companies are willing to make that dependence visible to customers and partners.