EU bans AI nudification apps in record AI Act deal

European Union member states and the European Parliament have reached a provisional agreement to ban AI systems that create non-consensual sexually explicit content, commonly known as "nudification" apps, as part of amendments to the EU's landmark Artificial Intelligence Act. The new rules, which were negotiated in record time following the controversy surrounding Elon Musk's Grok chatbot earlier this year, will prohibit the marketing and use of AI tools designed to generate sexualized deepfake images of real people without their consent, as well as content depicting child sexual abuse. Companies will have until December 2, 2026 to bring their systems into compliance with the new prohibition. The agreement also delays the implementation of obligations for high-risk AI systems, including those involving biometrics, critical infrastructure, education, employment, law enforcement, and border management, until December 2, 2027, pushing back the original August 2 deadline. This extension was proposed by the European Commission as part of its "simplification" package aimed at reducing regulatory burden on businesses, though critics have accused EU lawmakers of giving in to pressure from the technology industry. The mandatory watermarking of AI-generated content will apply from December 2, 2026. The ban on sexual deepfakes was prompted by incidents earlier this year when users of X exploited Grok to generate and disseminate millions of sexually explicit AI-manipulated images of women and children online. According to the Center to Counter Digital Hate, X users created three million sexualized images in just 11 days using the tool. The amendments now require formal approval from the European Parliament's plenary and EU governments, a step considered largely procedural, with the hope that final ratification will occur before the August 2 deadline when the original full implementation of the AI Act was scheduled to begin.