The EU has adopted new rules for dealing with sexualized AI deepfakes. Image: DPA
June 29, 2026, 4:57 p.mJune 29, 2026, 5:23 p.m
In the EU, AI applications for the abusive creation of sexualized deepfakes will be banned in the future. The member states gave the green light to the new rule, which will apply from December 2, 2026. The step was considered a formality after the agreement with the European Parliament. The EU AI Office should take the lead in enforcing the ban.
Deepfakes involve manipulating or creating deceptively real-looking image, video or sound recordings with the help of artificial intelligence. The new ban is intended to supplement the previous rules by targeting not only users but also providers of the corresponding digital tools.
What exactly should be forbidden
The new law should precisely define what is sexualized content and what is not. The ban is therefore aimed at AI applications that allow the creation of realistic images or videos of the intimate area – in particular of genitals, the anus, buttocks or exposed female breasts, as well as sexually explicit acts.
Satirical caricatures should therefore remain excluded. Non-consensual bikini pictures, such as those created by Grok and distributed on X, could therefore remain permitted. The new ban should also explicitly target the creation of content that depicts child sexual abuse.
Mandatory watermarks for AI will come later
Further changes to the AI law are intended to initially save the AI industry from additional obligations and also make it easier to use artificial intelligence in business. The companies behind ChatGPT, Claude and Co. now only have to comply with parts of the law that the European AI Office was originally supposed to enforce from August onwards from December 2026. By then at the latest, providers should clearly mark AI content as such – so they would have to mark generated images and videos with watermarks. (sda/awp/dpa)