EU lawmakers closed a political agreement on May 7, 2026, writing the bloc’s first explicit, named AI application prohibition directly into law. Nudifier apps, AI tools that generate sexualized or intimate imagery of real people without their consent, are now barred under the EU AI Act framework, according to European Parliament official news. The prohibition is integrated via amendment to the existing Act, alongside the high-risk compliance deadline extension covered separately in today’s companion brief.
The ban is precise in scope. It targets AI systems capable of generating pornographic deepfakes or intimate imagery without the depicted person’s consent, as confirmed by independent reporting from Deutsche Welle and RTE. What the ban does not do is resolve every adjacent question: platforms that host but don’t generate such imagery, open-source models with general image capabilities, and API providers whose systems could be misused all sit in gray territory that enforcement guidance will need to address.
For compliance officers, the operative deadline is August 2, 2026, the date the EU AI Office gains full enforcement powers, including the right to demand model access for safety audits. That date is separate from, and unaffected by, the high-risk system deadline extension in the same Omnibus deal. Organizations whose products can generate intimate imagery need a legal review now, before August 2, not after.
The amendment mechanism matters as much as the prohibition itself. By naming nudifier apps as a banned category in statute, rather than leaving them to guidance documents or enforcement discretion, the EU has established a legislative template. Future targeted prohibitions can follow this path: identify a specific application, define the harm, write the red line into the Act. A group of MEPs, according to Politico EU, issued a formal warning alongside the deal that EU legal frameworks remain ill-equipped for the most capable AI tools currently in restricted use, and called on ENISA to gain evaluation access to such systems. The warning signals that even as Parliament extends one deadline and enacts one prohibition, some legislators see a structural gap between what the law now covers and what the AI Office can realistically enforce by August.
Cross-jurisdictionally, the EU ban arrives while the US Take It Down Act targets similar conduct through a different mechanism, and the UK Online Safety Act addresses non-consensual intimate imagery under platform obligations. Three frameworks, one category of harm, and none of them interoperable. For organizations operating across jurisdictions, the compliance burden is additive, not unified.
The deeper compliance question isn’t whether nudifier apps are covered, they clearly are. It’s whether the API and platform layers that make such outputs possible have a clear enough view of how the prohibition applies to general-purpose systems with intimate-imagery capability. That question won’t be answered by the political agreement. It will be answered by the AI Office’s first enforcement guidance, expected before August 2. Watch for that document: it will draw the line between the named prohibition and the broader ecosystem.