Lawmakers in Europe have decided to delay critical aspects of the EU AI Act, aimed at regulating artificial intelligence, and have supported banning nudify apps. The European Parliament approved measures to extend deadlines for developers of high-risk AI systems — those posing significant risks to health, safety, or fundamental rights — until December 2027. Developers of AI systems governed by specific safety rules, such as toys or medical devices, have a proposed deadline of August 2028. Requirements for watermarking AI-generated content will be postponed until November 2026. Originally, these measures were to be implemented this August.
The Parliament also agreed to include a nudify apps ban in the updated AI Act, though specifics are not yet detailed. The ban will exclude AI systems with effective safeguards against creating such images. The decision follows significant backlash in the EU regarding Grok’s sexualized deepfakes earlier this year.
This vote extends uncertainty for businesses operating in Europe, which have experienced delays pending the EU’s postponed release of key guidelines and modifications to the law. It remains uncertain if these proposed changes will be enacted before the initial August deadline since the European Parliament cannot modify European law alone. Parliament must negotiate with the European Council, which consists of ministers from all 27 member states, to finalize the text.