The European Parliament has voted to postpone certain rules in EU AI law related to high-risk AI applications, including biometrics. Members also supported banning the NudeDefyer app, which creates explicit images of real people without their consent.
The digital omnibus package on simplification was adopted by a large majority on Thursday, paving the way for negotiations on the final law with the European Council. According to the proposal, the compliance deadline for high-risk AI system developers would be extended from August 2027 to December 2, 2027.
The proposed delay comes after the European Commission missed a February deadline to publish guidelines on high-risk AI systems under the EU’s AI law. This guideline is intended to provide clarity for companies providing and deploying these systems.
“Companies now need to be clear about whether they are at high risk,” Alba Kokalari, co-rapporteur of the Internal Market and Consumer Protection Committee, said earlier this month. “If Europe wants to be competitive, it must invest more and make AI easier to use, rather than penalizing companies that introduce innovative AI features into secure products.”
Members of Parliament (MEPs) also supported allowing service providers to process special categories of personal data, such as biometrics, for bias detection only if “absolutely necessary”. This rule applies not only to providers and implementers of high-risk systems, but also to other AI systems and models.
High-risk AI applications include applications in the areas of biometrics, critical infrastructure, law enforcement, essential services, employment, justice and the management of democratic processes. Systems classified as “high risk” include biometrics, classification, and emotional recognition.
Sexual deepfakes to be banned
AI-generated content is also subject to the updated rules. The proliferation of stripping apps creates an urgent need for clear regulatory bans, the proposal said.
The proposed ban comes in the wake of outrage over digitally altered images of undressed women, some of them minors, created by X Company’s AI chatbot Grok. The platform, owned by Elon Musk, was the subject of a European Commission investigation in January.
However, the ban does not prevent AI providers from developing the technical capabilities to generate images and videos, the document said. Providers will still be required to watermark AI-generated content, but the rule will be delayed until November 2026.
The digital omnibus package on simplification also suggests extending compliance deadlines for companies developing AI systems subject to sector-specific safety regulations, such as toys and medical devices, to August 2028.
To support the revitalization of EU companies, the document recommends extending support measures to both small and medium-sized enterprises (SMEs) and small and medium-sized enterprises (SMCs).
The seventh omnibus package on simplification was proposed by the European Commission in November last year. The EU is currently considering other proposals in the package, including those related to data and the establishment of a European business wallet.
The digital simplification plan is backed by countries such as France and Germany, as well as technology companies and lobbying groups.
Article topics
AI Law | Biometrics | Deepfake | EU | Europe | Law
