The European Parliament has taken another step towards revising the Artificial Ininformigence Act (AIA). The committees on the Internal Market and Civil Liberties voted by a large majority in favour of a simplification package that postpones certain deadlines, introduces new prohibitions, and creates more flexibility for businesses.
What is being postponed?
The core of the package is the postponement of application deadlines for high-risk AI systems. The background: important technical standards are not expected to be finalised in time for the original deadline of 2 August 2026. To avoid legal uncertainty, the Members of Parliament are proposing specific new dates.
- 2 December 2027 for high-risk AI systems explicitly listed in the legislation, including systems in the areas of biometrics, critical infrastructure, education, employment, law enforcement, justice, and border management.
- 2 August 2028 for AI systems falling under sector-specific EU product safety regulations or deployed as safety components in such products.
There is also a deadline extension on the subject of watermarking. Providers are to be given more time to label AI-generated audio, image, video, or text content accordingly. The Members of Parliament are proposing 2 November 2026 here — a shorter postponement than the February 2027 proposed by the Commission.
New Prohibition: AI “Nudifier” Apps
A key new element is the planned prohibition of so-called “nudifier” systems. These AI applications create or manipulate images to depict them in a sexually explicit or intimate manner while resembling an identifiable real person, without that person’s consent. The prohibition is not to apply where a system has effective safety mechanisms that effectively prevent the creation of such content.
“I am glad that it was possible to reach a compromise that is acceptable to the majority of Parliament and at least the centrist parties. And that this compromise included a proposal to ban so-called nudification apps, which I consider a legitimate expectation of our citizens,” stated Michael McNamara, co-rapporteur (Renew, Ireland).
More Flexibility for Businesses
The package also contains measures intfinished to create it simpler for European businesses to deal with the AI Act.
- Service providers may process personal data to detect and correct biases in AI systems, but only under strict conditions and where strictly necessary.
- Small Mid-Caps are to benefit in future from the same support measures as SMEs, in order to ease the transition as they grow.
- For products already subject to sector-specific EU safety regulations (such as medical devices, toys, or radio equipment), the requirements of the AI Act are to be less stringent, in order to avoid double regulation.
“We want predictable, simplified rules that eliminate overlaps with sectoral legislation and reduce fragmentation between Member States. If Europe wants to be competitive, we must increase investment and facilitate the apply of AI,” stated Arba Kokalari, co-rapporteur (EPP, Sweden).
What happens next?
The plenary of the European Parliament is expected to vote on the mandate on 26 March. Nereceivediations with the Council of the EU can then launch. The simplification package is part of the European Commission’s seventh Omnibus package of 19 November 2025, which also includes amfinishments to data protection and data apply legislation as well as the introduction of European business wallets.
Leave a Reply