The new European regulation published in the OJEU will require many organizations to revisit their artificial intelligence (AI) compliance plans. While it introduces simplification measures and extends certain transition periods, it also modifies key aspects of the rules applicable to high-risk systems, transparency, data processing, and innovation, making it necessary to reassess the regulatory impact of ongoing AI projects.
Following the final adoption by the Council of the EU of the AI Digital Omnibus Regulation on June 29, its official version was published today, July 24, in the Official Journal of the European Union (OJEU), and it will enter into force three days later, on July 27. The publication comes at the very last minute, as August 2 was the date initially set for mandatory compliance with the Artificial Intelligence Act (Regulation (EU) 2024/1689, “AI Act”). Indeed, the compliance timeline is one of the key changes introduced by the Omnibus Regulation.
The objective of this regulation is to facilitate the practical application of the AI Act by clarifying concepts, simplifying potentially overlapping regulatory frameworks, and rescheduling certain deadlines, among other measures.
This initiative is part of a broader legislative effort by the European Union aimed at simplifying 12 regulatory areas through various omnibus packages. This instrument is part of the seventh package (digital sphere) and is complemented by the general Digital Omnibus, currently pending final approval, which will introduce amendments to other key texts in the European digital ecosystem, such as the General Data Protection Regulation or the Data Act, among others.
Key changes
Among the key changes introduced by this regulation, the following are particularly noteworthy:
- The generation of non-consensual intimate or sexual content and the creation of child sexual abuse material using AI systems are now classified as prohibited practices. These prohibitions will take effect on December 2, 2026.
- The application of obligations for high-risk systems, initially scheduled for August 2, 2026, has been deferred. Specifically, the deadline is extended to December 2, 2027, for high-risk systems as defined in Annex III of the AI Act (e.g., in the fields of education or employment) and to August 2, 2028, for systems embedded in products regulated by EU sector-specific harmonization legislation (e.g., toys or watercraft).
- For AI systems that generate synthetic audio, image, video, or text content and that were placed on the market before August 2, 2026, a transitional period is introduced until December 2, 2026, to comply with the obligation to label their outputs in a manner that enables them to be identified as artificially generated. For systems placed on the market thereafter, this obligation will apply immediately, along with the other transparency obligations under article 50, which take effect on August 2, 2026.
- To avoid duplication with sector-specific legislation, a provision is introduced to allow the Commission, through delegated acts, to limit the application of certain obligations under the AI Act to high-risk systems where sector-specific regulations ensure an equivalent or higher level of protection. In addition, the definition of “safety component” is clarified.
- The scope of the obligations regarding artificial intelligence literacy is scaled back. The obligation now focuses on adopting measures to support the development of literacy, without requiring a specific level of literacy from each individual.
- The legal basis for processing special categories of personal data for the purpose of detecting and correcting bias is expanded, extending it to operators other than providers of high-risk systems.
- Measures to promote innovation, previously limited to SMEs and startups, are now extended to mid-cap companies.
Next steps
Following this legislative overhaul, organizations that develop, market, or use AI systems should review their compliance roadmaps to account for the new deadlines, revised obligations, and opportunities arising from the simplification measures. In addition, developments relating to the remaining digital omnibus initiatives will need to be closely monitored, as their adoption is also expected to have a significant impact on the governance of digital assets.