Please note, your browser is out of date.
For a good browsing experience we recommend using the latest version of Chrome, Firefox, Safari, Opera or Internet Explorer.

Digital Omnibus on AI: The First Amendments to the Artificial Intelligence Act Are Now in Force

SÉRVULO PUBLICATIONS 31 Jul 2026

On 27 July 2026, Regulation (EU) 2026/1744 of the European Parliament and of the Council of 8 July 2026 – the so-called Digital Omnibus on AI –, the first substantive amendment to the AI Act (Regulation (EU) 2024/1689) since its adoption in 2024, entered into force.

The Regulation postpones certain compliance deadlines, in particular for high-risk AI systems, extends regulatory simplifications to a broader range of companies, and introduces targeted amendments to the AI Act, Regulation (EU) 2018/1139 (on common rules for civil aviation) and Regulation (EU) 2023/1230 (Machinery Regulation).

 

The essentials: what changes

  • Postponement of obligations for high-risk AI systems. The deadline for the application of obligations relating to high-risk AI systems listed in Annex III (AI systems deployed in specific domains such as biometrics, critical infrastructure, employment, workers management and recruitment, credit scoring, education, law enforcement, among others) is deferred from 2 August 2026 to 2 December 2027. For high-risk AI systems embedded in regulated products under Annex I (medical devices, machinery, vehicles), the deadline is deferred to 2 August 2028.
  • Extension of simplifications to small mid-cap enterprises (SMCs). Regulatory flexibilities previously available only to small and medium-sized enterprises (SMEs) are now extended to small mid-cap enterprises, namely simplified technical documentation, proportionate requirements for quality management systems and proportionate caps on penalties.
  • Simplification of Providers’ quality management systems. The simplified compliance procedure under Article 17 of the AI Act, directed at providers of high-risk AI systems – previously reserved for microenterprises – is extended to all SMEs, including startups.
  • Broadening of regulatory sandboxes. An EU-level regulatory sandbox is established alongside national sandboxes, and the possibility of real-world testing outside sandboxes is expanded to high-risk AI systems under Annex I. The deadline for Member States to establish at least one national sandbox is deferred to 2 August 2027.
  • Transparency – marking of AI-generated content. The marking obligations directed at providers under Article 50(2) are deferred to 2 December 2026. However, the remaining marking obligations directed at deployers and the obligation to inform individuals of their interaction with AI systems under Article 50(1) apply from 2 August 2026.
  • AI literacy. The AI literacy obligation under Article 4 shifts from a requirement to ensure a specific level of literacy to a duty to take measures to promote the development of AI literacy. Companies should therefore demonstrate that they adopt measures to promote AI literacy.
  • Prohibition of AI systems that generate or manipulate sexually explicit content. A new prohibition is introduced targeting AI systems that enable the creation or manipulation of such content.
  • Amendments to civil aviation and the Machinery Regulation. Targeted amendments are introduced to Regulation (EU) 2018/1139 (on common rules for civil aviation) and Regulation (EU) 2023/1230 (machinery), ensuring sectoral coherence in the application of AI rules.

 

 

The key point: the new deadlines

Obligation

Original deadline

New deadline

High-risk AI systems — Annex III

2 August 2026

2 December 2027

High-risk AI systems — Annex I

2 August 2027

2 August 2028

Marking of AI-generated/manipulated content (Art. 50(2))

2 August 2026

2 December 2026

Establishment of national regulatory sandboxes

2 August 2026

2 August 2027

 

The Regulation entered into force on 27 July 2026, three days after its publication in the Official Journal.

 

 

 

What to do

The postponement of certain obligations should not be interpreted as an invitation to suspend compliance efforts. It is recommended to: (i) map the AI systems in use and classify them according to the AI Act risk framework, taking into account the amendments introduced by this Regulation; (ii) assess whether the organisation benefits from the new simplifications for SMEs or small mid-cap companies; (iii) promote AI literacy by adopting measures that enable staff involved in the operation and use of AI systems to develop the necessary knowledge; (iv) prepare technical documentation and quality management systems, where applicable, using the additional time to do so in a structured rather than reactive manner; and (v) monitor the implementing acts the Commission is expected to adopt by 2027, in particular regarding the sectoral scope of Annex I obligations.

 

The Regulation is available here: Regulation (EU) 2026/1744

 

 

 

Expertise Relacionadas
TMT