Source: The Digital Omnibus on AI enters into force (Lewis Silkin)

Law-firm briefing on Regulation (EU) 2026/1744, which amends the AI Act: high-risk deadlines moved to December 2027 and August 2028, softer AI literacy duty, marking grace period, new prohibitions.

Contents

Summary

Summarises the AI part of the EU’s Digital Omnibus package. The Commission proposed it in late 2025, Parliament and Council reached a provisional deal in May 2026, and it was published as Regulation (EU) 2026/1744.

Key claims

  • Published in the Official Journal 2026-07-24; in force 2026-07-27.
  • High-risk application dates: stand-alone Annex III systems move from 2026-08-02 to 2027-12-02; Annex I product-embedded systems to 2028-08-02.
  • Article 4 AI literacy is softened: organizations must take measures to support staff literacy but no longer guarantee a sufficient level for each person. Article 26(2) oversight training for high-risk deployers remains.
  • Article 50 transparency applies from 2026-08-02. Generative systems already on the market have until 2026-12-02 for machine-readable marking of outputs.
  • New prohibitions on “nudifier” systems and AI-generated CSAM.
  • Systems that providers self-assess as not high-risk under Art. 6(3) must still be registered, with simplified information.
  • The legal basis for processing special-category data to detect bias now covers all AI systems, under a strict-necessity test.
  • National regulatory sandboxes are due by 2027-08-02. SME relief is extended to small mid-caps.

Feeds

AI governance and compliance

Suggest changes →