The timeline has changed
Following the AI Omnibus entering into force on 27 July 2026, the European Commission lists 2 December 2027 for Annex III high-risk rules and 2 August 2028 for high-risk AI embedded in Annex I regulated products. A blanket statement that all high-risk obligations have applied since August 2026 would be outdated.
Transparency duties under Article 50 have applied since 2 August 2026, among other provisions. For certain systems placed on the market earlier, the Commission lists an Article 50(2) transition until 2 December 2026. The specific role and transitional circumstances need individual assessment.
Assess the use case
High-risk classification depends on intended purpose and statutory criteria. General writing assistance differs from an application used to select job candidates. Definitions and exceptions also matter within sensitive areas.
Create an inventory recording purpose, affected people, data types, provider and owner. Document where people make decisions and when results are used only as drafts.
What software can contribute
Controlled access, model policies and traceable workflows support business processes in AI Link. Legal assessment of the use case and the deploying organisation’s responsibility remain separate requirements.
This overview provides general information, not individual legal advice. Binding decisions require the applicable law and specific use to be considered; the sources below support that review.
Sources & further reading
Original research, provider documentation and further reading on the concepts and figures discussed here.