On May 19, 2026, the European Commission published draft guidelines on the classification of high-risk AI systems under Article 6 of Regulation (EU) 2024/1689, the EU AI Act. The guidelines are in the consultation stage, open for public comment until June 23, 2026. Once finalised, they will guide AI providers, deployers, and national market surveillance authorities across the European Economic Area in determining whether a given AI system meets the statutory high-risk threshold.
Article 6 of the EU AI Act provides two pathways to high-risk classification. Under Article 6(1), an AI system is high-risk if it operates as a safety component of a product covered by Annex I harmonisation legislation and the product requires third-party conformity assessment. Under Article 6(2), an AI system is high-risk if it falls within a use case listed in Annex III, subject to the filtering mechanism in Article 6(3) that can exclude systems posing no significant risk. The Commission has postponed the Article 6(2) application date from 2 August 2026 to 2 December 2027 and the Article 6(1) date from 2 August 2027 to 2 August 2028. The draft guidelines interpret key definitional terms and provide worked examples across eight Annex III sectors.
AI providers placing systems on the EU market, deployers using Annex III applications in employment, credit scoring, education, law enforcement, border control, or judicial proceedings, and national competent authorities acting as market surveillance bodies must apply the Article 6 classification logic before the postponed compliance dates. Providers of AI tools used in recruitment screening, employee performance evaluation, or consumer credit access will find the most directly applicable guidance in the draft guidelines' worked examples for those sectors.
The Article 6(3) filtering mechanism for Annex III systems posing no significant risk remains the subject of interpretive uncertainty. The draft guidelines set out general principles for applying the filter but do not address many commercially common product configurations in detail. Industry participants have raised concerns that the worked examples leave substantial grey areas unresolved. The Commission's final position may shift following consultation submissions before June 23, 2026.
Licentium advises clients on EU AI Act compliance strategy and Article 6 classification analysis and can draw on a partner network with regulatory counsel across EU member states. AI developers and operators determining whether specific systems qualify as high-risk before the postponed application dates may benefit from a classification assessment against the draft guidelines. Work we undertake includes EU AI Act high-risk classification reviews, compliance gap analysis, Annex III sector-specific risk assessments, and EU regulatory advisory for AI product launches.