On 19 May 2026, the European Commission published draft, non-binding guidelines on the classification of high-risk AI systems (HRAIs) under Regulation (EU) 2024/1689 (the EU AI Act). The guidelines consist of three documents: a general principles section, a guide to Annex I classification for AI as safety components in regulated products, and a guide to the Annex III listed high-risk use cases. A targeted consultation is open until 23 July 2026; the Commission expects to adopt final guidelines by end of 2026.
Article 6(1) of the EU AI Act classifies an AI system as high-risk when it serves as a safety component of a product in Annex I and the product requires third-party conformity assessment. Article 6(2) designates AI systems in Annex III use cases as high-risk. Article 6(3) allows a provider to rebut high-risk status by demonstrating the system poses no significant risk. The draft guidelines confirm that the Article 6(3) filter is unavailable where an AI component is not genuinely separable from a broader system that contributes to a high-risk output.
The Annex III use cases cover healthcare, biometric identification, employment screening, critical infrastructure management, education, law enforcement, border control, administration of justice, and democratic processes. Providers and deployers of AI systems in these categories must assess their classification against these guidelines. The guidelines treat agentic AI architectures and multi-component AI systems as requiring full-system classification analysis. Providers that previously relied on component-level analysis to argue non-high-risk status should review that position against the new interpretation of Article 6(3).
The guidelines are non-binding and will be finalised after the consultation closes 23 July 2026. The GPAI Code of Practice process addresses general-purpose AI model obligations under Chapter V of the EU AI Act separately; these guidelines do not cover those obligations. The Annex I list of regulated product categories remains unchanged.
Licentium advises AI system providers and deployers on EU AI Act compliance, including high-risk classification, conformity assessment procedures, and technical documentation requirements. We can assist with classification analysis and consultation submissions. Work we undertake includes EU AI Act compliance strategy, HRAI classification assessments, agentic AI regulatory analysis, AI governance documentation, and cross-sector AI regulatory advisory.