On 19 May 2026, the European Commission published draft guidelines(1) on how to classify Artificial Intelligence (AI) systems as “high-risk” under Article 6 of the EU AI Act, and opened a public consultation running until 23 June 2026. The high-risk AI obligations, particularly on providers, are extensive, and although the Digital Omnibus has pushed back the deadline for compliance with the Act’s obligations relating to high-risk AI systems, the question of whether a particular AI system should be categorised as high-risk is rarely straightforward. Getting the classification right is vital, but edge cases and grey areas are common.