A high-risk AI system under Art. 6 of the EU AI Act is either a safety component of a product covered by harmonisation legislation listed in Annex I, or a system in one of the areas listed in Annex III, such as employment and workers management, education, creditworthiness or critical infrastructure. An Annex III system is exceptionally not high-risk if it does not pose a significant risk; the provider documents this assessment.
For high-risk systems, the provider is responsible for risk management, technical documentation, the quality management system and conformity assessment, among other things, and the deployer for human oversight and monitoring of operation, among other things. Following the Digital Omnibus, the requirements apply from 2 December 2027 (Annex III) and from 2 August 2028 (Annex I).