EU AI Act: what changes for hospitals using high-risk AI
Full obligations for high-risk systems apply from August 2026, and hospitals count as deployers.
The EU AI Act (Regulation (EU) 2024/1689) sets risk-based rules for AI. A guide published by AI scribe vendor Tandem Health sets out what it means for healthcare organisations, which the Act treats as deployers when they use a third-party AI tool in their work.
Obligations began phasing in from February 2025, and the full requirements for high-risk AI systems apply from August 2026. For deployers of high-risk systems these include human oversight, appropriate use in line with the provider's instructions, staff training, monitoring, and reporting serious incidents.
Much clinical AI qualifies as high-risk because it is, or is part of, a medical device under the MDR or IVDR that needs a notified body.
