EU AI Act: High-Risk AI Classification Guidance Moves Into Consultation

The European Commission has opened a targeted consultation on draft guidelines for classifying high-risk AI systems under the EU AI Act, creating a new compliance milestone for healthcare, life sciences, and MedTech companies using AI.

The consultation opened on 19 May 2026 and is scheduled to close on 23 June 2026. The draft guidance is intended to clarify how organisations should determine whether an AI system is high-risk under the AI Act, including systems used as products or safety components in regulated sectors such as medical devices. (Digital Strategy)

This is particularly relevant for healthcare and life sciences companies developing or deploying diagnostic, predictive, clinical, regulatory, or decision-support AI tools. The classification question is now central: whether an AI system falls into a high-risk category will determine the level of governance, documentation, human oversight, risk management, and quality controls required.

There is also continuing movement around the AI Act implementation timeline. Recent EU-level discussions and provisional changes have indicated possible delays for certain high-risk AI obligations, including revised timelines for stand-alone and product-integrated high-risk systems. (Reuters)

For legal, compliance, and regulatory teams, the practical priority is not to wait for enforcement. Companies should begin mapping AI use cases, identifying systems that may fall into high-risk categories, and documenting governance controls now