EU Commission Publishes Draft High-Risk AI Classification Guidelines
View original source →The European Commission published draft guidelines on May 19 clarifying how AI systems are classified as high-risk under the EU AI Act, launching a public consultation open until June 23.
The publication resolves a three-plus month compliance gap that had frozen significant EU AI investment.
Key points:
• Guidelines issued under Article 6(5) clarify the two pathways to high-risk classification: AI systems as safety components in products subject to EU harmonization legislation, and AI systems within Annex III use-case categories
• Annex III categories include biometrics, critical infrastructure, education, employment, healthcare, and law enforcement
• The Commission provides practical examples of systems that should and should not be classified as high-risk
• Public consultation runs until June 23, 2026, after which final guidelines will be issued
The delay from February to May reflects genuine complexity in defining 'high-risk' across the broad range of AI applications in scope. Organizations that waited for final guidance rather than over-engineering pre-emptive compliance likely made the right call.
Organizations in biometrics, critical infrastructure, education, employment, healthcare, and law enforcement sectors should participate in the consultation. The guidelines will determine compliance obligations for years.
Why It Matters: The examples provided are the most practical compliance tool published since the AI Act was finalized. Organizations can now proceed with deployment planning — though final guidelines may adjust based on consultation responses.