European Commission Publishes AI High-Risk Draft Guidance, Opens Consultation Until June 23
Updated
Updated · ITPro · May 20
European Commission Publishes AI High-Risk Draft Guidance, Opens Consultation Until June 23
6 articles · Updated · ITPro · May 20
Draft guidance released by the European Commission sets out how to decide whether an AI system is high-risk under Article 6 of the AI Act, with public feedback open until June 23.
The text splits the test in two: AI embedded in regulated products or safety components, and AI use cases that could significantly harm health, safety or fundamental rights.
Article 6(3) is likely to cause the most compliance mistakes, with a four-part carve-out narrowly limiting exemptions; Annex III systems that profile natural persons remain high-risk, a key issue for HR and credit tools.
The guidance arrives months later than expected and alongside a revised enforcement timetable: some high-risk use cases start on Dec. 2, 2027, while AI integrated into products such as robotics and industrial machinery starts on Aug. 2, 2028.
The consultation targets providers, deployers, public authorities, researchers, civil society and the public as the EU moves from broad AI Act rules toward practical implementation.
Is the EU's landmark AI Act building genuine trust, or just an expensive compliance burden that stifles innovation?
EU rules offer AI exceptions, but one red line remains: profiling people. Will this single rule derail your company's AI strategy?
EU AI Act 2026: High-Risk AI Guidelines, Delayed Deadlines, and Global Regulatory Impact
Overview
On May 19, 2026, the European Commission published draft guidelines to clarify which AI systems are considered high-risk, a key step for implementing the upcoming AI Act. These guidelines provide illustrative, non-exhaustive examples and emphasize that classification depends on a comprehensive assessment of an AI system’s intended purpose. This involves reviewing all relevant materials, such as technical documents and marketing, rather than relying on isolated statements like disclaimers. The guidelines serve as an interpretative tool for classification, not as a final legal judgment, helping organizations understand and prepare for future compliance requirements.