Draft Commission guidelines on the classification of high-risk AI systems

Summary

Draft Commission guidelines setting out the Commission's interpretation of concepts relevant to classifying AI systems as high-risk under Article 6 of the AI Act, with practical examples of systems that should or should not be classified as high-risk, as Article 6(5) requires. They aim to help providers, deployers and market surveillance authorities apply Article 6 uniformly. One section covers AI systems that are safety components of products, or products, under the Union harmonisation legislation in Annex I that require third-party conformity assessment; another covers the use cases in Annex III. The draft was published for targeted stakeholder consultation.

Safety topics

Status notes

Published on 19 May 2026 in three parts, on general principles, on Annex I and on Annex III, so that each can be consulted and commented on separately; the content is also presented on the AI Act Single Information Platform. The examples are not exhaustive and may be updated over time. The targeted consultation was first open until 23 June 2026 and was extended by four weeks to 23 July 2026 at the request of several associations and other stakeholders; the Commission says the final guidelines will be adopted by the end of 2026. Following the political agreement on the AI Omnibus, the high-risk rules apply from December 2027 for stand-alone AI systems and from August 2028 for AI embedded in products. The guidelines will be complemented by other Commission guidelines on the obligations for high-risk AI systems. The Commission's report of 20 May 2026 under Article 112(1) of the AI Act noted that the classification guidelines were still in preparation. The Commission's page for the draft shows a last update of 23 July 2026.

Sources

Open the official source https://digital-strategy.ec.europa.eu/en/library/draft-commission-guidelines-classification-high-risk-ai-systems

Secondary source: https://digital-strategy.ec.europa.eu/en/news/consultation-guidelines-high-risk

Local copy

Verification

History

1. 2026-10-18related: [{"slug": "eu-ai-act", "id": "EU-03"}, {"slug": "eu-digital-omnibus-ai", "id": " → [{"slug": "eu-ai-act", "id": "EU-03"}, {"slug": "eu-digital-omnibus-ai", "id": " — two-way link added: EU-15 link(s) to EU-142026-10-18-fix-related-links · Claude Code (claude-opus-5-5)
2. 2026-10-17notes: The AI Act required these guidelines by 2 Feb 2026, but they were delayed. A dra → Published on 19 May 2026 in three parts, on general principles, on Annex I and o — rewritten from the stored files; the consultation extension, the planned adoption by end 2026 and the postponed high-risk dates added from the Commission's consultation page; the 2 Feb 2026 deadline removed2026-10-17-recheck-europe-1 · Claude Code (claude-opus-5-5)
3. 2026-10-17safety_topics: ["risk-based classification", "critical infrastructure", "public-sector use"] → ["risk-based classification", "evaluation & testing"] — 'critical infrastructure' and 'public-sector use' dropped: no stored text of this record names Annex III areas; 'evaluation & testing' added for the third-party conformity assessment that the Annex I route turns on2026-10-17-recheck-europe-1 · Claude Code (claude-opus-5-5)
4. 2026-10-17summary: Draft guidelines under AI Act Art. 6(5) on classifying AI systems as high-risk, → Draft Commission guidelines setting out the Commission's interpretation of conce — rewritten from the stored library page; the Article 6(3) filter is not shown in it and was removed2026-10-17-recheck-europe-1 · Claude Code (claude-opus-5-5)
5. 2026-10-17secondary_url: https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A52026DC0234 → https://digital-strategy.ec.europa.eu/en/news/consultation-guidelines-high-risk — COM(2026) 234 is not this instrument's page (it is the Article 112 report, EU-15); the Commission's consultation page on these draft guidelines, stored in this run, is proposed instead2026-10-17-recheck-europe-1 · Claude Code (claude-opus-5-5)

Related records

EU AI Act European Union · Partially in force

Digital Omnibus on AI (AI Omnibus) European Union · In force

AI Act Art. 112 review report 2026 European Union · Published (non-binding, current)

Citation

European Commission (AI Office). Draft Commission guidelines on the classification of high-risk AI systems. 2026-05-19. https://digital-strategy.ec.europa.eu/en/library/draft-commission-guidelines-classification-high-risk-ai-systems. Status: Draft / proposed; verified 2026-10-17 (level A). Local copy archive/EU-14/2026-10-17/official.html retrieved 2026-10-17 18:18:37 UTC, SHA-256 6a129ab6f8aab3dc6645052b88cb070a0ac350c966dd12bf74f402039a3f11a4. Every AI Policy Ever, version 1.18 (snapshot 2026-10-18), record EU-14.

Improve this record

Know of an amendment, a status change or an error? Report news or an error on GitHub; the form opens with this record filled in.