Read the general principles document, Annex I, and Annex III here.
The European Commission has published draft guidelines on when an AI system qualifies as high-risk under Article 6 of the EU Artificial Intelligence Act (AI Act), and when it does not. The guidelines are out for stakeholder consultation before finalisation and carry no binding force. Only the Court of Justice of the EU can give authoritative interpretations of the Act.
Why it matters: The EU AI Act is the world’s first comprehensive AI law with binding obligations, and the high-risk label is its most consequential classification. Companies selling or operating AI systems in the EU need to know whether their product clears this threshold before going to market: Being classified as high-risk means building a risk management system, maintaining technical documentation, putting data governance controls in place, enabling human oversight, and completing a conformity assessment — all before a single sale. Getting it wrong can attract fines of up to €30 million or 6% of global annual turnover, whichever is higher.
Who decides if a system is high-risk: The provider decides, not a regulator upfront. Classification turns on the intended purpose of the AI system, as described by the provider in instructions, documentation, promotional materials, and terms of service. Regulators can challenge the assessment after the fact, but the provider self-certifies before going to market. Providers cannot escape classification through their terms of service alone: asserting that the system excludes high-risk uses is not enough if promotional content or examples effectively promote them. For general-purpose AI providers, a model marketed as capable of CV screening or credit scoring, without consistently limiting those uses across all materials, attracts full high-risk compliance obligations.
Which AI systems are high-risk under Article 6(2): An AI system is high-risk if its intended purpose falls within one of eight areas the Act considers to pose the greatest risk to people’s rights, safety, and health. The list of qualifying use cases within each area is closed: only the specific use cases named in Annex III count, and only a formal legal amendment can change that list. Being in one of these sectors does not automatically make an AI system high-risk. High-risk classification does not ban the use of the system either. Instead, it requires the system to work accurately and the provider to properly manage risks to health, safety, and fundamental rights. The eight areas and their qualifying use cases are:
- Biometrics: Remote biometric identification systems, biometric categorisation based on sensitive attributes, emotion recognition
- Critical infrastructure: AI managing road traffic, water supply, gas, electricity, or heating networks
- Education and vocational training: AI deciding admission to schools or universities, evaluating learning outcomes, monitoring students…
Source link
Disclaimer
We strive to uphold the highest ethical standards in all of our reporting and coverage. We blogs.grocliq.com want to be transparent with our readers about any potential conflicts of interest that may arise in our work. It’s possible that some of the investors we feature may have connections to other businesses, including competitors or companies we write about. However, we want to assure our readers that this will not have any impact on the integrity or impartiality of our reporting. We are committed to delivering accurate, unbiased news and information to our audience, and we will continue to uphold our ethics and principles in all of our work. Thank you for your trust and support.
Website Upgradation is going on for any glitch kindly connect at [email protected]
