{"id":888,"date":"2025-09-08T10:30:49","date_gmt":"2025-09-08T10:30:49","guid":{"rendered":"https:\/\/naaia.ai\/news\/high-risk-ai-act-list-eu-ai-regulation-compliance\/"},"modified":"2026-06-02T13:44:06","modified_gmt":"2026-06-02T13:44:06","slug":"high-risk-ai-act-list-eu-ai-regulation-compliance","status":"publish","type":"post","link":"https:\/\/naaia.ai\/en\/high-risk-ai-act-list-eu-ai-regulation-compliance\/","title":{"rendered":"High-risk AI: Understanding the AI Act list to anticipate your obligations"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">The European regulation on artificial intelligence, the&nbsp;<a href=\"https:\/\/naaia.ai\/ai-act-conformity-governance-europe\/\">AI Act<\/a>, introduces a risk-based approach to regulate the use of AI.&nbsp;<a href=\"https:\/\/naaia.ai\/ai-act-high-risk\/\">It imposes specific requirements on&nbsp;<strong>high-risk AI systems<\/strong><\/a>, meaning those likely to have a significant impact on fundamental rights, health, or safety. A public consultation is currently open to clarify the classification of the systems concerned, possible exceptions, and the obligations related to their placement on the market.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">The AI Act List: Two Paths to Classify High-Risk AI<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The AI Act distinguishes two categories of systems that can be considered high-risk:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>AI systems integrated into products already regulated by the EU<\/strong>&nbsp;(Annex I)<\/li>\n\n\n\n<li><strong>AI systems listed based on their purpose or use case<\/strong>&nbsp;(Annex III)<\/li>\n<\/ol>\n\n\n\n<p class=\"wp-block-paragraph\">In both cases,&nbsp;<a href=\"https:\/\/naaia.ai\/ai-use-cases-by-industry-risks-compliance-regulation\/\">these systems are subject to the requirements of&nbsp;<strong>Title III<\/strong>&nbsp;of the regulation<\/a>, particularly in terms of risk management, transparency, and human oversight.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">1. Regulated AI systems: products covered by other European laws<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Some products are already subject to a conformity assessment under sector-specific legislation. When an AI system is integrated into one of these products, it is&nbsp;<strong>automatically considered high-risk<\/strong>.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Targeted products include:<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Medical devices (including in vitro)<\/li>\n\n\n\n<li>Machinery<\/li>\n\n\n\n<li>Motor vehicles and trailers<\/li>\n\n\n\n<li>Railway systems<\/li>\n\n\n\n<li>Aircraft, aeronautical equipment<\/li>\n\n\n\n<li>Marine equipment<\/li>\n\n\n\n<li>Gas appliances<\/li>\n\n\n\n<li>Pressure equipment<\/li>\n\n\n\n<li>ATEX equipment<\/li>\n\n\n\n<li>Toys<\/li>\n\n\n\n<li>Low-voltage electrical equipment<\/li>\n\n\n\n<li>Weighing instruments<\/li>\n\n\n\n<li>Measuring instruments<\/li>\n\n\n\n<li>Lifts<\/li>\n\n\n\n<li>Construction products<\/li>\n\n\n\n<li>Personal protective equipment<\/li>\n\n\n\n<li>Cableway installations for passenger transport<\/li>\n\n\n\n<li>Outdoor noise-emitting equipment<\/li>\n\n\n\n<li>Intelligent Transport Systems (ITS)<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">These products are considered sensitive because their proper functioning has a direct impact on&nbsp;<strong>public safety<\/strong>&nbsp;or&nbsp;<strong>user rights<\/strong>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">2. Full list of high-risk AI use cases<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Annex III<\/strong>&nbsp;of the AI Act regulation identifies&nbsp;<strong>eight categories of use cases<\/strong>&nbsp;for which AI systems are&nbsp;<strong>automatically considered high-risk<\/strong>, due to their potential impact on people\u2019s rights, health, or safety. These areas cover sensitive functions in social, economic, or democratic life. Below is the complete list:<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Category 1: Biometrics<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Remote biometric identification in real time or post-event in publicly accessible spaces<\/li>\n\n\n\n<li>Biometric categorization based on sensitive attributes (e.g., ethnic origin, sexual orientation, etc.)<\/li>\n\n\n\n<li>Emotion recognition from images or physiological signals<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">Category 2: Critical infrastructure<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>AI systems used as safety components for managing the supply of water, electricity, gas, or heating<\/li>\n\n\n\n<li>Systems linked to road traffic management or critical digital infrastructure<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">Category 3: Education and vocational training<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Access or admission to educational institutions<\/li>\n\n\n\n<li>Learning outcome assessment, student guidance, personalization of learning paths<\/li>\n\n\n\n<li>Exam monitoring (detection of suspicious behavior)<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">Category 4: Employment, human resources, and access to self-employment<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Automated recruitment: job posting, pre-selection, candidate scoring<\/li>\n\n\n\n<li>HR decision-making: promotion, dismissal, task assignment<\/li>\n\n\n\n<li>Monitoring employee performance and behavior<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">Category 5: Access to essential services (public and private)<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Granting or withdrawing social benefits, healthcare, or public aid<\/li>\n\n\n\n<li>Credit scoring and creditworthiness assessment (excluding fraud detection)<\/li>\n\n\n\n<li>Risk assessment and pricing in life and health insurance<\/li>\n\n\n\n<li>Processing and prioritization of emergency calls<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">Category 6: Law enforcement and public security<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Risk assessment for victimization or reoffending<\/li>\n\n\n\n<li>Use of systems like polygraphs or lie detectors<\/li>\n\n\n\n<li>Evaluation of the reliability of evidence in investigations<\/li>\n\n\n\n<li>Predictive profiling (excluding fully automated profiling)<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">Category 7: Migration, asylum, and border control<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Security or immigration risk assessment<\/li>\n\n\n\n<li>Assistance in examining asylum, visa, or residence permit applications<\/li>\n\n\n\n<li>Automated person identification (excluding simple document checks)<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">Category 8: Justice and democratic processes<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Decision support for judges or alternative dispute resolution<\/li>\n\n\n\n<li>Systems aimed at directly influencing voting or electoral behavior<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">In all these cases, systems must comply with the strict obligations of&nbsp;<strong>Title III<\/strong>&nbsp;of the regulation, unless they meet the&nbsp;<strong>exemption conditions<\/strong>&nbsp;defined in&nbsp;<strong>Article 6(2).<\/strong><\/p>\n\n\n\n<h2 class=\"wp-block-heading\">3. Exemption cases: When AI Systems Are Not High-Risk<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Some systems listed in&nbsp;<strong>Annex III<\/strong>&nbsp;may be excluded from the high risks AI systems scope if they meet specific&nbsp;<strong>functional conditions<\/strong>. The goal is to avoid overregulating low-impact technical uses. To qualify for this exception, an AI system must meet one of the following cases:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Perform a narrow procedural task<\/strong>: A system that carries out a well-defined and low-stakes task, such as converting unstructured data into structured data, sorting incoming documents, or detecting duplicates.<\/li>\n\n\n\n<li><strong>Enhance the output of a pre-existing human activity<\/strong>: AI is not autonomous in its decision but complements a human action, for example, reformulating a text in a professional tone or adapting it to an editorial style guide.<\/li>\n\n\n\n<li><strong>Detect patterns or anomalies without influencing decisions<\/strong>: AI intervenes after the fact to flag possible anomalies, without modifying or directly influencing human judgment, e.g., analyzing grading inconsistencies from a teacher.<\/li>\n\n\n\n<li><strong>Prepare an assessment without producing the final result<\/strong>: Applies to AIs that support upstream processing (indexing, translation, text or speech recognition) without participating in the final decision.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">These systems must nevertheless undergo a&nbsp;<strong>rigorous assessment<\/strong>&nbsp;to verify that they have&nbsp;<strong>no decision-making autonomy<\/strong>&nbsp;and&nbsp;<strong>no direct impact on individuals\u2019 rights<\/strong>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">AI Act Compliance: What Providers of High-Risk AI Must Do<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Providers of high risk AI systems must comply with several key obligations to ensure alignment with the AI Act:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Risk management throughout the system&#8217;s lifecycle<\/strong>&nbsp;(Article 9): implementation of a continuous, planned, and iterative process, addressing known or foreseeable risks.<\/li>\n\n\n\n<li><strong>Comprehensive technical documentation<\/strong><\/li>\n\n\n\n<li><strong>Transparency toward users<\/strong><\/li>\n\n\n\n<li><strong>Appropriate human oversight<\/strong><\/li>\n\n\n\n<li><strong>A quality management system<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">The goal is to ensure that the system operates as intended, within a controlled framework, with safeguards for affected individuals.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Understanding the categories and obligations of high-risk systems is essential to anticipate regulatory compliance. The AI Act provides a clear framework but calls for&nbsp;<strong>rigorous functional analysis<\/strong>&nbsp;of each use case.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Public consultation statement :<\/strong>&nbsp;A public consultation is currently open to clarify the classification of the systems concerned, possible exemptions, and obligations related to their placement on the market.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Do you design or use a potentially high-risk AI?<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Make sure you&#8217;re compliant with the AI Act,\u00a0<a href=\"https:\/\/naaia.ai\/request-demo\/\">request a Naaia demo<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>The European regulation on artificial intelligence, the&nbsp;AI Act, introduces a risk-based approach to regulate the use of AI.&nbsp;It imposes specific requirements on&nbsp;high-risk AI systems, meaning those likely to have a&hellip; <a href=\"https:\/\/naaia.ai\/en\/high-risk-ai-act-list-eu-ai-regulation-compliance\/\">Lire la suite<\/a><\/p>\n","protected":false},"author":2,"featured_media":889,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[46],"tags":[],"class_list":["post-888","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-governance-blog"],"_links":{"self":[{"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/posts\/888","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/comments?post=888"}],"version-history":[{"count":1,"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/posts\/888\/revisions"}],"predecessor-version":[{"id":2373,"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/posts\/888\/revisions\/2373"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/media\/889"}],"wp:attachment":[{"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/media?parent=888"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/categories?post=888"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/naaia.ai\/en\/wp-json\/wp\/v2\/tags?post=888"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}