{"id":5408,"date":"2026-09-10T08:00:00","date_gmt":"2026-09-10T06:00:00","guid":{"rendered":"https:\/\/www.lablogic.pl\/?p=5408"},"modified":"2026-10-20T18:12:21","modified_gmt":"2026-10-20T16:12:21","slug":"high-risk-ai-system","status":"publish","type":"post","link":"https:\/\/www.lablogic.pl\/en\/high-risk-ai-system\/","title":{"rendered":"Is the AI tool we use a high-risk system?"},"content":{"rendered":"\n

Micha\u0142 Rutkowski<\/strong><\/a> \u2022 for compliance teams and DPOs \u2022 published September 10, 2026 \u2022 updated October 20, 2026 \u2022 11 min read<\/p>\n\n\n\n

\n
\n
\n\n\n\n<\/div>\n<\/div>\n\n\n\n
\n
\n

Short answer<\/h2>\n\n\n\n

Whether a given tool is a high-risk AI system is not settled by its name or by how advanced its model is. What settles it is the intended purpose given to the system by the provider, and it is the provider that carries out the classification. An organisation that modifies that intended purpose so that a high-risk AI system results becomes the provider itself. Two independent routes lead to high risk. The first is a safety component of a product covered by Annex I or such a product itself, where the product is subject to a third-party conformity assessment. The second is a use case listed in Annex III. The second route has an exit under the conditions of Article 6(3), but profiling of natural persons closes that exit without exception. A deployer of a high-risk AI system performs the obligations under Article 26 from 2 December 2027 for Annex III and from 2 August 2028 for Annex I. For systems placed on the market or put into service before those dates, Article 111(2) is decisive.<\/p>\n<\/div>\n\n\n\n

Why this question arises at all<\/h2>\n\n\n\n

Regulation (EU) 2026\/1744 postponed the application of the requirements for high-risk AI systems. Sections 1\u20133 of Chapter III apply from 2 December 2027 to systems listed in Annex III and from 2 August 2028 to systems covered by Annex I. The postponement also covers the classification rules themselves, because Article 6 belongs to the first section of that chapter. For operators of high-risk AI systems placed on the market or put into service before those dates, including deployers, the Regulation applies only after a significant change in the design of those systems (Article 111(2)). The prohibitions in Article 5 apply to those systems without that limitation. In any case, for high-risk AI systems intended to be used by public authorities, providers and deployers must take the necessary steps to comply with the Regulation by 2 August 2030. It is easy to read these changes as a reason to put the subject off.<\/p>\n\n\n\n

What can be postponed is implementing the requirements, not establishing the scope. The requirements of Chapter III for which the provider is responsible range from data quality to conformity assessment. The deployer has a shorter list of obligations under Article 26, but those too require preparation. Classification therefore determines the budget and the order of work, not a deadline that is about to expire.<\/p>\n\n\n\n

One provision of Article 6 was not postponed. Paragraph 5 obliges the European Commission to issue guidelines on the practical implementation of that Article, and the deadline passed on 2 February 2026. The Commission published the draft guidelines on 19 May 2026 and closed the consultation on 23 July 2026. In October 2026 the document is still a draft, and the Commission has announced adoption of the final version by the end of 2026. An organisation that classifies its use cases today relies on the wording of the Regulation.<\/p>\n\n\n\n

What has to be established before a decision<\/h2>\n\n\n\n

High-risk AI system \u2014 two routes to classification<\/h3>\n\n\n\n

The first route concerns products. A system becomes a high-risk AI system by this route when two conditions are met at the same time (Article 6(1)). The system must be intended to be used as a safety component of a product covered by the Union harmonisation legislation listed in Annex I, or be such a product itself. The second condition is that this product is required to undergo a third-party conformity assessment under that legislation. Annex I covers toys, lifts and medical devices among other products, so the first route mainly concerns manufacturers. Machinery has been moved to Section B of Annex I, and for AI systems classified as high-risk under Article 6(1) that relate to products covered by that section the Regulation applies only to a narrow extent (Article 2(2)).<\/p>\n\n\n\n

The second route concerns use cases. A system listed in Annex III is, as a rule, a high-risk AI system (Article 6(2)). This route does not require any link to a product covered by Annex I. In my assessment this is the route through which most organisations outside manufacturing encounter classification, because Annex III describes mainly use cases rather than industries.<\/p>\n\n\n\n

The two routes are checked separately. A manufacturer that builds an AI module into its own products and at the same time uses a tool to assess job candidates faces two independent classification questions.<\/p>\n\n\n\n

When a system is not a safety component<\/h3>\n\n\n\n

Regulation 2026\/1744 added three paragraphs to Article 6 that clarify the first route. AI systems used solely for non-safety related aspects listed in the provision do not qualify as safety components (paragraph 1a). The provision lists user assistance and performance optimisation. It also lists service efficiency, automation or convenience and quality control. There is one exception to this exclusion. Where failure or malfunctioning of the system would endanger health and safety, the system remains a safety component (paragraph 1b).<\/p>\n\n\n\n

The third new paragraph concerns the conformity assessment procedure. A product that is required to undergo a third-party conformity assessment solely due to risks other than risks to health and safety is not considered to fulfil the condition in Article 6(1)(b) (paragraph 1c). The provision gives the example of risks relating to the distribution of radio spectrum or electromagnetic interference that do not affect health and safety.<\/p>\n\n\n\n

What Annex III describes<\/h3>\n\n\n\n

Annex III lists eight areas. Two of them concern the activities of public authorities or of those acting on their behalf. These are law enforcement (area 6) and migration, asylum and border control management (area 7). Area 8 covers AI systems that assist a judicial authority in researching and interpreting facts and the law and in applying the law to a concrete set of facts. It also covers similar use in alternative dispute resolution. It further includes AI systems intended to be used for influencing the outcome of an election or referendum or the voting behaviour of natural persons in the exercise of their vote in elections or referenda. This does not include AI systems to the output of which natural persons are not directly exposed. Examples are tools used to organise, optimise or structure political campaigns from an administrative or logistical point of view. Apart from support for judicial authorities, area 8 may also concern an ordinary organisation, as may the remaining five areas.<\/p>\n\n\n\n

In my assessment the area that most often affects organisations is area 4, which covers employment, workers\u2019 management and access to self-employment. It includes tools for the recruitment or selection of candidates, among them tools for placing targeted job advertisements and filtering applications. It also includes systems intended to be used to make decisions affecting the terms of work-related relationships and the promotion or termination of work-related contractual relationships. Finally, the area covers allocating tasks based on individual behaviour or personal traits or characteristics and monitoring and evaluating the performance and behaviour of workers.<\/p>\n\n\n\n

Area 5 covers access to essential services, including the creditworthiness assessment of natural persons and risk assessment and pricing in relation to natural persons in life and health insurance. Area 1 concerns biometrics, area 2 safety components of critical infrastructure and area 3 education and vocational training. The creditworthiness category expressly excludes systems used for detecting financial fraud.<\/p>\n\n\n\n

The Annex describes the intended purpose of a system, not a category of tool. The same tool based on a language model stays outside the Annex when it suggests the content of a letter and enters area 4 when it filters candidates’ applications. If the provider did not foresee such a purpose and the system becomes a high-risk AI system because of it, the organisation giving it that purpose becomes its provider (Article 25(1)(c)).<\/p>\n\n\n\n

\n\n

Remember<\/p>\n\n\n

It is the use case that is classified, not the purchased licence.<\/p>\n\n<\/div>\n\n\n\n

When the Article 6(3) derogation really applies<\/h3>\n\n\n\n

The derogation requires two conditions to be met together. The first is general. The system must not pose a significant risk of harm to the health, safety or fundamental rights of natural persons. The provision adds that this includes not materially influencing the outcome of decision making. The second condition is one of the four situations listed in the provision.<\/p>\n\n\n\n