guidance

High-risk § 3.3.1

Overview of use cases and horizontal issues

Official source ↗

(209)

Point 3 of Annex III lists as high-risk four AI system use cases intended to be used in the area of education and vocational training, namely:

(a)

AI systems intended to be used to determine access or admission to or to assign natural persons to educational and vocational training institutions or programmes at all levels;

(b)

AI systems intended to be used to evaluate learning outcomes of natural persons;

(c)

AI systems intended to be used to assess the appropriate level of education thatan individual will receive or will be able to access, in the context of or within educational and vocational training institutions at all levels; and

(d)

AI systemsintended to be used to monitor and detect prohibited behaviour of students during tests in the context of or within educational and vocational training institutions at all levels. i. The concept of 'educational and vocational training institution'

(210)

An AI system will only be classified as high-risk under point 3 of Annex III if it is intended to be used in the context of or within educational and vocational training institutions at any level. The reference to ‘educational and vocational training institutions’ in conjunction with the reference to ‘at all levels’ should be understood in a broad manner so as to encompass all levels of formal and non-formal education, including early childhood, primary, secondary, tertiary and vocational education and training, as well as adult education.

(211)

As explained in the Guidelines on prohibited artificial intelligence practices31, the reference to ‘educational institutions’in the AI Act is broad, encompassing both public and private institutions, with no limitation on the types or ages of pupils or students, or the specific environment (online, in person, in a blended mode, etc.). Educational institutions are normally accredited or sanctioned by the relevant national education authorities or equivalent authorities, and may provide a certificate (with participation as a precondition for obtaining a certificate).

(212)

This includes not only formal educational and vocational training institutions, but also non-formal institutions, such as adult education and continuing educational institutions, which may offer personalised learning and skill development opportunities, an d which may provide a certificate upon completion of the training or course.

(213)

In particular, formal education refers to the structured education system that runs from primary to tertiary education and may also include specialised programmes for technical and professional training. Non-formal education refers to any deliberate, voluntary and planned programme of personal and social education that aims to convey and practice values, and develop a wide range of skills and competences, outside the formal education curriculum. Vocational training is an integral part of formal education in several EU education systems.

(214)

AI systems falling within one of the use cases listed in point 3 of Annex III are those that are intended to be used in the context of or within educational and vocational training institutions at all levels. Such systems primarily impact individuals engaged in learning, such as students or apprentices. Where an AI system is used in the context of an educational or vocational training institution, but it is mainly directed to the training of educators and training staff, including teachers and academic staff, that system will fall outside the use cases listed in point 3 of Annex III, although it may fall within one of the employment use cases listed in point 4 of Annex III.

(215)

AI systems falling within the use cases listed in point 3 of Annex III will often involve the automated processing of personal data to evaluate the personal aspects of a natural person, in particular to analyse or predict aspectsrelated to that person’s academic or vocational training path. Such systems will often perform profiling of natural persons and may lead to automated decision-making within the meaning of Article 22 GDPR. They should always be classified as high-risk, even if the system fulfils the conditions of the filter mechanism listed in Article 6(3) AI Act. ii. Interplay with the prohibited practices and the use case in point 1(c) of Annex III

(216)

Article 5(1)(f) AI Act prohibits the placing on the market, the putting into service for this specific purpose, or the use of AI systems to infer emotions of a natural person in the area of education institutions, except where the use of the system is intended to be placed on the market or put into service for medical or safety reasons. Where an AI system to infer emotions of natural persons in the area of education falls outside the scope of this prohibition, it may nevertheless be classified as high-risk, either because it falls within one of the use cases listed in point 3 of Annex III or the use case listed in point 1(c) of Annex III.

  1. Paragraph255, Commission Guidelines on prohibited artificial intelligence practices established by Regulation (EU) 2024/1689 (AI Act).