Applicable to: AI providers and deployers.
Date of application: 2 February 2025.
Objective: to prohibit the use of AI systems to infer emotions of a natural person in the areas of workplace and education institutions.
Regulation (EU) No 1689/2024 on Artificial Intelligence (AI Act) of 13 June 2024 introduced a series of prohibitions within the European Union on practices involving the use of AI systems that pose an ‘unacceptable’ risk (the highest level assigned to certain AI systems in the risk-based approach adopted by the AI Act).
In this article, we will analyse the practice covered by Article 5.1.f) of the AI Act, under which the following is prohibited:
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 areas of workplace and education institutions, except where the use of the AI system is intended to be put in place or into the market for medical or safety reasons.
The elements that characterise the prohibited practice are:
1) the fact that the AI system is placed on the market, put into service or used within the European Union
2) to infer emotions of a natural person
3) in the area of the workplace or education and training institutions
4) excluded from the prohibition are AI systems intended for medical or safety
On this point, AI Act has clarified that this prohibition applies to AI systems for the purpose of identifying and inferring emotions or intentions of natural persons based on their biometric data (physical or behavioural).
The concept of “emotions” should be understood in a wide sense, including (these examples are not exhaustive) happiness, sadness, anger, embarrassment and amusement. However, physical states such pain or fatigue should not be included in this notion (recital 18)
As specified in the Guidelines approved by the European Commission on 29 July 2025, the use of a webcam and a voice recognition system by a call centre to infer their employees’ emotions to the purpose of deciding in order to decide which tasks to assign to each of them is subject to the prohibition in question.
On the contrary, the prohibition does not apply to ad AI system that infers a driver’s fatigue to suggest when to take breaks to avoid accidents.
The Guidelines have, also, clarified that the notions of “workplace” and “education institutions” should be interpreted broadly, including in the first scope every physical or virtual space where natural persons engage in tasks assigned by their employer, independently from the status as an employee, contractor professional or trainee (the prohibition is, in fact, also applicable during the selection and hiring process.
Regarding education institutions, the notion should include both public and private institutions.
AI Act contains an explicit exception to this prohibition for emotion recognition systems used for medical or safety reasons, such as systems for therapeutical use.
On this basis, the use of an AI system should, for example, be permitted to assist employees and students with autism and improve accessibility for blind or deaf people.
Articoli Correlati