UE-OSHA - Enhancing worker health and safety through Ai-driven systems: implications of the Artificial Intelligence act Discussion paper, april 2026

European Agency for Safety and Health at Work

Stefania Marassi, John Bolte

ENHANCING WORKER HEALTH AND SAFETY THROUGH AI-DRIVEN SYSTEMS: IMPLICATIONS OF THE ARTIFICIAL INTELLIGENCE ACT

DISCUSSION PAPER

CONTENTS

1. Introduction
2. Example of a prohibited AI practice: AI emotion recognition technologies in the workplace
    2.1 AI systems not classified as AI emotion recognition systems under the AI Act and therefore not prohibited
    2.2 The ‘medical or safety reasons’ exception: when AI emotion recognition in the workplace is permitted
3 High-risk AI systems in the AI Act and AI-driven OSH systems
    3.1 AI systems that are intended to be used as safety components of a product or are themselves products
    3.2 Stand-alone high-risk AI systems in the workplace
4 The interplay between the AI Act, EU OSH legislation and the GDPR
    4.1 The AI Act as complementary legislation
    4.2 Employers’ obligations under the AI Act and information (and consultation) rights
5 Key takeaways on the practical application of the AI Act to AI-driven OSH systems
6 Concluding remarks
7 References
Annex I. Text of selected relevant articles and recitals (40) of the AI Act

© European Agency for Safety and Health at Work, 2026
Reproduction is authorised provided the source is acknowledged.

fonte: osha.europa.eu