AI at work and workers' rights: what the EU AI Act says

When AI agents join a workplace, the people already there have a stake in how it happens. Some AI systems help workers do their jobs; others are used to hire, assign, monitor or evaluate them. The EU AI Act draws a sharp line between the two and gives workers specific rights where AI is used to make decisions about them. This page summarizes what the Act says. It is a plain-language overview, not legal advice.

The short answer: the EU AI Act treats AI used for recruitment, promotion, termination, task allocation based on personal traits or behaviour, and performance monitoring as high-risk. Employers using such systems must inform workers' representatives and affected workers beforehand. People affected by certain high-risk AI decisions can ask for an explanation. Using AI to infer workers' emotions is prohibited, with medical and safety exceptions.

Which workplace uses count as high-risk

Annex III of the Act lists employment, workers' management and access to self-employment as a high-risk area. It covers (EU AI Act, Annex III, point 4):

• AI systems intended to be used for the recruitment or selection of people, in particular to place targeted job advertisements, to analyse and filter job applications, and to evaluate candidates.
• AI systems intended to be used to make decisions affecting terms of work-related relationships, the promotion or termination of work-related contractual relationships, to allocate tasks based on individual behaviour or personal traits or characteristics, or to monitor and evaluate the performance and behaviour of persons in such relationships.

An agent that drafts reports for a team is not on this list. An agent that ranks those team members for promotion is.

Workers must be told first

Article 26(7) says that before putting into service or using a high-risk AI system at the workplace, deployers who are employers shall inform workers' representatives and the affected workers that they will be subject to the use of the high-risk AI system. This is done following the rules and procedures of EU and national law and practice on informing workers and their representatives (EU AI Act, Article 26(7)).

Separately, Article 26(11) requires deployers of high-risk systems listed in Annex III that make, or assist in making, decisions related to people to inform those people that they are subject to the use of the system (Article 26(11)).

For how these rules and several US state and city laws apply to recruitment and screening tools, see our guide to AI in hiring.

The right to an explanation

Under Article 86(1), a person affected by a decision that a deployer takes on the basis of output from a high-risk AI system listed in Annex III has the right to obtain clear and meaningful explanations of the role of the AI system in the decision-making procedure and the main elements of the decision taken. The right applies where the decision produces legal effects, or similarly significantly affects that person in a way they consider to have an adverse impact on their health, safety or fundamental rights (EU AI Act, Article 86).

The scope is specific: Annex III high-risk systems, decisions with legal or similarly significant effects, and the person's own view that the impact is adverse. Article 86 also sets out exceptions, so check the full text for a given case.

Human oversight, by competent people

Employers using high-risk systems must assign human oversight to people with the necessary competence, training and authority, and the support to do it (Article 26(2)). Those people must be able to understand the system's limits, avoid over-relying on it, interpret its output, override it and stop it (Article 14). In a workplace, this means a person who can actually overrule the system on a hiring or performance decision.

What is banned outright

Article 5 prohibits the use of AI systems to infer the emotions of a person in the workplace and in education institutions, except where the system is intended for medical or safety reasons (EU AI Act, Article 5(1)(f)).

For how AI is used to monitor and evaluate workers in practice, what research finds and what to ask, see AI monitoring at work. For its health and safety side, from stress to robots working beside people, see AI and workplace safety.

A note on dates

The Act's obligations apply in stages. The Digital Omnibus on AI, Regulation (EU) 2026/1744, published in the Official Journal on 24 July 2026 (NI Assembly EU law tracker), moved the date the Annex III high-risk rules apply, including those for employment, to 2 December 2027 (European Commission). Before relying on a specific date, check the current consolidated text and official guidance.

Beyond the legal minimum: a fair transition

The law sets a floor. How AI changes work is still largely decided inside each organization. The International Labour Organization's research on generative AI and jobs is offered as a foundation for social dialogue and targeted policy responses to manage the transition, and it frames the key question as how far jobs can retrain and retain existing workers rather than replace them (ILO Working Paper 140, 2025).

In practice, organizations that bring workers in early tend to get better systems: the people doing the work know where an agent will fail. Good practice includes:

• Explaining which AI systems are used, for what, and who oversees them.
• Involving workers or their representatives before systems that affect them go live. For how unions have bargained over AI, see AI and unions.
• Giving people a clear route to question an AI-assisted decision about them, and a safe way to report a system they think breaks the rules; see raising concerns about AI at work.
• Offering training to work with new systems, and routes into the new tasks they create. Where AI does lead to job cuts, consultation and notice rules apply; see AI and layoffs.