AI in hiring and HR under the EU AI Act
Yes, in most cases. Annex III of the AI Act lists employment and workers management as a high-risk area: AI that recruits, selects, evaluates, promotes or dismisses people is high risk, and emotion recognition at work is prohibited.
Not legal advice. Sources: Regulation (EU) 2024/1689 (the AI Act).
What Annex III lists
Point 4 of Annex III covers AI systems intended to be used:
- for recruitment or selection, in particular to place targeted job advertisements, to analyse and filter job applications, and to evaluate candidates;
- to make decisions affecting work relationships: promotion or termination, allocating tasks based on individual behaviour or personal traits, or monitoring and evaluating the performance and behaviour of workers.
These are high risk under Article 6(2), with the obligations of Chapter III.
What is prohibited at work
Article 5(1)(f) bans AI systems that infer the emotions of people at the workplace, except for medical or safety reasons. Analysing candidates' facial expressions or voice in video interviews to judge their enthusiasm or honesty falls under this ban. It has applied since 2 February 2025.
The exceptions, and their limit
Article 6(3) says an Annex III system is not high risk when it does not pose a significant risk of harm, for example when it only performs a narrow procedural task (such as converting CVs into a standard format), or only improves the result of work a person already did.
The exception never applies when the system profiles people: evaluating or predicting their work performance, reliability or behaviour is profiling. Most screening and ranking tools do this. A provider who relies on the exception must document the assessment and register the system (Article 6(4)).
What the employer (deployer) must do
Most employers use a hiring tool built by someone else: they are its deployer. Under Article 26 they must use it according to its instructions, assign trained people to oversee it, check that the input data they control is relevant, keep the logs for at least six months, and inform workers' representatives and the affected workers before using it at the workplace. People who are subject to a decision can ask for an explanation (Article 86).
What the vendor (provider) must do
The company that builds the tool is its provider: risk management, data governance and bias examination, technical documentation, logging, human oversight by design, accuracy and robustness, a quality management system, a conformity assessment, CE marking and registration (Articles 9 to 17, 43 to 49). An employer that builds its own screening tool is a provider too.
Key dates
| Date | What applies |
|---|---|
| 1 August 2024 | The AI Act enters into force. |
| 2 February 2025 | Prohibited practices are banned; AI literacy duties start (Articles 4 and 5). |
| 2 August 2025 | Obligations for general-purpose AI models apply to new models (Articles 53 to 55); penalties apply. |
| 2 August 2026 | Most of the Act applies: high-risk systems in Annex III areas, transparency duties (Article 50), deployer obligations. |
| 2 August 2027 | High-risk rules apply to AI in products covered by Annex I law (Article 6(1)); general-purpose models placed on the market before 2 August 2025 must comply. |
Dates checked against Regulation (EU) 2024/1689 (the AI Act) on 17 November 2026. In November 2025 the European Commission proposed a "Digital Omnibus" that would postpone the high-risk obligations, linking them to the availability of standards (at the latest December 2027 for Annex III areas and August 2028 for Annex I products). Check whether it has been adopted, and its final dates, before relying on the high-risk dates shown here.
Questions
Is using ChatGPT to write job ads high risk?
Writing the text of a job ad is not one of the Annex III uses. Placing targeted job ads, filtering applications or evaluating candidates is.
Is an applicant tracking system high risk?
Only the parts that use AI to filter, rank or evaluate applications. Storing applications and scheduling interviews is not an AI use listed in Annex III.
Do we need to tell candidates?
Yes when a high-risk system is used to make or assist decisions about them (Article 26(11)), and workers and their representatives must be informed before the system is used at work (Article 26(7)).
More guides
Related free tools
Published ; awaiting review.