About: Philippe Ehrenström, La décision individuelle automatisée discriminatoire dans le recrutement assisté par l’intelligence artificielle, in: Florence Guillaume/Jonathan Bory (eds.), Droit et intelligence artificielle, Bern 2025, pp. 171-184 (https://staempflirecht.ch/fr/droit-et-intelligence-artificielle/ean-9783727225093):

The article analyses the risks of discrimination associated with the use of artificial intelligence systems in recruitment and examines the instruments that Swiss law, supplemented by European AI and data protection law, offers to candidates.

The author starts from the observation that many employers already automate the sorting of CVs or the evaluation of applications using models that learn from historical data and can therefore reproduce or even amplify existing discrimination, for example based on gender, origin, age or disability.

He first recalls the Swiss framework for combating discrimination in recruitment, which is characterised by the absence of a general ban in the private sector. Protection derives mainly from respect for personality rights and various special standards.

The article then recalls that algorithmic recruitment has been classified as a ‘high-risk AI system’, particularly when the tool is used to decide on access to employment. This results in a significant risk to the candidate’s personal rights.

The decision to reject a candidate solely on the basis of a score generated by an AI system constitutes an automated individual decision. The data subject has, in principle, the right not to be subject exclusively to such a decision, to request the intervention of a natural person and to obtain information on the logic underlying the processing, its significance and its intended effects.

These rights open up the possibility of using the information obtained as evidence in a lawsuit to challenge a discriminatory refusal to hire. The article discusses the procedural avenues and the particular difficulties they may present in Swiss law.

In conclusion, the author considers that Swiss data protection law offers an important lever against discriminatory automated recruitment decisions, but that practical obstacles remain and that broader reflection on a right to non-discrimination adapted to AI systems remains necessary in Swiss labour law.

Me Philippe Ehrenström, attorney, LLM, CAS in Data Protection, CAS in Law and Artificial Intelligence