Abstract :
[en] Artificial intelligence (AI) is profoundly transforming recruitment, raising hopes of more efficient, reliable and objective processes, but at the same time raising concerns about the reproduction or amplification of existing biases. Indeed, the European AI Act classifies AI-systems for recruitment as "high-risk systems", requiring special safeguards and attention.
This study examines the use of AI in recruitment and selection processes in Belgium. Its main objective is to analyse recruiters' current practices and assess their awareness of the risks of gender discrimination linked to the use of AI systems. The study highlights a growing but uneven adoption of AI in recruitment, mainly concentrated
on the preparatory phases ahead of the process itself. While recruiters recognise the efficiency gains, they have legitimate concerns about dehumanisation, data confidentiality and the risk of discrimination. Awareness of the risks of gender bias remains insufficient overall, particularly among internal recruiters and in smaller companies. The paradox lies in the fact that AI, perceived as potentially more objective, can in fact reproduce or amplify existing discrimination if it is not rigorously supervised, which is what some recruiters sometimes tend to forget, in favour of a strong belief in the virtues of AI. In this respect, the results of our study reveal a fundamental tension: on the one hand, the hope that AI can neutralise human bias by standardising processes; on the other, the fear that it will lock organisations into models that reproduce the past, while losing the wealth of contextual information and human creativity that may pick out atypical but relevant candidates.
The recommendations put forward here aim to create a responsible ecosystem combining
public regulation, responsibility on the part of developers and user companies, and vigilance on the part of recruitment professionals. The challenge is to make AI a lever for equality rather than a vector for reproducing gender inequalities, in compliance with the legal framework and with a view to continuously improving recruitment practices.
