The benefits and risks of AI in recruitment

The increasing use of artificial intelligence (AI) tools in the recruitment process offers significant benefits in terms of efficiency and accuracy. Using AI can enable rapid and thorough analysis of applications, identifying relevant skills, assessing the match between the candidate and the open positions, and reducing the unconscious biases of human actors.

However, this practice raises major concerns: first, regarding the transparency of the choices made, and secondly, in terms of processing of personal data relating to the candidates.

The use of AI in recruitment raises the risk of discriminatory criteria being systematised if the AI design and training tools are not properly configured. In this way, algorithmic discrimination could target minority groups of people, whether on the basis of ‘race’, religion or gender.

The use of AI and its compliance with the GDPR

Under the General Data Protection Regulation (GDPR), employers are required to ensure that any collection and use of personal data is justified, proportionate to the purpose and complies with the principles of privacy and data security. These principles also apply to the recruitment process.

Compliance with the use of AI tools in recruitment procedures requires particular attention to several principles of the General Data Protection Regulation (GDPR), and in particular:

• The principle of transparency. Employers must provide clear information on how their data will be used by the recruiters.
• The prohibition of decisions based solely on automated processing, if this could significantly impact the rights of data subjects. AI may be used as a tool to facilitate a recruiter’s work, but the recruiter must retain the decisive role in the recruitment process.
• The consent as lawfulness basis for the processing of personal data is not possible. Indeed, the consent of candidates is not valid due to the weak position of the candidate in relation to the recruiter. The legitimate interest basis seems the most relevant, which requires a balancing exercise demonstrating that the recruitment procedure based partly on AI does not infringe the rights and interests of the data subject.