A multi-stage HR-in-the-loop approach to enhance fairness perceptions of AI selection systems

In the era of rapid advancements in artificial intelligence (AI), integrating AI systems into the personnel selection processes has become increasingly prevalent. As debates escalate concerning potential biases and unfairness in AI-driven decision-making, it becomes imperative to delve into how job applicants perceive fairness during the AI-based selection process. Drawing on Organizational Justice Theory, we propose a multi-stage, multi-disciplinary framework to systematically categorize and analyze fairness perceptions throughout the selection process, spanning pre-assessment, in-assessment, and post-assessment stages. Building on this framework, we advocate for four strategic approaches to facilitate deliberate design and effective implementation of AI selection systems: promoting human–AI joint decision-making, providing transparency of AI involvement and explanations of AI decisions, developing inherently fair AI selection systems, and implementing personalized communication. We also offer new insights and provide directions for future interdisciplinary research in this burgeoning field.