Acceptance of AI-based human resource practices: Sectoral differences between technology-intensive and traditional industries

Abstract

This study examines employee acceptance of AI-based human resource management practices in technology-intensive and traditional sectors. Data were collected from 360 employees, with 60 participants from each of six sectors. Biotechnology, information technology, and manufacture of computer, electronic and optical products formed the technology-intensive group, while retail, textiles, and hotels formed the traditional group. Acceptance was evaluated across recruitment, performance appraisal, training planning, and talent management. The explanatory variables were perceived usefulness, perceived ease of use, trust in AI, perceived algorithmic fairness and transparency, and loss of human touch. The subgroup regressions show that usefulness, ease of use, and trust positively predict acceptance in technology-intensive sectors. In traditional sectors, fairness and transparency positively predict acceptance, while loss of human touch negatively predicts acceptance. Five significant interactions identify specific differences between sector groups. The positive association of fairness and transparency with acceptance is stronger in traditional sectors for recruitment and performance appraisal, and the negative association of loss of human touch is stronger in traditional sectors for recruitment. The positive association of trust is stronger in technology-intensive sectors for training planning, while that of usefulness is stronger in technology-intensive sectors for talent management. Acceptance is descriptively higher in training planning and talent management than in recruitment and performance appraisal. These findings show that employee evaluations of AI-based HR practices vary by sector group and HR function, supporting attention to usefulness, trust, fairness, and human involvement in the particular functions where their associations with acceptance differ.