The digital transformation paradox: Unravelling the interplay of technology acceptance, employee well-being, and organizational performance in the era of artificial intelligence

Abstract

Despite 72 percent of organizations adopting artificial intelligence, nearly half of implementations fail due to employee resistance, a paradox that challenges technology acceptance models. Prior research has examined either cognitive acceptance or motivational responses in isolation, leaving the interplay between these pathways and their ethical contingencies unresolved. This research addresses this critical gap through a sequential explanatory mixed-method design combining a quantitative survey of 1,785 employees in AI-adopting organizations with in-depth qualitative interviews with 19 participants, analysed using SmartPLS 4 for structural equation modelling and NVivo 14 for thematic analysis. The findings reveal three key insights. First, psychological empowerment mediates the relationship between technology characteristics and job satisfaction more strongly than technology acceptance, a counterintuitive finding that challenges the technology acceptance model's 30-year dominance. Second, job satisfaction emerges as the strongest predictor of organizational performance. Third, ethical leadership moderates both pathways, such that high ethical leadership amplifies empowerment effects by 34 percent. This research advances a novel Ethical Dual-Pathway Technology Acceptance Model (EDP-TAM), offering actionable guidance for organizations to implement AI that simultaneously optimises performance and safeguards employee well-being.