Resisting the machine: modeling emotional friction and academic decline in enforced AI learning environments

Abstract Background The rapid integration of artificial intelligence (AI) in higher education has shifted from voluntary engagement to mandatory usage, introducing psychological and emotional strain. Traditional technology adoption models often fail to address the challenges posed by compulsory digital environments, where system design flaws can lead to cognitive overload and student resistance. Objectives This study aims to develop and validate a student-centered framework that examines the influence of AI system design—specifically system reliability, task routinization, and information richness—on students’ negative technological experiences, and how these experiences subsequently affect techno-resistance and academic performance. Methods A quantitative research design was employed, using data collected from 229 undergraduate students across Malaysian universities. Exploratory factor analysis, correlation analysis, and multiple regression were used to examine relationships among key constructs and to validate the hypothesized model. Results Findings reveal that system reliability and information richness significantly reduce students’ negative experiences, while task routinization does not. Negative technological experience emerged as a key mediator, amplifying techno-resistance and decreasing academic performance. The results emphasize that the mandatory implementation of AI technologies does not automatically translate into improved educational outcomes. Conclusions This study reconceptualizes techno-resistance as an adaptive response to cognitive and emotional strain resulting from the poor design of AI systems in compulsory educational settings. It advances theoretical understanding by integrating digital dissonance with sociotechnical and cognitive load perspectives, and offers practical implications for the design of more student-aligned AI educational technologies.