HOW ATTITUDE SHAPES GENERATIVE AI ADOPTION INTENTIONS AMONG GRADUATE STUDENTS

Despite attitude's theoretical centrality in technology acceptance, UTAUT-based models have largely omitted it as a mediating mechanism. This study reinstates attitude within this framework by examining whether it mediates the relationship between performance expectancy and behavioral intention in voluntary Generative AI adoption among graduate business students. Drawing on TAM and UTAUT, we analyzed survey data from 252 MBA students at an internationally accredited business school in Peru. Using ordinary least squares regression with HC3 robust standard errors and nonparametric bootstrap mediation analysis, we found that performance expectancy significantly influenced both attitude and behavioral intention. Attitude accounted for 60.6% of the total effect in the simple model and 46.3% after adjusting for effort expectancy, risk perception, prior training, use frequency, age, and sex. These results position attitude as a crucial evaluative mechanism that transforms cognitive beliefs into intended use, not a redundant construct. This study contributes to educational technology adoption research by restoring attitude's explanatory role in UTAUT-based models and offers practical implications for business schools integrating Gen AI into MBA learning environments.