Bridging fit and fear: behavioral, technological, and emotional drivers of university students’ self-reported generative artificial intelligence use in learning contexts

This cross-country study developed and tested a theoretical model explaining university students’ self-reported use of generative artificial intelligence tools in learning contexts. The proposed model integrates task-technology fit, the theory of planned behavior, and fear of missing out. We collected the data from Kuwait ( n = 229) and Türkiye ( n = 147) and analyzed them using partial least squares structural equation modeling, followed by country-based multigroup analysis and artificial neural network analysis. In the Kuwait sample, fear of missing out, task-technology fit, attitude, individual-technology fit, and subjective norms had significant positive effects on generative artificial intelligence use. By contrast, perceived behavioral control was not significant ( R ² = 0.632). In the Turkish sample, fear of missing out, individual-technology fit, and subjective norms showed significant effects, whereas attitude, perceived behavioral control, and task-technology fit did not ( R ² = 0.567). In the combined analysis, all variables were statistically significant, and the model’s explanatory power was 59.6%. Individual-technology fit emerged as the strongest predictor in neural network analysis. These findings indicate that the relative importance of the predictors differs across the Kuwaiti and Turkish populations. Because cultural values were not measured directly, these differences should be read as contextual rather than as confirmed cultural mechanisms. The study contributes to the development of strategies that will support the responsible and effective use of generative artificial intelligence tools in higher education.