Balancing empowerment and threat: deciphering the sustainable integration of Gen AI in creative pedagogy

Personal and Ubiquitous Computing · Springer Nature · 2026-10-24 · DOI: 10.1007/s10639-026-14175-x

Аннотация

Generative artificial intelligence (GenAI) is increasingly used in design education, yet students’ post-adoption evaluations involve both perceived benefits and risks. This study examined associations of threat and coping appraisals, knowledge management factors, and perceived AI attributes with satisfaction and continuance intention. Participants were 478 design students at higher education institutions in three Chinese cities. Cross-sectional survey data were analyzed using covariance-based structural equation modeling (CB-SEM), artificial neural networks (ANN), and necessary condition analysis (NCA). Perceived threat was negatively associated with satisfaction and had the largest absolute standardized CB-SEM coefficient among the eight antecedents. The other seven antecedents were positively associated with satisfaction in the structural model. Perceived intelligence ranked highest in ANN importance, followed by anthropomorphism and animacy. Satisfaction was positively associated with continuance intention in the structural model. None of the eight satisfaction antecedents met both prespecified NCA criteria. Satisfaction met both criteria for continuance intention, indicating an empirical necessity pattern within the observed sample. These findings connect risk appraisal, knowledge-related experiences, and perceived interaction quality to post-adoption evaluations in design education. The combined analyses distinguish linear associations, predictor importance, and empirical necessity as complementary perspectives on post-adoption GenAI use.

Читать у издателя →

Похожие статьи