Explaining reported generative AI engagement in higher education: an extended TAM with ethical compatibility and...

The rapid integration of generative artificial intelligence (AI) tools into higher education has intensified conversations regarding usefulness, ethical alignment, and responsible engagement. Unlike traditional technology acceptance studies that focus on initial use, this study examines AI use intensity among active university users. Building on an extended Technology Acceptance Model (TAM), the model incorporates AI-Alignment Construct, reliance-based trust in AI outputs, and normative alignment within academic contexts. Data were collected from 637 university students and analyzed using variance-based structural equation modeling. The results indicate that perceived usefulness remains the strongest predictor of AI use. Furthermore, reliance-based trust and AI-Alignment Construct demonstrate statistically significant correlations with engagement, whereas moderation hypotheses were not supported. These findings suggest that ethical and trust-related mechanisms operate primarily at the attitudinal alignment level rather than as boundary conditions within this cross-sectional framework. Moreover, the study contributes by repositioning TAM within a post-adoption engagement context and clarifying the bounded conceptualization of ethics and trust in AI-mediated learning environments. Practical implications emphasize calibrated AI integration, transparent governance, and assessment design aligned with academic integrity. Finally, the findings are associative in nature and should be interpreted within the methodological constraints of self-reported, cross-sectional data.