<p>Generative Artificial Intelligence (GenAI) is reshaping higher education, yet empirical evidence regarding adoption heterogeneity across distinct institutional types remains limited. To address this gap, a study within an extended Unified Theory of Acceptance and Use of Technology (UTAUT) framework examined adoption behaviors among 1,062 faculty members (739 from universities and 323 from technical and vocational colleges) using a hybrid approach of PLS-SEM and fsQCA. SEM results reveal a shared dominance of intrinsic motivation, where perceived enjoyment serves as the strongest predictor for both groups, surpassing traditional utilitarian drivers in professional settings. However, divergent adoption patterns emerged: the university teachers follow a path driven primarily by performance and enjoyment, whereas the college teachers exhibit a pattern significantly influenced by social norms and effort expectancy. Complementing these findings, fsQCA uncovered causal asymmetries, identifying a conservative convergent model characterized by synergistic high-threshold logic for the university faculty, versus a flexible divergent model featuring adaptive equifinality for the college staff. Theoretically, these findings highlight the necessity of accounting for institutional typology as a boundary condition for UTAUT and recontextualize the role of psychological gratification in human-AI collaboration. Practically, the results guide educational administrators to transcend uniform digitalization mandates, advocating tailored strategies that harmonize rational utility with emotional engagement: fostering autonomy-driven exploration for university faculty while constructing socially supportive ecosystems for technical and vocational college teachers.</p>