Bridging the AI-TPACK Chasm: The Impact of Faculty AI Literacy on Pedagogical Quality and Scholarly Output in Higher Education

Generative artificial intelligence (AI) is rapidly reshaping higher education, yet evidence on how faculty AI literacy relates to concrete outcomes remains limited. Using a cross-sectional survey with embedded qualitative open-ended responses, this study examined whether faculty AI literacy predicts self-reported pedagogical practices quality and scholarly productivity. An online questionnaire was completed by 691 university faculty members and modeled relationships among four AI literacy dimensions—conceptual knowledge, application skills, ethical/critical awareness, and pedagogical integration (AI-TPACK)—and two outcomes, pedagogical practices quality and scholarly productivity (in the past three years). Hierarchical regression and ANOVA were used for the quantitative analyses, and narrative responses were thematically analyzed to contextualize and explain key patterns. Conceptual knowledge and pedagogical integration showed the strongest unique positive associations with both outcomes. Ethical/critical awareness showed a small negative association with pedagogical quality once design-oriented competencies were controlled, suggesting a potential ‘caution trap’ in which risk awareness without integration corresponds to more conservative teaching choices. Qualitative comments highlighted boundary-setting, institutional policy gaps, and unequal access to training across disciplines and career stages. The findings support targeted faculty development that integrates conceptual, ethical, and design-based AI competencies; however, because measures are self-reported and cross-sectional, conclusions should be interpreted as associations, rather than causal effects.