Generative AI literacy in Tanzanian higher education: comparing instructors and students practices using the knowledge-skills-attitudes framework

Abstract

Generative artificial intelligence tools are increasingly shaping teaching and learning in higher education, yet evidence from African contexts remains limited. We examined generative artificial intelligence literacy among 320 instructors and 385 students across 10 higher education institutions in Tanzania. Both groups reported substantial engagement, but the literacy profile was uneven: applied use and self-reported skills were relatively strong, whereas conceptual and ethics-related knowledge were comparatively weaker. Students reported higher experience, use, skills, and openness, while instructors reported higher knowledge; attitudes were broadly similar across groups, and most differences were small to moderate in magnitude. Structural modelling showed that experience and use were positively associated with knowledge and skills, and that openness was strongly associated with attitudes. Exploratory cross-level analyses further indicated that instructor openness and attitudes were associated with students’ attitudes and reported use. These findings highlight the need for clearer institutional guidance, curriculum-embedded digital competence, and professional development that supports responsible and context-sensitive use in Tanzanian higher education.