On the Measurement of AI Literacy Among Students in Higher Education: A Scoping Review

Artificial intelligence (AI) literacy is essential in higher education. This scoping review maps how AI literacy is measured among higher education students. Using Arksey and O’Malley’s (2005) five-stage framework, we searched five databases, screened 190 records, and included 39 studies. We synthesized definitions, instruments, construct dimensions, theoretical frameworks, related variables, and geographic contexts. Findings indicate emerging consensus that AI literacy is multidimensional, commonly covering knowledge, application, evaluation, and ethics. Research was largely quantitative and self-report; the Artificial Intelligence Literacy Scale (AILS) (Wang et al., 2022) was the most frequently used measure. Few studies used objective tests or mixed methods, limiting evidence of demonstrated competence. AI literacy was often examined alongside attitudes, self-efficacy, and adoption intentions. Studies were concentrated in East Asia, with fewer from Europe, North America, and the Middle East. This review highlights needs for performance-based assessment, clearer boundaries between AI and generative AI literacy, and broader geographic coverage.