<![CDATA[This study explores the integration of artificial intelligence (AI), specifically large language models (LLMs) like ChatGPT, in evaluating cultural literacy within higher education (HE) curricula. Recognizing the increasing importance of cultural literacy in preparing globally competent graduates, the research investigates the extent to which university course outlines incorporate these competencies across disciplines. The primary objectives are to develop an effective framework for assessing cultural literacy levels in course content and to evaluate the utility of ChatGPT as an analytical tool in this process. Using a mixed-methods approach, the study first employed ChatGPT to generate a four-level classification framework for cultural literacy, which distinguishes between courses with no evidence, potential for embedding, implicit presence, and explicit integration of cultural literacy. Subsequently, a dataset of 584 publicly available course outlines from a university’s website was analyzed. ChatGPT applied the framework to categorize each course, with human coders independently validating a subset of the assessments to ensure reliability. Inter-rater reliability metrics confirmed high agreement between AI-generated and human evaluations, underscoring the robustness of the methodology. The findings reveal that many courses, especially in disciplines with high international student populations such as Technology, Engineering, and Health, lack explicit cultural literacy content. However, the framework demonstrated effectiveness in systematically assessing these competencies and offered actionable insights for curriculum development. The study concludes that AI-powered tools like ChatGPT present a promising, efficient means for continuous curriculum auditing and enhancement to foster inclusive and globally aware graduates. It also highlights the potential for broader application across institutions and disciplines, emphasizing the importance of transparent validation processes to ensure the credibility and reproducibility of AI-driven assessments in higher education.]]>
