‘ChatGPT can make mistakes’: pedagogically problematic discourses on generative AI in higher education

Fill in the blank: Generative AI (genAI) is a ________. This seemingly simplistic exercise reveals more than you might think, especially if you are a teacher. While most agree that words have implications, few have interrogated the pedagogical consequences of the dominant genAI discourses circulating in higher education. This Point of Departure argues that these discourses – often originating from big tech companies – shape how teachers and, ultimately, students understand and engage with genAI. Drawing on a poststructuralist view of discourse analysis and Lea and Street’s [1998. Student writing in higher education: An academic literacies approach. Studies in Higher Education 23, no. 2: 157–72] Academic Literacies framework, we argue that dominant utopian and dystopian framings focus on technological capabilities at the expense of pedagogical consequences. In response, we propose a human-centred model of discursive engagement that foregrounds discourses of curiosity, criticality, and collaboration. Ultimately, we aim to help teachers appreciate how the discourses they use to describe genAI affect how their students learn.