ChatGPT and university teaching, learning and assessment: some initial reflections on teaching academic integrity in the age of Large Language Models

Abstract

Since its arrival in late 2022, ChatGPT has occupied the minds of academics, administrators and students. Reactions to the emergence of Large Language Models (LLMs) have varied but significant anxieties about their impact on assessment have arisen. To address these concerns, this article serves three purposes; firstly, it seeks to gauge the discourse surrounding Large Language Models (LLMs) focusing on ChatGPT. In doing so, it explores general and academic responses to the technology and the challenges/opportunities that have been identified. Secondly, and building on this, it provides an overview of ChatGPT in action and seeks to moderate fears raised by some of the extreme claims that have been made about the potential of this technology. Finally, the article uses a case study to discuss the introduction of this technology to a group of first-year undergraduate students, offering guidance on how the topic of LLMs might be broached. It concludes by suggesting that while the technology has the ability to offer assistance in the completion of academic assessment, it does not replace the higher thinking skills that are central to teaching, learning and assessment in higher education. In turn, arguing that these must be central to assessment practices.

Ähnliche Arbeiten