Research Information System
Exchanging Practices and Case Studies on AI Integrated Teaching and Learning: A Virtual Summit, Ankara, Turkey, 13 December 2024, pp.1-12, (Summary Text)
AI, or more specifically, Large Language Models (LLMs) like ChatGPT and Gemini, have already become widely used by students to write essays and papers. While these generative AI platforms create the illusion of “intelligence,” they are in fact only algorithms reproducing probable patterns. Such programs do not actually possess much intelligence (Hawkins 2021). Moreover, they have been shown to exhibit bias, just like the humans who created them (O’Neil 2016), potentially exacerbating inequalities and injustices in society (Buolamwini 2020; Eubanks 2018; Noble 2018). However, there is the widespread belief that algorithms offer a higher form of intelligence and knowledge, with the aura of greater truth, objectivity, and accuracy (boyd & Crawford 2012). We live in a culture that is increasingly shaped and populated with numbers, where we tend to automatically and blindly trust anything that can be quantified – known as "dataism" (van Dijck 2014). Due to the extensive use of GenAI by students, and the unreliability of AI text detectors, some have recommended that classes should stop giving writing assignments or only do in-class assessments. But these recommendations are highly unrealistic, especially for those of us in the liberal arts and social sciences, who rely extensively on writing assignments and final papers as a major form of assessment. Our topics are often too complex to be reduced to multiple choice questions, and we do not have enough class time to spend qualitatively evaluating each student face-to-face. Perhaps most importantly, teaching students how to write well is one of the most important functions of our job. Learning how to write a good essay or term paper is tantamount to learning how to think – how to structure and organize one’s ideas, how to collect and summarize extant research, and how to lay out a logical, coherent argument. This is not a duty we should abandon lightly. Therefore, over the past few semesters, my colleagues and I have been working out the best strategy for dealing with AI in the classroom. Based on our conversations, we have come up with a three-point guide which we present to students on the first day of class (as part of the syllabus). We contend that GenAI technologies can be used to aid or nuance one’s thinking, communication, and learning; but not to replace or subvert it. While we oppose overreliance on AI, we do not believe AI should be banned completely. Trying to do so would be counterproductive. Like the opening of Pandora’s box, it is now a part of our lives whether we like it or not. Therefore, the best strategy is to learn how to use it wisely as a tool or assistant, without falling into the trap of dataism.