This research aims to display how ChatGPT responds to questions that are by nature controversial and debatable as their values depend on the agency or power that answers them. The approaches employed in this work are Michel Foucault’s discourse analysis and Norman Fairclough’s critical discourse analysis (CDA). It is contended that discourses are ubiquitously everywhere in social practices. To Fairclough and Foucault, language production is strongly connected to power and knowledge production. Therefore, the work, utilising a qualitative method in the examination, wishes to identify the discourse(s) that can affect ChatGPT’s responses and discuss its implications for knowledge and power production. The research discovered that ChatGPT’s answers unveiled a prejudice towards one discourse, established knowledge, and suppressed alternative opinions. Also, the definitions and analyses provided by ChatGPT were insufficient because the political and ideological dimensions of the debate were ignored, which can be considered major factors in the course of misconceptualising and misunderstanding the two selected conversation topics. This indicates that ChatGPT’s knowledge production, as an example or model from the ongoing technological revolution, cannot be relied on in education because it produces biased knowledge and information.

Author: Karzan Aziz Mahmood

Published in: Canada International Conference on Education, 2024

• Date of Conference: 23-25 July, 2024
• DOI: 10.20533/CICE.2024.0081
• Electronic ISBN: 978-1-913572-65-5
• Conference Location: Toronto Metropolitan University, Toronto, Canada