Unveiling gender bias in AI: A classroom reflection on media representation and AI literacy in postgraduate education

This reflection explores an in-class activity designed for postgraduate students in media and communication, aimed at enhancing their AI literacy, particularly regarding gender-related stereotypes in AI-generated media. In this seminar on gender representations, students were tasked with creating prompts devoid of any gender-specific hints and using AI to generate corresponding images. The class then analysed the results to identify any instances of gender bias in the AI outputs and discussed how these biases manifested visually, despite the neutral prompts. The exercise highlighted the subtle yet pervasive nature of gender bias in AI systems, encouraging students to critically examine the ways technology can perpetuate stereotypes. By focusing on how AI interprets and represents gender, the activity deepened students’ understanding of the ethical implications of using AI in media creation. Discussions extended to the importance of recognizing and addressing these biases in professional settings. This practice-based reflection outlines the pedagogical strategies employed to foster critical thinking and ethical awareness among postgraduate students, providing them with essential skills to navigate AI technologies responsibly. The activity also aligned with broader goals of integrating AI literacy in media education, preparing students to engage thoughtfully with generative tools in their future careers. This article was published open access under a CC BY licence: https://creativecommons.org/licenses/by/4.0/ .