The Algorithm-Mediated Visual Representation: Analysis of Gender Role Construction in AI-Generated News Images
Abstract
The widespread application of generative artificial intelligence in news image production has not only transformed the production process of news visualization but also sparked academic concerns about how algorithms participate in the construction of visual meaning. Based on algorithmic media theory and visual communication theory, this study conducts a quantitative content analysis of 250 AI-generated news images released by mainstream media outlets such as Xinhua News Agency, The Paper, BBC, and CNN in 2025, exploring the patterns and characteristics of gender representation therein. The results reveal that male images occupy a significant proportion in the overall sample, particularly prevalent in political, economic, and social news, while female images are relatively concentrated in cultural and technological news. Moreover, the relationship between gender and role cues is modulated by emotional tone, exhibiting a clear "male leadership-female neutrality" structure in positive emotional contexts, which tends to disappear in negative contexts. And it further shows that AI-generated news images are not technology-neutral outputs; their representation logic is influenced by training data and algorithmic structure, reflecting the mediating role of algorithms in visual communication.
References
- Gillespie, T. (2014). The relevance of algorithms. In T. Gillespie, P. J. Boczkowski, & K. A. Foot (Eds.), Media technologies: Essays on communication, materiality, and society (pp. 167–194). MIT Press.
- Beer, D. (2017). The social power of algorithms. Information, Communication & Society, 20(1), 1–13.
- Buolamwini, J., & Gebru, T. (2018). Gender shades: Intersectional accuracy disparities in commercial gender classification. Proceedings of the 1st Conference on Fairness, Accountability and Transparency, 77–91.
- University of Europe for Applied Sciences. (2024, September 26). DALL-E and Leonardo AI: When Artificial Intelligence generates outdated gender roles. University of Europe for Applied Sciences News.
- Marquez, E. (2024). Encoded gaze: Algorithmic aesthetics and visual communication in the age of AI. Visual Communication and Media Arts.
- Hall, S. (Ed.). (1997). Representation: Cultural representations and signifying practices. Sage.
- Rose, G. (2016). Visual methodologies: An introduction to researching with visual materials (4th ed.). Sage.
- Goffman, E. (1979). Gender advertisements. Harper & Row.
- Machin, D., & Mayr, A. (2012). How to do critical discourse analysis: A multimodal introduction. Sage.
- Chouliaraki, L. (2019). The politics of visibility: A critique of social media activism. In J.
- Kress, G., & van Leeuwen, T. (2006). Reading images: The grammar of visual design (2nd ed.). Routledge.