What is beauty? Society has always had its own ideas about beauty: in ancient times, these were carved in stone; later, they were painted on canvas; and in the 20th century, they were transferred to cinema and magazines. Today, algorithms apply these ideals and produce flawless skin, slim bodies, and eternal youth on a scale and at a pace that was previously unimaginable. Frank Richter discusses what this does to our self-image with Charlotte Tumescheit, mathematician and former postdoc at the University of Zurich, now a researcher at Idiap Research.
In the “UnrealBody” project, Charlotte investigated how image generators reproduce and reinforce ideals of beauty. The project was developed in collaboration with Janna Hastings, Paula Muhr, Lucas Caluori, Andrea B. Horn, and Sandra Koenig Heuer. Together, the team created a video designed to raise awareness among teenagers and young adults about the risks of AI-generated body images.
This episode of our podcast Schampar digital explores why AI models automatically generate unrealistic bodies from their training data, how this affects us, and why even knowing that an image is AI-generated does not protect us from negative effects. Charlotte also gives specific tips on what we can do in our everyday lives, from being more conscious about our own feed to spending more time offline.
The video was produced using Unreal Engine and AI-generated material, deliberately employing the same technologies that it critically examines. A pilot study involving students from the University of Zurich and international participants provided initial insights and was directly incorporated into the further development of the format.
Links and resources
- Project “UnRealBody: Inoculating Young People Against Harm from Unrealistic AI-Based Body Images“
- Offline Club Zürich
- Milton, A., Hambleton, A., Roberts, A., Davenport, T., Flego, A., Burns, J. & Hickie, I. (2021). Body Image Distress and Its Associations From an International Sample of Men and Women Across the Adult Life Span: Web-Based Survey Study. JMIR Formative Research, 5(11), e25329. https://doi.org/10.2196/25329
- Groesz, L. M., Levine, M. P. & Murnen, S. K. (2002). The effect of experimental presentation of thin media images on body satisfaction: A meta-analytic review. International Journal of Eating Disorders, 31(1), 1–16. https://doi.org/10.1002/eat.10005
- Fioravanti, G., Bocci Benucci, S., Ceragioli, G. et al (2022). How the Exposure to Beauty Ideals on Social Networking Sites Influences Body Image: A Systematic Review of Experimental Studies. Adolescent Res Rev 7, 419–458. https://doi.org/10.1007/s40894-022-00179-4
- McComb, S. E. & Mills, J. S. (2020). A systematic review on the effects of media disclaimers on young women’s body image and mood. Body Image, 32, 34–52. https://doi.org/10.1016/j.bodyim.2019.10.010
- Dhadly, P. K., Kinnear, A. & Bodell, L. P. (2023). #BoPo: Does viewing body-positive TikTok content improve body satisfaction and mood? Eating Behaviors, 50, Article 101761. https://doi.org/10.1016/j.eatbeh.2023.101747