Women's Faces in Beauty Algorithms: Dehumanization through AI-Based Applications

Abstrak

This study aims to examine how the algorithms of AI-based facial processing applications—Meitu, Hypic, Beautycam, and Wink—apply the principles of dehumanization that shape female beauty standards. The urgency of this research lies in understanding how visual technology constructs women's beauty identities through filters that subtly reproduce aesthetic biases, thereby shaping collective perceptions of beauty. The research approach uses qualitative methods with algorithmic and multimodal analysis. Data were obtained by processing photos of 10 female participants with different facial characteristics—skin tone, facial shape, and skin texture—using various AI filter features, interviewing participants about their experiences with these applications, and observing Instagram posts hashtagged #Meitu, #Hypic, #BeautyCam, and #Wink. Analysis of visual posts was conducted using Kress and van Leeuwen's Visual Grammar framework, which was then correlated with textual data in the form of post captions and comments. The findings are read through the perspective of Donna Haraway's Cyborg Theory to examine the body-technology relationship. The results of this study indicate that facial processing applications trigger dehumanization by homogenizing women's visual identities. Although Haraway views this as a form of empowerment, this research proves otherwise. The practice of facial manipulation actually alienates women from authority over their digital bodies, while perpetuating the collective memory of discriminatory beauty standards through this visual circulation. Haraway's concept of the cyborg can only be realized if digital infrastructure is first decolonized.