Identity Negotiation in AI Styling Interactions on Short-Video Platforms

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

After generative AI entered everyday decision-making, how algorithmic aesthetic evaluation affects the identity construction of young users has become a question worth attention. This study adopts content analysis and examines 109 interaction units from 15 AI styling videos on short-video platforms. The findings show that identity negotiation in AI styling interactions mainly takes three forms, accommodative confirmation, collaborative adaptation, and defensive repair, and shows the feature of a separation between action power and interpretive power. While adopting AI suggestions, users maintain the interpretive power over the meaning of their own image through questioning, adjustment, and irony, and they turn algorithmic mismatch into performative content production. This study provides a new perspective for understanding human-machine communication and subjectivity construction in the era of generative AI.