The Visibility of AI Content Labels: A Comparative Study of Label Presentation Across Four Short-Video Platforms
Abstract
The continuous entry of AI-generated content (AIGC) into short-video platforms has made AI labels an important cue for users to identify content sources, yet having a label does not mean the label is prominent enough. This paper selects 50 videos carrying AI labels from each of Douyin, Kuaishou, Bilibili, and Xiaohongshu, and compares them across three aspects: label position, visual salience, and information depth. The study finds that the AI labels of all four platforms are visible on the first screen, but most are located in auxiliary information areas with limited visual prominence; the platforms also differ in label wording and follow-up explanatory information. This shows that AI content labeling must address not only the question of "whether there is a label" but also "how to make it seen" and "what can be known after seeing it".
References
- Cyberspace Administration of China, Ministry of Industry and Information Technology, Ministry of Public Security. (2022, November 25). Provisions on the administration of deep synthesis of internet information services. https://www.cac.gov.cn/2022-12/11/c_1672221949318230.htm.
- Cyberspace Administration of China, Ministry of Industry and Information Technology, Ministry of Public Security, National Radio and Television Administration. (2025, March 14). Measures for the labeling of AI-generated synthetic content (Guoxinban Tongzi [2025] No. 2). https://www.cac.gov.cn/2025-03/14/c_1743654685899683.htm.
- Standardization Administration of China. (2025, September 1). Cybersecurity technology--Method for labeling AI-generated synthetic content (GB 45438-2025). https://www.sac.gov.cn/xw/bzhyw/art/2025/art_31a0f706b91d4d41b38f8f68b3d09621.html.
- European Parliament. (2024, June 13). Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act). https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32024R1689.
- California Legislature. (2024, September 19). California AI Transparency Act (SB 942). https://legiscan.com/CA/text/SB942/id/2964649.
- Zhao, X. (2025). South Korea passes the basic act on artificial intelligence.Prosecutorial View, (10), 64-65.
- 'Are the short videos you watch labeled AI-generated?'. (2025).Communist, (22), 46.
- Xiao, R. (2025, September 8). AI-generated content labeling is both a 'firewall' and an 'escort'.Consumer Daily, A01.
- Li, J., Chen, S., Ren, S., et al. (2026). Optimization of risk grading and labeling methods for AI-generated short-video communication.Youth Journalist, (3), 105-112.
- He, Y. (2008). Visual salience and information delivery--On the visual hierarchy of layout organization.Journal of Zhanjiang Normal College, 29(4), 137-140.
- Research Office One, General Office of the CPPCC Committee of the Autonomous Region. Standardizing prompt-language labels on short-video platforms.Huaxing Times, 2026-03-25(004).
- Yin, J., & Zhou, M. (2020). Research on short-video APP design based on user experience.Packaging Engineering, 41(6), 198-204.
- Luo, Y., & Wang, J. (2025, September 3). AIGC labeling measures take effect in September; platforms and large model companies respond by 'adding watermarks'.The Beijing News. https://m.bjnews.com.cn/detail/1756879784168649.html#1.
- Announcement on autonomous labeling of AI-generated synthetic content. (2026, February 12).Xiaohongshu. https://www.xiaohongshu.com/explore/698db691000000000a0310cf?xsec_token=CBmQG3G7qH3Up7WPjpDrXPVZ37kzHONdL-_bnKaDAj934=&xsec_source=app_share.
- Huo, Z. (2021).Research on fashion short-video interface design based on eye-tracker testing[Doctoral dissertation, Beijing Institute of Fashion Technology].