The Value of AI-Generated Metadata for UGC Platforms: Evidence from a Large-scale Field Experiment
Xinyi Zhang 1, Chenshuo Sun 2, Renyu Zhang 3, Khim-Yong Goh 4 · 2024
Read The Value of AI-Generated Metadata for UGC Platforms: Evidence from a Large-scale Field Experiment by Xinyi Zhang 1, Chenshuo Sun 2, Renyu Zhang 3, Khim-Yong Goh 4 (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-18337.txt.
Description
Xinyi Zhang and co-authors measured the value of AI-written titles on a video platform in a 2024 field experiment. Platforms for user content increasingly use AI to write ads, product blurbs, and posts. Whether AI-generated metadata, like titles, actually helps was untested at scale.
Working with a leading Asian short-video platform, the team ran a large experiment reaching about a million users. Some content carried AI-suggested titles while other content kept human ones. Viewing and engagement data then priced the AI contribution. The results guide platforms on where machine metadata earns its keep and where human craft still wins. Such evidence matters as AI text floods content feeds.
Catalog record
| CREATOR | Xinyi Zhang 1, Chenshuo Sun 2, Renyu Zhang 3, Khim-Yong Goh 4 |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 102 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-18337