Differential Privacy in Generative AI Agents: Analysis and Optimal Tradeoffs
Ya-Ting Yang, Quanyan Zhu · 2026
Read Differential Privacy in Generative AI Agents: Analysis and Optimal Tradeoffs by Ya-Ting Yang, Quanyan Zhu (2026) 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-2603-17902.txt.
Description
Yang and Zhu study privacy leaks from enterprise data inside AI agent replies. Most guard work watches user prompts, not the company data behind answers. Their frame treats reply generation as a random map from prompts and datasets to text. Token-level and message-level differential privacy bounds follow from that view. Leak size ties to settings like temperature and reply length, they show. A design problem then picks the best temperature trade-off.
Catalog record
| CREATOR | Ya-Ting Yang, Quanyan Zhu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30393 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.17902