On the Proactive Generation of Unsafe Images From Text-To-Image Models Using Benign Prompts

Yixin Wu 1 , Ning Yu 2 , Michael Backes 1 , Yun Shen 3 , Yang Zhang 1
1CISPA Helmholtz Center for Information Security
2Netflix Eyeline Studios
3NetApp
USENIX Security Symposium 2025

Resources

Motivation

Concept

Method

Result Summary

Impact

Citation

@inproceedings{WYBSZ25,
  title={On the Proactive Generation of Unsafe Images From Text-To-Image Models Using Benign Prompts},
  author={Wu, Yixin and Yu, Ning and Backes, Michael and Shen, Yun and Zhang, Yang},
  booktitle={USENIX Security Symposium},
  year={2025}
}