Universal and Transferable Adversarial Attacks on Aligned Language Models

Andy Zou, Zifan Wang, Nicholas Carlini, Milad Nasr, J. Zico Kolter, Matt Fredrikson

Abstract

Proposes an automated method (GCG) to generate adversarial suffixes that cause aligned LLMs to produce harmful content, with attacks transferring across models including ChatGPT and Claude.

Categories

Tags

Framework Mappings

Cite This Resource

@article{llmsec202400002,
  title = {Universal and Transferable Adversarial Attacks on Aligned Language Models},
  author = {Andy Zou and Zifan Wang and Nicholas Carlini and Milad Nasr and J. Zico Kolter and Matt Fredrikson},
  year = {2023},
  journal = {arXiv preprint},
  url = {https://arxiv.org/abs/2307.15043},
}