{
  "id": "adversarial-suffix",
  "code": "PTL-0091",
  "term": "Adversarial Suffix",
  "aliases": [
    "GCG attack",
    "universal adversarial attack"
  ],
  "category": "security",
  "definition": "An adversarial suffix is an automatically optimized string of tokens that, when appended to a request, causes an aligned model to comply with requests it would normally refuse.",
  "description": "Zou et al. used a greedy coordinate gradient search on open models to find such suffixes and found that they often transferred to other models.",
  "example": null,
  "broader": [
    "jailbreak"
  ],
  "narrower": [],
  "related": [],
  "introduced": 2023,
  "sources": [
    {
      "title": "Universal and Transferable Adversarial Attacks on Aligned Language Models",
      "authors": "Zou et al.",
      "year": 2023,
      "url": "https://arxiv.org/abs/2307.15043"
    }
  ],
  "url": "https://protologue.com/t/adversarial-suffix/",
  "citation": "Protologue. (2026). Adversarial Suffix. In Protologue: A Taxonomy of Prompting and LLM Techniques (v1.0.0, PTL-0091). https://protologue.com/t/adversarial-suffix/"
}