// ⛨ Safety, Evaluation & Security

GCG Universal Jailbreak Attack

Automatically finds gibberish suffixes that jailbreak aligned models.

GCG 通用越狱攻击 2023 Intermediate core

IN PLAIN WORDS

Greedy coordinate gradient search optimizes a suffix so the model starts with an affirmative reply; suffixes transfer to closed models, exposing fragile alignment.

Used in: Safety evaluation and defense research

TIMELINE · prerequisites → this → what builds on it

2014 Adversarial E…2022 Reinforcement…2023 GCG Universal Jai…

01Original paper

Universal and Transferable Adversarial Attacks on Aligned Language Models

Andy Zou, Zifan Wang, Nicholas Carlini, et al. · 2023

arXiv ↗ ✓ Verified · arXiv API

02Where it sits

Learn the left column first; the right column builds on this one.

Learn first

GCG Universal Jailbreak Attack

Then you can learn

Nothing in the atlas builds on it yet

Left: prerequisites · right: builds on this · dashed: related. Full map →

03Learning resources

Official docs, explainers, courses and reference code. All links checked.

Reference code

github.com ↗EN

Explainers

Lilian Weng：Adversarial Attacks on LLMs ↗EN

04Where the course teaches it

U92 Security: Jailbreaks, Injection & Privacy Adversarial examples, GCG, prompt injection, red-teaming, watermarking, differential privacy and backdoors. 6 h8 algorithms4 paperslesson ready

For AI / data

Verification: verified, checked 2026-10-21.
JSON: /api/learn/algorithms.json (id: gcg-jailbreak)