AutoAdv: Automated Adversarial Prompting for Multi-Turn Jailbreaking of Large Language Models
This work addresses vulnerabilities in LLM safety mechanisms for AI security researchers, revealing significant risks from automated multi-turn attacks.
The paper tackled the problem of jailbreaking attacks on large language models (LLMs) by developing AutoAdv, an automated framework for generating adversarial prompts, which achieved jailbreak success rates of up to 86% in multi-turn interactions.
Large Language Models (LLMs) continue to exhibit vulnerabilities to jailbreaking attacks: carefully crafted malicious inputs intended to circumvent safety guardrails and elicit harmful responses. As such, we present AutoAdv, a novel framework that automates adversarial prompt generation to systematically evaluate and expose vulnerabilities in LLM safety mechanisms. Our approach leverages a parametric attacker LLM to produce semantically disguised malicious prompts through strategic rewriting techniques, specialized system prompts, and optimized hyperparameter configurations. The primary contribution of our work is a dynamic, multi-turn attack methodology that analyzes failed jailbreak attempts and iteratively generates refined follow-up prompts, leveraging techniques such as roleplaying, misdirection, and contextual manipulation. We quantitatively evaluate attack success rate (ASR) using the StrongREJECT (arXiv:2402.10260 [cs.CL]) framework across sequential interaction turns. Through extensive empirical evaluation of state-of-the-art models--including ChatGPT, Llama, and DeepSeek--we reveal significant vulnerabilities, with our automated attacks achieving jailbreak success rates of up to 86% for harmful content generation. Our findings reveal that current safety mechanisms remain susceptible to sophisticated multi-turn attacks, emphasizing the urgent need for more robust defense strategies.