Pandora:通过检索增强生成投毒对 GPTs 进行越狱攻击
Pandora: Jailbreak GPTs by Retrieval Augmented Generation Poisoning
AI总结:
研究针对LLMs(特别是GPTs)的间接越狱攻击,提出名为Pandora的RAG投毒方法,通过恶意构造内容操纵提示影响RAG过程,在四种场景下成功实施越狱且成功率高于直接攻击。
摘要:
大型语言模型(LLMs)获得了极大的普及,并日益广泛地应用于各个领域。因此,确保这些模型的安全至关重要。越狱攻击通过操纵 LLMs 生成恶意内容,已被视为一项重大的安全漏洞。尽管现有研究主要集中于对 LLMs 的直接越狱攻击,但对间接方法的探索仍然有限。将各种插件集成到 LLMs 中,尤其是检索增强生成(RAG),使得 LLMs 能够将外部知识库纳入其响应生成过程中(如 GPTs),这为间接越狱攻击引入了新的途径。为了填补这一空白,我们研究了对 LLMs(特别是 GPTs)的间接越狱攻击,引入了一种名为 Retrieval Augmented Generation Poisoning 的新型攻击向量。该方法名为 Pandora,通过提示操纵利用 LLMs 和 RAG 之间的协同作用,以生成意料之外的响应。Pandora 使用恶意构造的内容来影响 RAG 过程,从而有效地发起越狱攻击。我们的初步测试表明,Pandora 在四种不同场景下成功实施了越狱攻击,其成功率高于直接攻击,其中 GPT-3.5 的成功率为 64.3%,GPT-4 的成功率为 34.8%。
英文摘要:
Large Language Models~(LLMs) have gained immense popularity and are being increasingly applied in various domains. Consequently, ensuring the security of these models is of paramount importance. Jailbreak attacks, which manipulate LLMs to generate malicious content, are recognized as a significant vulnerability. While existing research has predominantly focused on direct jailbreak attacks on LLMs, there has been limited exploration of indirect methods. The integration of various plugins into LLMs, notably Retrieval Augmented Generation~(RAG), which enables LLMs to incorporate external knowledge bases into their response generation such as GPTs, introduces new avenues for indirect jailbreak attacks. To fill this gap, we investigate indirect jailbreak attacks on LLMs, particularly GPTs, introducing a novel attack vector named Retrieval Augmented Generation Poisoning. This method, Pandora, exploits the synergy between LLMs and RAG through prompt manipulation to generate unexpected responses. Pandora uses maliciously crafted content to influence the RAG process, effectively initiating jailbreak attacks. Our preliminary tests show that Pandora successfully conducts jailbreak attacks in four different scenarios, achieving higher success rates than direct attacks, with 64.3\% for GPT-3.5 and 34.8\% for GPT-4.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。