arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2402.19200cs.CRcs.CL
收藏

PRSA: 针对真实世界提示词服务的提示词窃取攻击

PRSA: Prompt Stealing Attacks against Real-World Prompt Services

Zhejiang University(浙江大学);Stony Brook University(石溪大学)

Yong Yang, Changjiang Li, Qingming Li, Oubo Ma, Haoyu Wang, Zonghui Wang, Yandong Gao, Wenzhi Chen, Shouling Ji

AI总结:

针对真实世界提示词服务的泄露风险,提出实用攻击框架PRSA,通过有限输入输出分析窃取提示词,在两类服务中显著提升攻击成功率,并基于互信息发现指导防御设计,已上报厂商推动防护落地。

摘要:

近年来,大语言模型(LLM)凭借其卓越性能获得了广泛关注。提示词是LLM功能与性能的核心,因此是极具价值的资产。对高质量提示词日益增长的依赖推动了提示词服务的大幅增长,但这种增长也扩大了提示词泄露的可能性,增加了攻击者复制原有功能、打造竞品并严重侵犯开发者知识产权的风险。尽管存在这些风险,真实世界提示词服务中的提示词泄露问题仍未得到充分研究。\n本文提出PRSA,一种用于提示词窃取的实用攻击框架。PRSA通过极有限的输入-输出分析推断提示词的详细意图,能够成功生成可复现原有功能的窃取提示词。大量评估证明PRSA在两类主要的真实世界提示词服务中均具备有效性。具体而言,与已有研究相比,它在提示词市场的攻击成功率从17.8%提升至46.1%,在LLM应用商店的攻击成功率从39%提升至52%。值得注意的是,在攻击OpenAI GPT Store中最受欢迎的教育类应用之一、对话量超100万的“Math”时,PRSA发现了一个此前未被披露的隐藏彩蛋。此外,我们的分析显示,提示词与其输出之间的互信息越高,泄露风险就越大。这一发现为设计和评估两种应对PRSA带来的安全威胁的潜在防御方案提供了指导。我们已将这些发现上报给PromptBase和OpenAI等提示词服务提供商,并正与它们积极合作落实防御措施。

英文摘要:

Recently, large language models (LLMs) have garnered widespread attention for their exceptional capabilities. Prompts are central to the functionality and performance of LLMs, making them highly valuable assets. The increasing reliance on high-quality prompts has driven significant growth in prompt services. However, this growth also expands the potential for prompt leakage, increasing the risk that attackers could replicate original functionalities, create competing products, and severely infringe on developers' intellectual property. Despite these risks, prompt leakage in real-world prompt services remains underexplored. In this paper, we present PRSA, a practical attack framework designed for prompt stealing. PRSA infers the detailed intent of prompts through very limited input-output analysis and can successfully generate stolen prompts that replicate the original functionality. Extensive evaluations demonstrate PRSA's effectiveness across two main types of real-world prompt services. Specifically, compared to previous works, it improves the attack success rate from 17.8% to 46.1% in prompt marketplaces and from 39% to 52% in LLM application stores, respectively. Notably, in the attack on "Math", one of the most popular educational applications in OpenAI's GPT Store with over 1 million conversations, PRSA uncovered a hidden Easter egg that had not been revealed previously. Besides, our analysis reveals that higher mutual information between a prompt and its output correlates with an increased risk of leakage. This insight guides the design and evaluation of two potential defenses against the security threats posed by PRSA. We have reported these findings to the prompt service vendors, including PromptBase and OpenAI, and actively collaborate with them to implement defensive measures.

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。

↑