arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.27140cs.CR
收藏

在来自不受信任第三方渠道的对抗性提示下衡量移动大语言模型代理的安全性

Measuring the Security of Mobile LLM Agents under Adversarial Prompts from Untrusted Third-Party Channels

William & Mary(威廉与玛丽学院);Carnegie Mellon University(卡内基梅隆大学);Nanyang Technological University(南洋理工大学)

Chenghao Du, Quanfeng Huang, Tingxuan Tang, Zihao Wang, Adwait Nadkarni, Yue Xiao

AI总结:

研究移动LLM代理在对抗性环境下的安全风险,设计评估一系列案例研究,涵盖多种攻击类型并涉及多个先进移动代理,通过大量试验揭示系统漏洞,映射攻击到相关框架发现新途径,证明其在现实中可被利用。

摘要:

大语言模型(LLMs)改变了软件开发,催生了基于LLM的代理应用。然而,在对抗性移动环境中部署此类代理的安全影响仍知之甚少。本文首次对移动LLM代理的安全风险进行系统研究。设计并评估了一系列对抗性案例研究,涵盖从弹出式广告到涉及恶意软件安装和跨应用数据泄露的高级端到端工作流程。评估了三种架构的八个先进移动代理,进行了2000多次对抗性和配对良性试验。结果揭示了系统漏洞,通过将这些攻击映射到MITRE ATT&CK移动框架,发现了LLM驱动自动化特有的新的权限提升和持久化途径。研究结果首次提供了端到端证据,表明在现实对抗环境中移动LLM代理是可被利用的,不受信任的第三方渠道是移动生态系统的固有部分。

英文摘要:

Large Language Models (LLMs) have transformed software development, enabling AI-powered applications known as LLM-based agents that promise to automate tasks across diverse apps and workflows. Yet, the security implications of deploying such agents in adversarial mobile environments remain poorly understood. In this paper, we present the first systematic study of security risks in mobile LLM agents. We design and evaluate a suite of adversarial case studies, ranging from opportunistic manipulations such as pop-up advertisements to advanced, end-to-end workflows involving malware installation and cross-app data exfiltration. Our evaluation covers eight state-of-the-art mobile agents across three architectures, with over 2,000 adversarial and paired benign trials. The results reveal systemic vulnerabilities: low-barrier vectors such as fraudulent ads succeed with over 80% reliability, while even workflows requiring the circumvention of operating-system warnings, such as malware installation, are consistently completed by advanced multi-app agents. By mapping these attacks to the MITRE ATT&CK Mobile framework, we uncover novel privilege-escalation and persistence pathways unique to LLM-driven automation. Collectively, our findings provide the first end-to-end evidence that mobile LLM agents are exploitable in realistic adversarial settings, where untrusted third-party channels (e.g., ads, embedded webviews, cross-app notifications) are an inherent part of the mobile ecosystem.

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。

↑