SafeMobile: Chain-level Jailbreak Detection and Automated Evaluation for Multimodal Mobile Agents

TL;DR

This paper introduces SafeMobile, a system that detects jailbreak attempts in multimodal mobile agents by analyzing behavioral sequences and using large language models, enhancing security in complex interaction scenarios.

Contribution

The work presents a novel risk discrimination mechanism and an automated assessment scheme for mobile multimodal agents, addressing limitations of existing security measures.

Findings

Improved detection of risky behaviors in high-risk tasks.

Reduction in the probability of agents being jailbroken.

Enhanced recognition of security threats through behavioral sequence analysis.

Abstract

With the wide application of multimodal foundation models in intelligent agent systems, scenarios such as mobile device control, intelligent assistant interaction, and multimodal task execution are gradually relying on such large model-driven agents. However, the related systems are also increasingly exposed to potential jailbreak risks. Attackers may induce the agents to bypass the original behavioral constraints through specific inputs, and then trigger certain risky and sensitive operations, such as modifying settings, executing unauthorized commands, or impersonating user identities, which brings new challenges to system security. Existing security measures for intelligent agents still have limitations when facing complex interactions, especially in detecting potentially risky behaviors across multiple rounds of conversations or sequences of tasks. In addition, an efficient and consistent automated methodology to assist in assessing and determining the impact of such risks is currently lacking. This work explores the security issues surrounding mobile multimodal agents, attempts to construct a risk discrimination mechanism by incorporating behavioral sequence information, and designs an automated assisted assessment scheme based on a large language model. Through preliminary validation in several representative high-risk tasks, the results show that the method can improve the recognition of risky behaviors to some extent and assist in reducing the probability of agents being jailbroken. We hope that this study can provide some valuable references for the security risk modeling and protection of multimodal intelligent agent systems.