GUI agent: Guided Exploration of User-Sensitive Screens

2026-06-24

AI Summary

This work addresses the security risks posed by large language model (LLM) agents operating in open graphical user interface (GUI) environments, where they may inadvertently trigger actions involving sensitive user information due to an inability to recognize sensitive states or request user intervention. To mitigate this, the paper proposes the first framework for identifying user-sensitive GUI states and generating interpretable queries to solicit user confirmation. The approach leverages demonstration-guided exploration, enabling LLM agents to traverse the GUI state space, automatically detect potentially sensitive operations, and formulate clear, context-aware queries. This mechanism establishes a controllable user takeover protocol and yields the first automatically constructed dataset of GUI-sensitive queries, laying a foundation for developing secure and reliable intelligent GUI agents.

Abstract

LLM agents are increasingly being used to automate tasks for users within an open GUI environment. They inevitably encounter screens containing user-sensitive information, for which takeover of task execution by the user is highly desirable or even necessary. State-of-the-art LLM-driven agents are usually fine-tuned to complete tasks regardless of the safety implications of their actions. This makes their real-world deployment difficult and adversely affects the reliability. Therefore, it is crucial to identify and categorize user-sensitive states and define user-sensitive queries. This dataset would be to engineers to recognize and request handover to the user in critical scenarios. This short paper develops an explorer agent that systematically explores the query space starting from one demonstrated task to identify queries that, if executed, would lead to user-sensitive states in a GUI environment.

Problem

• GUI agent
• user-sensitive information
• task automation
• LLM safety
• handover

Innovation

• GUI agent
• user-sensitive states
• guided exploration
• LLM safety
• task handover