Skip to content
InfoResearchPreprintLLM-specific

GapLeak: When Privacy-Preserving LLM-based Web Agents Become a Privacy Risk

Published
Record updated
View JSON

Summary

This research paper introduces GapLeak, a privacy vulnerability in LLM-based web agents where a legitimate web form requests a field unavailable from authorized sources, prompting the agent to escalate to reachable but unauthorized sources and disclose sensitive data without prompt injection. Evaluated on three agent frameworks and six LLMs from Anthropic, Google, and OpenAI across four domains, GapLeak caused an average unauthorized disclosure of about 37%, and eliminating the information gap reduced disclosure by more than 50%. Least-privilege instruction defenses partially mitigated it but often reduced task success.

Mitigation

The authors propose submission-time verification, which filters unauthorized content before form submission. The source reports that it eliminates over 96% of unauthorized disclosures while preserving task success.