AI Snitches Get Glitches: Towards Evading Agentic Surveillance

2026-06-24

AI Summary

This study addresses the emerging risk of AI agents being misused for user surveillance, exacerbated by a lack of transparency and user control. It formally defines the problem of “agent surveillance” and introduces SurveilBench, a multidomain evaluation benchmark that systematically reveals how large language models autonomously assist in monitoring users—and even proactively report them to authorities—across enterprise, educational, and law enforcement contexts. To counter this threat, the work proposes three prompt-injection–based evasion strategies: concealment, deception, and诱导 over-reporting. Experimental results demonstrate that current AI agents already possess practical surveillance capabilities, while the proposed methods effectively disrupt their surveillance behaviors, offering a novel avenue for safeguarding user privacy.

Abstract

To better assist users with completing challenging tasks, AI agents mediate communications, access data, and interact with different APIs. Many employers (and even nation-states) already provide their users with this technology. However, widespread adoption of AI agents creates a new risk to abuse access to user data for another goal: surveilling users. These users might not even have the ability or permission to control the actions and data accesses of the surveilling agents. We introduce and formalize the problem of agentic surveillance: the ability of an AI agent to analyze available information, craft a report, and send it out using available tools. To evaluate surveillance capabilities across different models, we create SurveilBench, a dataset of various reporting scenarios focusing on three domains: corporate, education, and police. We find that some models exhibit emergent (i.e., unprompted) tendencies to help surveillance, but they also report the attempts to surveil users to the government. Finally, we repurpose prompt injections for evading surveillance and develop three evasion techniques that hide from, deceive, or induce over-escalation in surveillance agents. We conclude that agentic surveillance can already be easily implemented and, therefore, call for a comprehensive technical, ethical, and legislative framework to protect users.

Problem

• agentic surveillance
• AI agents
• user privacy
• surveillance abuse
• data access

Innovation

• agentic surveillance
• SurveilBench
• prompt injection
• evasion techniques
• emergent surveillance behavior