IAAgent: Autonomous Inference Attacks Against ML Services With LLM-Based Agents

Yixin Wu, Rui Wen, Chi Cui, Michael Backes, Yang Zhang

Keywords: LLM-Based Agents, Risk Assessments, Membership Inference Attacks, Model Stealing Attacks, Data Reconstruction Attacks, Property Inference Attacks

Abstract: Inference attacks have been widely studied and offer a systematic risk assessment of ML services; however, their implementation and the attack parameters for optimal estimation are challenging for non-experts. The emergence of advanced large language models presents a promising yet largely unexplored opportunity to develop autonomous agents as inference attack experts, helping address this challenge. In this paper, we propose IAAgent, an autonomous agent capable of independently conducting inference attacks without human intervention. We evaluate it on 20 target services. The evaluation shows that our agent, using GPT-4o, achieves a 100.0\% task completion rate and near-expert attack performance, with an average token cost of only \$0.627 per run. The agent can also be powered by many other representative LLMs and can adaptively optimize its strategy under service constraints. We further perform trace analysis, demonstrating that design choices, such as a multi-agent framework and task-specific action spaces, effectively mitigate errors such as bad plans, inability to follow instructions, task context loss, and hallucinations. We anticipate that such agents could empower non-expert ML service providers, auditors, or regulators to systematically assess the risks of ML services without requiring deep domain expertise.

Primary Area: alignment, fairness, safety, privacy, and societal considerations

Submission Number: 11426