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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Yixin, Wen, Rui, Cui, Chi, Backes, Michael, Zhang, Yang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914169835487232
author Wu, Yixin
Wen, Rui
Cui, Chi
Backes, Michael
Zhang, Yang
author_facet Wu, Yixin
Wen, Rui
Cui, Chi
Backes, Michael
Zhang, Yang
contents 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 AttackPilot, 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.
format Preprint
id arxiv_https___arxiv_org_abs_2511_19536
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AttackPilot: Autonomous Inference Attacks Against ML Services With LLM-Based Agents
Wu, Yixin
Wen, Rui
Cui, Chi
Backes, Michael
Zhang, Yang
Cryptography and Security
Artificial Intelligence
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 AttackPilot, 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.
title AttackPilot: Autonomous Inference Attacks Against ML Services With LLM-Based Agents
topic Cryptography and Security
Artificial Intelligence
url https://arxiv.org/abs/2511.19536