---
id: 20260912-T0-09
title: "黑盒红队测试Agentic AI：分类驱动框架自动发现风险"
title_en: "Black-Box Red Teaming of Agentic AI: Taxonomy-Driven Framework for Automated Risk Discovery"
url: https://ai.daily.yangsir.net/daily/20260912-T0-09
issue_date: 2026-09-12
publish_date: 2026-09-11T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2609.09647
---

# 黑盒红队测试Agentic AI：分类驱动框架自动发现风险

一篇新论文提出针对Agentic AI系统的黑盒红队测试框架。Agentic系统正在快速进入生产环境，它们读取不可信输入、调用具有真实权限的工具并自主行动，安全面远超纯聊天模型。但现有评估方法仍停留在单轮对话层面，无法覆盖Agentic场景的风险。该论文提出分类驱动的自动化风险发现框架，试图填补这一空白。目前为论文阶段，实际效果有待验证。

## English Version

**Black-Box Red Teaming of Agentic AI: Taxonomy-Driven Framework for Automated Risk Discovery**

A new paper proposes a black-box red teaming framework for agentic AI systems. Agentic systems are rapidly moving to production, where they read untrusted inputs, call tools with real permissions, and act autonomously, expanding the security surface far beyond chat-only models. Yet standard evaluations remain single-turn and fail to cover agentic scenarios. The paper introduces a taxonomy-driven framework for automated risk discovery to fill this gap. It remains at the paper stage, with practical effectiveness yet to be validated.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2609.09647)

**详情页**：https://ai.daily.yangsir.net/daily/20260912-T0-09

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*