---
id: 20260510-T0-06
title: "多智能体AI系统权限失控？研究提议建立身份治理基础设施"
title_en: "Researchers Propose Identity Infrastructure to Fix Multi-Agent AI Authorization"
url: https://ai.daily.yangsir.net/daily/20260510-T0-06
issue_date: 2026-05-10
publish_date: 2026-05-09T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2605.05440
---

# 多智能体AI系统权限失控？研究提议建立身份治理基础设施

arXiv发表新论文探讨多智能体AI系统的授权传播问题。研究指出，当前安全讨论过度关注提示注入，忽视了多智能体系统中的权限失控风险：非人类主体在检索数据和委托任务时，系统授权约束可能被打破。论文提出将身份治理作为基础设施，确保AI系统在多级任务中的权限一致性，为企业构建AI工作流提供了安全架构参考。

## English Version

**Researchers Propose Identity Infrastructure to Fix Multi-Agent AI Authorization**

A new arXiv paper addresses authorization propagation in multi-agent AI systems. The research points out that current security discussions focus too heavily on prompt injection, ignoring the risk of broken authorization invariants when non-human principals retrieve data or delegate tasks. The paper proposes treating identity governance as infrastructure to ensure permission consistency across multi-level tasks, providing a security architecture reference for enterprises building AI workflows.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2605.05440)

**详情页**：https://ai.daily.yangsir.net/daily/20260510-T0-06

---

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