arXiv cs.AI· Alexandra E. Michael, Franziska Roesner·· 12 小时前AI 评分36
AI 智能体如何请求权限：从界面到执行，AI 智能体的用户级权限机制
How Agents Ask for Permission: User Permissions for AI Agents, from Interfaces to Enforcement
AI 导读
一项研究调研了 21 个用户级 AI 智能体权限系统，并构建了权限指定、推导与执行方式的分类体系，同时对比了 5 个商用智能体。研究发现，尽管相关提案多样，但在可用性、透明度和用户控制方面仍存在明显缺口，学术智能体与商用智能体之间的差距尤为突出。
正文 · 原文
Abstract:As AI agents grow in prevalence, so too do their risks to users, creating a need for robust and usable systems for agent permissions. Many recent proposals implement product-level permissions, where the same security policy is applied for all users of a given agent framework. Yet different users have varying needs, necessitating support for user-level permission policies in agentic AI systems.
We survey 21 proposed user-level agent permissions systems and construct a taxonomy of approaches to specifying, deriving, and enforcing user-level permissions. We further analyze and compare against five commercial agents. We find that despite a wide diversity of proposals, sizable gaps remain in usability, transparency, and user control, particularly between academic and commercial agents.
| Comments: | Accepted at the Third Workshop on Agents in the Wild @ NeurIPS 2026. 9 pages, 2 tables, plus appendices |
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.13718 [cs.CR] |
| (or arXiv:2607.13718v3 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2607.13718 arXiv-issued DOI via DataCite |
Submission history
From: Alexandra E. Michael [view email]
[v1]
Wed, 15 Jul 2026 11:36:22 UTC (239 KB)
[v2]
Mon, 20 Jul 2026 11:32:26 UTC (239 KB)
[v3]
Thu, 15 Oct 2026 22:49:38 UTC (238 KB)
来源：arXiv cs.AI · arxiv.org