---
id: 20260715-T0-19
title: "AI最小自主性理论：超越最小权限原则"
title_en: "Minimum Autonomy Theory for AI Surpasses Principle of Least Privilege"
url: https://ai.daily.yangsir.net/daily/20260715-T0-19
issue_date: 2026-07-15
publish_date: 2026-07-14T04:00:00.000Z
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2607.09744
---

# AI最小自主性理论：超越最小权限原则

arXiv:2607.09744v1 公告类型：新论文 摘要：最小权限原则——即一个身份应仅持有其任务严格所需的权限——几十年来一直是访问控制的基础原语。我们认为，这一原则对于智能体AI系统而言并不足够，这类系统需要动态决策和自主行动能力。本文提出“最小自主性”理论，主张在赋予AI系统自主权时，应仅授予其完成任务所必需的最低程度的自主决策能力，并建立可量化的自主性边界。通过对比传统权限控制与自主性控制，我们展示了该理论在防止AI系统过度授权、降低意外行为风险方面的有效性。实验数据表明，在模拟的智能体任务场景中，应用最小自主性原则可将系统越权操作减少约73%，同时保持任务完成效率在95%以上。该工作为构建安全可控的自主AI系统提供了新的理论框架。

## English Version

**Minimum Autonomy Theory for AI Surpasses Principle of Least Privilege**

arXiv:2607.09744v1 announces a new paper proposing the 'Minimum Autonomy' theory for AI agents. The principle of least privilege—granting only the permissions strictly needed for a task—has long been foundational for access control, but the authors argue it is insufficient for intelligent agent systems that require dynamic decision-making and autonomous action. The new theory advocates granting AI agents only the minimum autonomous decision-making capability necessary to complete tasks, with quantifiable autonomy boundaries. Compared to traditional privilege control, this approach reduces unauthorized operations by approximately 73% in simulated agent tasks while maintaining task completion efficiency above 95%. The work provides a novel theoretical framework for building safe and controllable autonomous AI systems.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2607.09744)

**详情页**：https://ai.daily.yangsir.net/daily/20260715-T0-19

---

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