arXiv cs.AI· Yue Zhao, Jiate Li, Li Li, Yi Nian, Jinbo Liu, Xiaolin Zhou, Xiyang Hu·· 10 天前AI 评分24
如何让 AI 智能体的声明可审计
Auditable Claims about AI Agents
AI 导读
针对组织对 AI 智能体所作声明（如每封外发邮件均经人工批准、每次操作均有日志）难以核查的问题，研究者提出声明可审计需在裁决前固定策略、范围、可定论记录及记录者与判定规则。
正文 · 原文
Abstract:Organizations make claims about their AI agents: a person approves every external email, every action is logged, an evaluation shows the agent is safe to deploy. Article 12 of the EU AI Act requires high-risk systems to allow the automatic recording of events but does not say which records settle a given claim. The position is one sentence: to be checked, a claim about an agent must first name its policy, its scope, the records that would settle it, and who writes them. Adapting the preconditions of an assurance engagement, we call a claim auditable when these elements and a decision rule are fixed before any verdict and the records are obtainable. This extends the Policy Checkability dimension of our Auditable Agents framework from single actions to claims. Agents add three conditions: coverage by an independent record, authorization bound to each action's arguments, and completeness beyond integrity. Under an explicit model, we prove that support is impossible without each wherever its hypotheses hold. A claim-check table applies the method to six common claims, anchored in current NIST, IETF, and OWASP drafts. A worked case follows one claim through five evidence states. We close with a practice box and steps for operators, buyers, auditors, and standard setters.
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2610.07459 [cs.AI] |
| (or arXiv:2610.07459v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07459 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Xiyang Hu [view email]
[v1]
Mon, 5 Oct 2026 22:11:11 UTC (945 KB)
来源：arXiv cs.AI · arxiv.org