arXiv cs.AI· Rong Pan, Yili Hong, Min Xie·· 11 小时前AI 评分34
面向 AI 系统的可靠性工程：挑战、方法与方向
Reliability Engineering for AI Systems: Challenges, Methods, and Directions
AI 导读
一篇 arXiv 论文将传统可靠性工程方法（失效定义、运行包络、FMEA、加速测试、现场监控、可靠性增长）迁移到 AI 系统，提出按组件、运行回路、智能体行为、网络与治理划分的四级故障诊断框架。
正文 · 原文
Abstract:AI reliability concerns whether an AI system performs its intended function dependably over a stated period and under stated operating conditions, with clear evidence. As these systems become more autonomous, this function should include more than only a correct output. Retrieval, memory, tool use, permissions, human oversight, and interactions among systems must operate consistently and safely, and, for a generative AI, so must the reasoning process that produces the output. Average benchmark accuracy only measures system capability; it does not quantify this broader reliability claim. This paper adapts established reliability engineering methods, from failure definitions and operational envelopes to FMEA, accelerated testing, field monitoring, and reliability growth, to AI systems. A four-level diagnostic framework classifies failures as component, operational-loop, agentic-conduct, or network and governance failures. Test, evaluation, verification, and validation (TEVV), sequential monitoring, and FRACAS create and refresh evidence. The SMART framework provides statistical guidance for measurement, analysis, assessment, and test planning, and the NIST's AI risk management framework provides organizational guidance for governance, evaluation, monitoring, and mitigation. Three cases are discussed to illustrate these programs. It is shown that while the established reliability engineering principles can still provide a usable foundation, new measurements and safety guardrails are urgently needed as these systems are self-evolving at a fast pace.
| Subjects: | Artificial Intelligence (cs.AI); Applications (stat.AP) |
| MSC classes: | 86T05, 86T01 |
| ACM classes: | A.1; I.2 |
| Cite as: | arXiv:2609.35316 [cs.AI] |
| (or arXiv:2609.35316v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2609.35316 arXiv-issued DOI via DataCite |
Submission history
From: Rong Pan [view email]
[v1]
Mon, 28 Sep 2026 14:48:59 UTC (56 KB)
[v2]
Sun, 1 Nov 2026 03:01:54 UTC (812 KB)
[v3]
Tue, 3 Nov 2026 22:27:19 UTC (812 KB)
来源：arXiv cs.AI · arxiv.org