English(EN) The Irreversibility Budget: Fleet-Level Risk Accounting and Admission Control for Agent Operating Systems

新“不可逆性预算”论文解决AI代理舰队风险

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-02 04:00

一篇新论文提出了“不可逆性预算”来管理与AI代理舰队相关的风险。该系统旨在核算和控制跨多个代理、工作流和租户的累积残值风险，即使单个代理保持在其各自授权的门限内，也能防止舰队超出主体的风险限制。该提议的预算将不可逆性视为一种资源，对每种效应收取其残值损失，并在总预算透支后拒绝进一步的操作。虽然一项对照研究表明该预算可以抵御舰队级别的透支，但准确定价代理行为的异质性和相关效应仍然是实际部署中的一个重大开放性挑战。 AI

影响提出了一种新颖的AI代理舰队风险核算框架，解决了主体风险限制的潜在透支问题。

排序理由研究论文发表在arXiv上，详细介绍了AI代理的新风险管理系统。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 30 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 研究论文发表在arXiv上，详细介绍了AI代理的新风险管理系统。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Bardia Mohammadi, Laurent Bindschaedler · 2026-09-02 04:00

不可逆转预算：面向代理操作系统的大规模风险核算与准入控制

arXiv:2609.00275v1 Announce Type: new Abstract: Fleets of LLM agents now externalize effects that cannot be fully undone: they move money, deploy code, delete data, and disclose information. Current controls check one effect at a time, so a fleet of individually authorized agents…

本文如何被排名

报道来源 [1]

不可逆转预算：面向代理操作系统的大规模风险核算与准入控制

相关实体

相关话题