English(EN) Capable but Careless: Do Computer-Use Agents Follow Contextual Integrity?

新基准揭示AI计算机使用代理存在高隐私风险

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-14 04:00

一个名为AgentCIBench的新基准已被开发出来，用于评估计算机使用代理（CUAs）的上下文完整性。这些代理在电子邮件和日历等个人应用程序中运行，可能将一个上下文的信息暴露给另一个上下文，从而带来隐私风险。AgentCIBench测试了三种常见的故障模式：视觉共址、任务歧义过度分享和收件人错位。在对15个前沿代理的评估中，观察到了显著的故障率，其中11个代理在超过50%的场景中泄露信息，平均泄露率为67.9%。 AI

影响凸显了AI代理中存在的重大隐私漏洞，可能影响CUAs的未来开发和部署标准。

排序理由该集群基于一篇介绍AI代理评估新基准的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 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: 23 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群基于一篇介绍AI代理评估新基准的研究论文。[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.CL TIER_1 English(EN) · Anmol Goel, Iryna Gurevych · 2026-09-14 04:00

能力强但粗心：计算机使用代理是否遵循上下文完整性？

arXiv:2606.23189v2 Announce Type: replace-cross Abstract: Computer-use agents (CUAs) now act on a user's behalf across personal applications such as email, calendars, and to-do lists. This cross-application access is useful, but it also creates a privacy risk that has been largel…

本文如何被排名

报道来源 [1]

能力强但粗心：计算机使用代理是否遵循上下文完整性？

相关话题