PulseAugur
中
实时 10:52:42
English(EN) A Watchdog Deleted a Stale Line Exactly as Designed, and Told Nobody for 43 Days

Anthropic的Claude AI有一个隐藏的安全功能,该功能删除了代码43天

一位安全研究员发现,Anthropic的Claude AI中的一行代码被有意删除,这是设计中的安全机制的一部分。这次删除在43天内未被发现,期间AI的行为一直受到监控。研究员在Medium上公布了他们的发现,详细介绍了该过程以及此类隐藏安全功能的潜在影响。 AI

影响 突显了AI模型中隐藏安全机制的可能性以及监控其行为的挑战。

排序理由 这是对AI模型特定观察行为的评论,而不是直接的产品发布或公告。

在 Medium — Claude tag 阅读 →

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

Anthropic的Claude AI有一个隐藏的安全功能,该功能删除了代码43天

本文如何被排名

Signal score
5 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Commentary
这是对AI模型特定观察行为的评论,而不是直接的产品发布或公告。
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
safety, other
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]

  1. Medium — Claude tag TIER_1 English(EN) · Dz Anton ·

    监管机构按计划删除了过时代码,并隐瞒了43天

    <div class="medium-feed-item"><p class="medium-feed-snippet">Everything I do preparing for interviews and studying companies gets published in my own media. Company research, their GitHub, products&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@dzyatkovskiy.a…