<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>状态机架构 on AI内参</title>
    <link>https://neican.ai/tags/%E7%8A%B6%E6%80%81%E6%9C%BA%E6%9E%B6%E6%9E%84/</link>
    <description>Recent content in 状态机架构 on AI内参</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Thu, 14 May 2026 16:10:04 +0800</lastBuildDate>
    <atom:link href="https://neican.ai/tags/%E7%8A%B6%E6%80%81%E6%9C%BA%E6%9E%B6%E6%9E%84/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>当AI开始“甩锅”：智能体协作架构的信任危机与地基重构</title>
      <link>https://neican.ai/insights/article-20260514161004848-2/</link>
      <pubDate>Thu, 14 May 2026 16:10:04 +0800</pubDate>
      <guid>https://neican.ai/insights/article-20260514161004848-2/</guid>
      <description>本文深度剖析了Claude智能体中爆发的“角色混淆”漏洞，指出这是长上下文时代AI系统在指令优先级与归因逻辑上的结构性缺陷。文章进一步探讨了从单纯Prompt工程向严密状态机（FSM）与上下文治理转型的发展路径，强调了提升AI Agent信任边界的紧迫性。</description>
    </item>
  </channel>
</rss>
