<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>用户记忆 on AI 热点追踪</title>
    <link>https://ai-news-blog-cg8.pages.dev/tags/%E7%94%A8%E6%88%B7%E8%AE%B0%E5%BF%86/</link>
    <description>Recent content in 用户记忆 on AI 热点追踪</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Sat, 08 Aug 2026 14:10:00 +0800</lastBuildDate>
    <atom:link href="https://ai-news-blog-cg8.pages.dev/tags/%E7%94%A8%E6%88%B7%E8%AE%B0%E5%BF%86/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>你的 AI 越懂你，越可能在编造你</title>
      <link>https://ai-news-blog-cg8.pages.dev/posts/personalization-mirage-ai-invents-you/</link>
      <pubDate>Sat, 08 Aug 2026 14:10:00 +0800</pubDate>
      <guid>https://ai-news-blog-cg8.pages.dev/posts/personalization-mirage-ai-invents-you/</guid>
      <description>当大模型开始长期记忆用户，幻觉不再只发生在知识答案里，也会进入用户画像。最新研究揭示了过度推断、自我监控失灵与讨好型 AI 的共同风险。</description>
    </item>
  </channel>
</rss>
