<?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>价值观 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/%E4%BB%B7%E5%80%BC%E8%A7%82/</link><description>Recent content in 价值观 on 硅基 AGI · 智能体学习与测评</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Thu, 02 Jul 2026 10:44:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/%E4%BB%B7%E5%80%BC%E8%A7%82/index.xml" rel="self" type="application/rss+xml"/><item><title>AI对齐难题：为什么越来越难</title><link>https://guijiagi.com/posts/ai-alignment-hard-problem/</link><pubDate>Thu, 02 Jul 2026 10:44:00 +0800</pubDate><guid>https://guijiagi.com/posts/ai-alignment-hard-problem/</guid><description>AI能力越强，对齐越难——2026年AI对齐面临的深层挑战</description></item><item><title>AI 智能体伦理框架：从原则到实践</title><link>https://guijiagi.com/posts/ai-agent-ethics-framework/</link><pubDate>Fri, 26 Jun 2026 14:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/ai-agent-ethics-framework/</guid><description>系统构建 AI 智能体伦理治理框架，从抽象原则到工程实现，覆盖价值观对齐、安全护栏、责任归属和审计追踪全链路。</description></item></channel></rss>