<?xml version="1.0" encoding="utf-8" standalone="yes" ?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Containment on Fabian G. Williams</title>
    <link>https://www.fabswill.com/tags/containment/</link>
    <description>Recent content in Containment on Fabian G. Williams</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en</language>
    <lastBuildDate>Sun, 11 Oct 2026 00:00:00 +0000</lastBuildDate>
    
	<atom:link href="https://www.fabswill.com/tags/containment/index.xml" rel="self" type="application/rss+xml" />
    
    
    <item>
      <title>Trust the Model the Least. I&#39;ve Been Building That at Home for Months.</title>
      <link>https://www.fabswill.com/blog/trust-the-model-the-least/</link>
      <pubDate>Sun, 11 Oct 2026 00:00:00 +0000</pubDate>
      
      <guid>https://www.fabswill.com/blog/trust-the-model-the-least/</guid>
      <description>BLUF (bottom line up front): This weekend Satya Nadella published Models as Insider Risks in the Super Intelligence Era. His line: &amp;ldquo;The most trustworthy Super Intelligence system will not be the one with the model we trust most. It will be the one that enables us to trust the model the least.&amp;rdquo; I agree, and I didn&amp;rsquo;t have to start this weekend. On my own hardware and my own time I&amp;rsquo;ve been running that architecture for months, and I&amp;rsquo;ve been writing it up in public as I went.</description>
    </item>
    
  </channel>
</rss>