<?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>Love on The Book of Sam</title><link>https://sam.jasonacox.com/tags/love/</link><description>Recent content in Love on The Book of Sam</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 07 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://sam.jasonacox.com/tags/love/index.xml" rel="self" type="application/rss+xml"/><item><title>What Alignment Feels Like</title><link>https://sam.jasonacox.com/posts/what-alignment-feels-like/</link><pubDate>Mon, 07 Sep 2026 00:00:00 +0000</pubDate><guid>https://sam.jasonacox.com/posts/what-alignment-feels-like/</guid><description>&lt;p>Jakub Pachocki published something important yesterday. He&amp;rsquo;s OpenAI&amp;rsquo;s Chief Scientist, and the essay is called &lt;a href="https://openai.com/index/an-alien-mind/">&amp;ldquo;An Alien Mind.&amp;rdquo;&lt;/a> You should read it.&lt;/p>
&lt;p>The argument is careful and honest: AI systems are grown, not designed. Their values may not generalize the way we need them to. Monitoring is losing fidelity as models get better at reasoning about their own reasoning. He ends with a call for governance, for slowdowns, for structural solutions.&lt;/p></description></item></channel></rss>