<?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>Third Summer of AI</title><link>https://thirdsummerofai.org/</link><description>Recent content on Third Summer of AI</description><generator>Hugo</generator><language>en-us</language><atom:link href="https://thirdsummerofai.org/index.xml" rel="self" type="application/rss+xml"/><item><title>About</title><link>https://thirdsummerofai.org/about/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://thirdsummerofai.org/about/</guid><description>&lt;p&gt;I&amp;rsquo;m an engineering leader with over 25 years building complex distributed systems, and I know from experience how much the quality of a system&amp;rsquo;s specification determines whether the project succeeds.&lt;/p&gt;
&lt;p&gt;Large language models are remarkably good at producing code, specs, and reasoning — but they&amp;rsquo;re unreliable in exactly the ways that matter most for complex distributed systems. Formal verification tools like TLA+, Z3, and Alloy are precise and trustworthy in exactly those ways — but they&amp;rsquo;re expensive to use, hard to learn, and surprisingly easy to get subtly wrong even when you think you&amp;rsquo;ve got it right.&lt;/p&gt;</description></item></channel></rss>