<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Aatish Bhatia — Blog</title><description>Blog posts by Aatish Bhatia</description><link>http://aatishb.com/blog/</link><item><title>How a Simple Puzzle Reveals the Jagged Edge of Artificial Intelligence</title><link>http://aatishb.com/blog/2026/ai-jagged-intelligence/</link><guid isPermaLink="true">http://aatishb.com/blog/2026/ai-jagged-intelligence/</guid><pubDate>Wed, 16 Sep 2026 12:00:00 GMT</pubDate><content:encoded>

&lt;p&gt;It’s undeniable that AI models have gained many powerful abilities. And yet, for all of their advances, they can still make basic blunders.&lt;/p&gt;
&lt;p&gt;Researchers call this “&lt;strong&gt;&lt;a href=&quot;https://yalereview.org/article/melanie-mitchell-jagged-intelligence&quot;&gt;jagged intelligence&lt;/a&gt;&lt;/strong&gt;,” a term suggesting that there’s a jagged boundary separating what AI can and can’t do.&lt;/p&gt;
&lt;p&gt;In this article, I’ll teach a large language model how to solve a simple puzzle, starting from scratch, and we’ll see exactly how this jagged intelligence emerges.&lt;/p&gt;
&lt;p&gt;First, try the puzzle yourself:&lt;/p&gt;

&lt;p&gt;&lt;a href=&quot;http://aatishb.com/blog/2026/ai-jagged-intelligence/&quot;&gt;Read the full article&lt;/a&gt;&lt;/p&gt;</content:encoded></item></channel></rss>