<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Prompt on Research Logs</title><link>https://williamresearch.com/tags/prompt/</link><description>Recent content in Prompt on Research Logs</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Mon, 24 Aug 2026 10:00:00 +0700</lastBuildDate><atom:link href="https://williamresearch.com/tags/prompt/index.xml" rel="self" type="application/rss+xml"/><item><title>🤖 AI's Forbidden Thought: The More You Say No, the More It Looks</title><link>https://williamresearch.com/posts/ai-forbidden-thought/</link><pubDate>Mon, 24 Aug 2026 10:00:00 +0700</pubDate><guid>https://williamresearch.com/posts/ai-forbidden-thought/</guid><description>&lt;h2 id="i-dont-see-her"&gt;“I don&amp;rsquo;t see her”&lt;/h2&gt;
&lt;p&gt;Picture a robot covering one eye with a metal hand. With the other hand, he points directly at the elegant robot standing behind him and insists:&lt;/p&gt;








&lt;blockquote class="blockquote-regular"&gt;
 &lt;p&gt;I don&amp;rsquo;t see her.&lt;/p&gt;

&lt;/blockquote&gt;
&lt;p&gt;It sounds convincing until you ask the obvious question: if he is not paying attention to her, how does he know exactly where she is?&lt;/p&gt;
&lt;p&gt;That is roughly what happens when we tell an AI system:&lt;/p&gt;</description></item></channel></rss>