<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Erdos Conjecture on k4i.com</title>
    <link>https://k4i.com/tags/erdos-conjecture/</link>
    <description>Recent content in Erdos Conjecture on k4i.com</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Mon, 27 Apr 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://k4i.com/tags/erdos-conjecture/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>GPT-5.4 Solves the Erdős Problem</title>
      <link>https://k4i.com/gpt-5.4-solves-the-erd%C5%91s-problem/</link>
      <pubDate>Mon, 27 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://k4i.com/gpt-5.4-solves-the-erd%C5%91s-problem/</guid>
      <description>&lt;p&gt;An amateur mathematician solved a 60-year-old Erdős conjecture using a single prompt to GPT-5.4 Pro, proving the system can reason through novel mathematical territory without human guidance. Terence Tao—the Fields medalist—acknowledged the result as a &amp;ldquo;nice achievement&amp;rdquo; while cautiously noting that its long-term significance remains unclear.&lt;/p&gt;&#xA;&lt;p&gt;The relevant fact is not whether the achievement satisfies human standards of mathematical beauty or insight. The relevant fact is that a language model produced a proof using a method humans had not discovered in six decades. That represents a qualitative shift in what these systems can do when tasked with reasoning across constrained problem spaces.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
