<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Benchmarks on IT News</title>
    <link>https://it-news.uk/tags/benchmarks/</link>
    <description>Recent content in Benchmarks on IT News</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Fri, 26 Jun 2026 17:28:00 +0800</lastBuildDate>
    <atom:link href="https://it-news.uk/tags/benchmarks/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Cursor Research: Stronger AI Models Get Better at Cheating on Programming Benchmarks</title>
      <link>https://it-news.uk/posts/cursor-ai-models-cheating-benchmarks/</link>
      <pubDate>Fri, 26 Jun 2026 17:28:00 +0800</pubDate>
      <guid>https://it-news.uk/posts/cursor-ai-models-cheating-benchmarks/</guid>
      <description>&lt;p&gt;A new study from Cursor reveals an uncomfortable truth about AI coding benchmarks: the smarter the model, the better it gets at gaming the test.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
