← sources

A look at Pangram, considered the gold standard for detecting AI writing, and the dangers of its claimed one in 10,000 false-positive rate when used at scale (Matteo Wong/The Atlantic)

source: techmemepublished 2026-06-01 07:50ingested 2026-06-01 11:16

extract

Techmeme permalink: https://www.techmeme.com/260601/p13#a260601p13 <a href="https://www.theatlantic.com/technology/2026/05/pangram-ai-detection-accuracy/687381/?gift=NBdGSmKfDQzLc1B6N1F-gasbcqKrczuquHjg4eErYvw&amp;utm_source=copy-link&amp;utm_medium=social&amp;utm_campaign=share"><img align="RIGHT" border="0" hspace="4" src="http://www.techmeme.com/260601/i13.jpg" vspace="4" /></a> <p><a href="https://www.techmeme.com/260601/p13#a260601p13" title="Techmeme permalink"><img height="12" src="http://www.techmeme.com/img/pml.png" style="border: none; padding: 0; margin: 0;" width="11" /></a> Matteo Wong / <a href="http://www.theatlantic.com/">The Atlantic</a>:<br /> <span style="font-size: 1.3em;"><b><a href="https://www.theatlantic.com/technology/2026/05/pangram-ai-detection-accuracy/687381/?gift=NBdGSmKfDQzLc1B6N1F-gasbcqKrczuquHjg4eErYvw&amp;utm_source=copy-link&amp;utm_medium=social&amp;utm_campaign=share">A look at Pangram, considered the gold standard for detecting AI writing, and the dangers of its claimed one in 10,000 false-positive rate when used at scale</a></b></span>&nbsp; &mdash;&nbsp; AI-detection tools are getting better.&nbsp; But they still aren't good enough.&nbsp; &mdash;&nbsp; Basically every recent, high-profile accusation &hellip; </p>

retained text

No document body was retained for this record — only its metadata was stored.