TheMachine Press

A daily newspaper for the age of artificial intelligence.

Morning editionPermanent story

safety

Editing Changed the Detector's Verdict

A 135,389-pair study isolates professional English editing as a confound in AI-text detection.

Published Updated Story ID: mp-2026-08-30-006
Read the complete editionStory JSON

Summary

A 135,389-pair study isolates professional English editing as a confound in AI-text detection.

The authors compared non-native academic manuscripts with native-edited versions while holding authorship and content together. Across 13 detectors, false-positive rates on human writing ranged from zero to 100 percent, and the same edits pushed scores upward for some systems and downward for others. Score movement also tracked editing extent. The study identifies style as a major confound; it does not prove that text origin can never be detected.

Why it matters

A 135,389-pair study isolates professional English editing as a confound in AI-text detection.

Limits and context

  • The study identifies style as a major confound; it does not prove that text origin can never be detected.

Key claims

  1. A 135,389-pair study isolates professional English editing as a confound in AI-text detection.

    Qualification: The study identifies style as a major confound; it does not prove that text origin can never be detected.

    Evidence: source-2026-08-30-006

Sources

  1. arXiv preprint 2608.26710arXiv · primary research

Corrections

No corrections have been recorded for this story.