Study Finds AI Humanizers Can Distort Facts While Evading Detection
A new arXiv paper audits AI 'humanizers' that rewrite machine-generated text to evade detectors, finding public leaderboards score evasion and embedding similarity but not factual accuracy.
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-11-03 05:00 · arXiv cs.CL
Humanized but Still True? Auditing Facts and Detection in AI-Text Humanizers