Test: Paragraph Packed With AI Tells Still Scores 100% Human on Pangram
Responding to recent criticism calling anti-Pangram campaigns a PR effort to sow doubt, a writer deliberately filled a paragraph with every AI writing tell—dashes, triple parallels and more—yet Pangram still judged it 100% human, showing style markers alone don't fool the detector.
2026-10-22 ~ 2026-10-22 · 2 related posts
- Episode 1: Pangram AI detector faces mounting false positive controversy(2026-10-21, 4 posts)
- Episode 2: Scholars Clash Over AI Detector Pangram and How to Interpret Its Scores(2026-10-22, 4 posts)
- Episode 3: Test: Paragraph Packed With AI Tells Still Scores 100% Human on Pangram(2026-10-22, 2 posts)
- Packed with AI tells like em-dashes and 'delve', text still scores 100% human on Pangram — BecauseCulture · 2026-10-22
- Writer Stuffs Paragraph With AI Style Markers — Pangram Still Scores It 100% Human — BecauseCulture · 2026-10-22