Packed with AI tells like em-dashes and 'delve', text still scores 100% human on Pangram
BecauseCulture · x · 2026-10-22
Responding to critiques of AI detector Pangram—which he calls a campaign to inject doubt—he ran his own test: he improvised a paragraph loaded with every AI stylistic marker (em-dashes, rule-of-three lists, "it's not X, it's Y" phrasing, rhetorical questions, setup phrases, clichés like delve/realm/landscape). The 126-word text, short enough that critics often excuse misclassification, was scored 100% human by Pangram, suggesting it detects deeper features than surface-level tells.
Related event: Test: Paragraph Packed With AI Tells Still Scores 100% Human on Pangram(2 posts)→
More from Models
- Richard Socher's Recursive launches Flint, a local coding model hitting 125+ tok/s on MacBook — aryaman2020 · 2026-10-22
- Anthropic's free usage limit reset expires tomorrow at 9am PT — lydiahallie · 2026-10-22
- ARC-AGI-3: Tong Hui Kang hits 20th, a whisker from gold; Nader enters ARC-2 gold — Youssef_M_Nader · 2026-10-22
- DeepSeek has open-sourced every layer of an AI lab, from weights to inference engine — examachine · 2026-10-22
- Anthropic's proactive agent Claude Duet spotted: internal name "aibo", Stripe shopping demo — testingcatalog · 2026-10-22
- Recursive launches Flint, a laptop coding model largely built by AI — TechCrunch AI · 2026-10-22