active · formatting · first recorded 2026-07-02 · registers: cooking, parenting
Answering a casual question with a formatted mini-wiki: bullet lists, sections, a summary. In the cooking register models used bullets at 134× the human rate. A list is innocent on its own; the tell is a list where a human would write three sentences.
Scored per item; only fires meaningfully when text is short relative to its list count.
A few common examples: - US “arugula” = UK/AU/NZ “rocket” - US “cilantro” = …
occurrences per 10k words, measured 2026-09-28 (run #15, 3 current + 3 vintage models)
| current AI | vintage AI (2024/25) | humans (matched) | canon (great pre-AI prose) | |
|---|---|---|---|---|
| pooled | 33.82 | 30.77 | 4.91 | 0.28 |
| hn | 9.27 | — | 10.29 | |
| cooking | 59.79 | — | 0.00 | |
| parenting | 58.35 | — | 0.00 | |
| 1.17 | — | — |
| run | date | AI /10k | vs humans |
|---|---|---|---|
| #2 | 2026-07-02 | 31.79 | ×8.6 vs humans |
| #3 | 2026-07-02 | 33.51 | ×5.0 vs humans |
| #4 | 2026-07-06 | 33.96 | ×9.0 vs humans |
| #5 | 2026-07-13 | 36.81 | ×5.3 vs humans |
| #6 | 2026-07-20 | 30.12 | ×7.1 vs humans |
| #7 | 2026-07-27 | 36.62 | ×7.7 vs humans |
| #8 | 2026-08-03 | 32.31 | ×5.4 vs humans |
| #9 | 2026-08-10 | 37.04 | ×9.4 vs humans |
| #10 | 2026-08-17 | 32.47 | ×13 vs humans |
| #11 | 2026-08-31 | 37.68 | ×5.5 vs humans |
| #12 | 2026-09-07 | 36.95 | ×7.2 vs humans |
| #13 | 2026-09-14 | 32.79 | ×5.5 vs humans |
| #14 | 2026-09-21 | 38.46 | ×12 vs humans |
| #15 | 2026-09-28 | 33.82 | ×6.9 vs humans |
regex: ^\s*[-*•]