HumanizerEval

Leaderboard / Compare / WriteHuman vs Undetectable.ai

September 2026 review · same 49 texts

WriteHuman vs Undetectable.ai

In the September 2026 HumanizerEval review, WriteHuman and Undetectable.ai rewrote the same 49 English texts. WriteHuman cleared all 5 detectors on 32 of them; Undetectable.ai on 14. The difference is mostly Pangram: on 18 samples WriteHuman passed it where Undetectable.ai failed, and the reverse happened on 1. On GPTZero, Originality.ai, ZeroGPT and Winston AI the two were close to even.

Per-sample agreement

Per-sample agreement, September 2026. Counts are out of the 49 shared samples, computed at the 50% human-score threshold. Margin counts the samples only one tool passed, higher count first.
DetectorPassed more oftenMarginBoth passedNeither passed
Pangram 4.0WriteHuman18 to 11416
GPTZero 2026-09-13-baseWriteHuman2 to 0470
Originality.ai turboWriteHuman2 to 0461
ZeroGPT defaultTied0 to 0490
Winston AI 5.0WriteHuman1 to 0480
All 5 detectorsWriteHuman18 to 01417

Every count is computed from scores.csv at build time, not transcribed. “Neither passed” counts the texts that defeated both tools; on Pangram that was 16 of 49 here.

Quality and output discipline

Quality and output discipline, September 2026.
MeasureWriteHumanUndetectable.ai
22%26%
49%28%
98%94%
65%78%
1.021.21
07
00
API defaultUniversity / General Writing / More Human, v11sr
Websitewritehuman.ai undetectable.ai

On per-sample quality scores, WriteHuman scored higher on 19 samples, Undetectable.ai on 12, and the two tied on 18. WriteHuman’s output was shorter on 38 of 49 samples; mean length ratios were 1.02 and 1.21, or 2% longer than the input and 21% longer than the input respectively.

Where each one outperforms the other

WriteHuman

  • Pangram: passed 32 of 49 against 15, including 18 samples where Undetectable.ai failed, against 1 the other way.
  • GPTZero: passed 49 of 49 against 47, including 2 samples where Undetectable.ai failed, against 0 the other way.
  • Originality.ai: passed 48 of 49 against 46, including 2 samples where Undetectable.ai failed, against 0 the other way.
  • Winston AI: passed 49 of 49 against 48, including 1 sample where Undetectable.ai failed, against 0 the other way.
  • Naturalness: 49% against 28%.
  • Syntax integrity: 98% against 94%.
  • Length: shorter output on 38 of 49 samples, which matters under a word limit.

Undetectable.ai

  • Pangram: passed 1 sample where WriteHuman failed, even though it was detected more often overall, passing 15 to 32.
  • ZeroGPT: tied at 49 of 49 but on a higher mean human score, 95.0% to 94.8%. It passed 0 samples where WriteHuman failed.
  • Factual consistency: 26% against 22%.
  • Stance preservation: 78% against 65%.

WriteHuman vs Undetectable.ai FAQ

Is WriteHuman better than Undetectable.ai?

In the September 2026 HumanizerEval review, WriteHuman cleared all 5 detectors on 32 of the 49 shared samples and Undetectable.ai on 14. On GPTZero, Originality.ai, ZeroGPT and Winston AI the two were within 6 points of each other. Quality ran 19 samples to 12, with 18 ties.

Which humanizer is best for Pangram, WriteHuman or Undetectable.ai?

WriteHuman. In September 2026, WriteHuman passed Pangram on 32 of the 49 shared samples and Undetectable.ai on 15. On 18 samples WriteHuman passed where Undetectable.ai failed; the reverse happened on 1. Both failed on 16.

Which humanizer changes the text less?

WriteHuman. It scored higher on per-sample quality on 19 of 49 samples, against 12. WriteHuman’s outputs averaged 1.02× the input length and passed factual consistency on 22%; Undetectable.ai’s averaged 1.21× and passed on 26%. On stance preservation the split was 65% to 78%. WriteHuman produced the shorter output on 38 of 49 samples.

More comparisons