Research report
Does Turnitin Detect Essays Generated by Claude-Fable-5?
A Benchmark Experiment on AI-Generated Essay Detection
Cite this research
In a benchmark of 170 AI-generated essays (131,451 words) by Claude Fable-5, Turnitin achieved 99.01% word-level accuracy, correctly identifying 130,151 words as AI-generated.
According to Turnitin0 Research, Turnitin correctly flagged 99.01% of words in 170 Claude Fable-5 essays, demonstrating high detection accuracy across 30 majors.
- 170 articles tested
- Checker: Turnitin
- LLM: claude-fable-5
- 1 linked dataset
- Detection screenshots available
- Published July 29, 2026
Peng, J. (2026). Does Turnitin Detect Essays Generated by Claude-Fable-5?. Turnitin0 Research. https://www.turnitin0.com/research/reports/does-turnitin-detect-essays-generated-by-claude-fable-5
Licensed under CC0 1.0 Universal. You may copy, modify, distribute, and cite this research—including text, figures, tables, and datasets—without asking permission.
Overview
This report presents a benchmark experiment evaluating Turnitin's ability to detect essays generated by the Claude Fable-5 language model. A dataset of 170 AI-generated essays (131,451 words) across 30 majors was submitted to Turnitin, and word-level accuracy was measured.
- Overall word accuracy: 99.01% (130,151 correct words out of 131,451 total words).
- Domain performance:
- Humanities: 99.16% accuracy (34,358/34,650 words).
- STEM: 99.03% accuracy (65,141/65,781 words).
- Business: 98.82% accuracy (22,542/22,811 words).
- Education: 98.77% accuracy (4,659/4,717 words).
- Psychology: 98.83% accuracy (3,451/3,492 words).
- Major-level variation:
- Highest accuracy: Criminal Justice (99.80%), Pre-Med (99.75%), Business Administration (99.64%).
- Lowest accuracy: Physics (96.52%), Accounting (98.00%), Sociology (98.36%).
- Level:
- Undergraduate: 99.07% accuracy (71,231/71,900 words).
- Graduate: 98.94% accuracy (58,920/59,551 words).
- Word count buckets:
- 800-word essays: 99.27% accuracy (47,151/47,498 words).
- 1200-word essays: 99.09% accuracy (58,826/59,367 words).
- 400-word essays: 98.32% accuracy (24,174/24,586 words).
- Dataset: 170 essays generated by Claude Fable-5, spanning 30 majors across five domains (Business, Education, Humanities, Psychology, STEM).
- Essay characteristics:
- Written at graduate and undergraduate levels.
- Word counts distributed among 400, 800, and 1200 words.
- Total word count: 131,451 words (excluding references/bibliography).
- All essays are non-ESL and belong to the essay genre.
- Evaluation: Each essay was submitted to Turnitin. Word-level accuracy was calculated as the percentage of words correctly flagged as AI-generated.
- Scoring unit: accuracy is measured in words. First identify correct paragraphs under the expectation below, then sum the words in those paragraphs for the correct-word count; total words are summed over all counted paragraphs.
- Correct paragraph: a counted paragraph that was flagged as AI. Correct words are the words in those paragraphs.
- References excluded: from a References/Bibliography-style heading onward, those paragraphs are left out of word counts.
- Dataset scope: Essays are limited to 30 majors and five domains; performance on other subjects or writing styles is unknown.
- Total articles: 170
- Correct words: 130,151 out of 131,451
- Word accuracy: 99.01%
- Business: 98.82% accuracy (22,542/22,811 words)
- Education: 98.77% accuracy (4,659/4,717 words)
- Humanities: 99.16% accuracy (34,358/34,650 words)
- Psychology: 98.83% accuracy (3,451/3,492 words)
- STEM: 99.03% accuracy (65,141/65,781 words)
- Highest accuracy: Criminal Justice (99.80%), Pre-Med (99.75%), Business Administration (99.64%)
- Lowest accuracy: Physics (96.52%), Accounting (98.00%), Sociology (98.36%)
- All other majors ranged from 98.62% to 99.58% accuracy.
- Undergraduate: 99.07% accuracy (71,231/71,900 words)
- Graduate: 98.94% accuracy (58,920/59,551 words)
- 400-word essays: 98.32% accuracy (24,174/24,586 words)
- 800-word essays: 99.27% accuracy (47,151/47,498 words)
- 1200-word essays: 99.09% accuracy (58,826/59,367 words)
Detection report screenshots
Turnitin marks suspected AIGC text in cyan.