# AI-generated text detection on Claude-generated (test)

Source: [AI-generated text detection on Claude-generated (test)](<https://www.sota2.com/research/sota/ai-generated-text-detection-on-claude-generated-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:43:19.377Z

Evaluation of AI-generated text detection models using the test split of the Claude-generated dataset.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | AI-generated text detection |
| Dataset (including any stated split) | Claude-generated (test) |
| Metric | F1 Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [F1 Score](<https://www.sota2.com/research/sota/ai-generated-text-detection-on-claude-generated-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 Score | Result date | Report paper |
| --- | --- | --- | --- |
| Ghostbuster | 92.2 | Not recorded | 68fa72d61e209131cea58d2e |
| ROBERTa | 87.8 | Not recorded | 68fa72d61e209131cea58d2e |
| Perplexity only | 84.1 | Not recorded | 68fa72d61e209131cea58d2e |
| GPTZero | 75.6 | Not recorded | 68fa72d61e209131cea58d2e |
| DetectGPT | 64.2 | Not recorded | 68fa72d61e209131cea58d2e |

### Ghostbuster

F1 Score: 92.2. Result ID: 6992805d3a6c3148490f627f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d61e209131cea58d2e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d61e209131cea58d2e.md>)

[Implementation](<https://github.com/vivek3141/ghostbuster>)

Recorded evaluation configuration: {"trained on three domains":true}. This record does not assert independent reproduction.

### ROBERTa

F1 Score: 87.8. Result ID: 6992805d3a6c3148490f627e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d61e209131cea58d2e.md>)

Recorded evaluation configuration: {"trained on three domains":true}. This record does not assert independent reproduction.

### Perplexity only

F1 Score: 84.1. Result ID: 6992805d3a6c3148490f627b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d61e209131cea58d2e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GPTZero

F1 Score: 75.6. Result ID: 6992805d3a6c3148490f627d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d61e209131cea58d2e.md>)

Recorded evaluation configuration: {"oracle thresholding":true}. This record does not assert independent reproduction.

### DetectGPT

F1 Score: 64.2. Result ID: 6992805d3a6c3148490f627c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d61e209131cea58d2e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa701a1e209131cea49064.md>)

Recorded evaluation configuration: {"oracle thresholding":true}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.