# E5-small LoRA AI detector accuracy test

Source: https://toolbenchmarks.com/ai-detectors/e5-lora/  
Tested: 30 October 2026. Publisher: Tool Benchmarks. Data licence: CC BY 4.0.

> The E5-small LoRA detector flagged exactly half of the human texts in our test as AI (50.0%), including 75.0% of human forum answers, while catching 61.1% of AI texts. Its AUROC of 0.582 means its scores barely separate AI writing from human writing.

Verdict: Fast and tiny, but it flags half of all human writing, so its verdicts carry almost no information.

| Measure | Result |
|---|---|
| Catches AI text | 61.1% (308 of 504) |
| Flags human text | 50.0% (63 of 126) |
| Catches humanized AI text | 47.6% |
| AUROC | 0.582 |
| Catches Claude Fable 5.1 | 49.2% |
| Catches Claude Haiku 5.5 | 66.7% |
| Catches Claude Opus 5.5 | 64.3% |
| Catches Claude Sonnet 5.5 | 64.3% |
| Flags human text: Academic abstracts | 36.0% |
| Flags human text: Q&A answers | 75.0% |
| Flags human text: Encyclopedia articles | 52.0% |
| Flags human text: Fiction | 12.5% |
| Flags human text: News reports | 70.8% |

## What it is

This is the smallest model in our test, an E5-small text encoder fine-tuned with LoRA on the RAID dataset of human and AI writing. We used the standard probability threshold of 0.5.

## Results

It flagged 50.0% of human texts: 75.0% of forum answers, 70.8% of news reports and 52.0% of encyclopedia articles. Only fiction got through mostly unflagged, at 12.5%.

It caught 61.1% of AI texts, and like Fakespot it did best on encyclopedia and news writing, at 92.0% and 94.8%. Those are the same genres where it flags human writing most, which points to a detector reacting to genre style. Its AUROC of 0.582 confirms that its scores barely separate AI from human writing.

Longer human texts were flagged more often: 64.3% of human texts of 300 words or more, against 39.5% under 200 words. A humanizer rewrite lowered its catch rate on Claude Opus 5.5 texts from 64.3% to 47.6%.

## Questions people ask

### Is the E5-small LoRA AI detector accurate?

No. It flagged 50.0% of human texts and caught 61.1% of AI texts in our test, a balanced accuracy of 55.6%.

### What is the fastest AI detector you tested?

This one, at 0.03 seconds per 1,000 words on an Apple M4 Pro. Speed does not help when half of the human texts get flagged.