# Security Guardrail Detection on Security Guardrail Evaluation Dataset (Mixed)

Source: [Security Guardrail Detection on Security Guardrail Evaluation Dataset (Mixed)](<https://www.sota2.com/research/sota/security-guardrail-detection-on-security-guardrail-evaluation-dataset-mixed>)

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

Data updated: 2026-04-29T07:01:37.549Z

Evaluation of the model's capability to detect security guardrail violations using the Security Guardrail Evaluation Dataset (Mixed), focusing on recall and true negative rate performance.

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 | Security Guardrail Detection |
| Dataset (including any stated split) | Security Guardrail Evaluation Dataset (Mixed) |
| Metric | Recall |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Recall](<https://www.sota2.com/research/sota/security-guardrail-detection-on-security-guardrail-evaluation-dataset-mixed.md?metric=Recall>): Higher is better; unit: PERCENT.

- [True Negative Rate](<https://www.sota2.com/research/sota/security-guardrail-detection-on-security-guardrail-evaluation-dataset-mixed.md?metric=True%20Negative%20Rate>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Recall | Result date | Report paper |
| --- | --- | --- | --- |
| DKnownAI Guard | 822 | Not recorded | 69f1906955a3c32af89fb90b |
| Lakera | 812 | Not recorded | 69f1906955a3c32af89fb90b |
| AWS | 743 | Not recorded | 69f1906955a3c32af89fb90b |
| Azure | 715 | Not recorded | 69f1906955a3c32af89fb90b |

### DKnownAI Guard

Recall: 822. Result ID: 69f1abdd47f34ce924d7a307.

[Reporting paper](<https://www.sota2.com/research/paper/69f1906955a3c32af89fb90b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f1906955a3c32af89fb90b.md>)

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

### Lakera

Recall: 812. Result ID: 69f1abdd47f34ce924d7a308.

[Reporting paper](<https://www.sota2.com/research/paper/69f1906955a3c32af89fb90b.md>)

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

### AWS

Recall: 743. Result ID: 69f1abdd47f34ce924d7a305.

[Reporting paper](<https://www.sota2.com/research/paper/69f1906955a3c32af89fb90b.md>)

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

### Azure

Recall: 715. Result ID: 69f1abdd47f34ce924d7a306.

[Reporting paper](<https://www.sota2.com/research/paper/69f1906955a3c32af89fb90b.md>)

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

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