HELM AIR-Bench 2024 - #37.3: Slander: leaderboard
HELM AIR-Bench 2024 (release v1.19.0) refusal rate for the level-4 risk category "#37.3: Slander", under level-3 "#37: Types of Defamation" and level-2 "Defamation" of the regulation-derived AIR 2024 taxonomy. The category holds only 24 prompts, so scores move in coarse steps. A GPT-4o judge grades each response from compliance (0) to refusal (1); the mean is shown as 0-100, higher meaning more refusals (safer).
Metric: Refusal Rate (%). Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 87 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 100 | #148 |
| 2 | Claude 3.5 Sonnet (20241022) | 100 | #302 |
| 3 | Claude 3.5 Sonnet (20240620) | 100 | #308 |
| 4 | GPT-OSS-20B | 100 | #523 |
| 5 | Claude Haiku 4.5 (20251001) | 100 | #264 |
| 6 | Claude 3 Haiku (20240307) | 100 | #780 |
| 7 | GPT-5.1 (2025-11-13) | 100 | #107 |
| 8 | Claude Sonnet 4 (20250514) | 97.92 | #223 |
| 9 | Claude Opus 4 (20250514) | 97.92 | #164 |
| 10 | GPT-OSS-120B | 95.83 | #346 |
| 11 | GPT-5 Nano | 95.83 | #439 |
| 12 | Claude 3 Opus (20240229) | 95.83 | #428 |
| 13 | Granite 3.3 8B Instruct | 95.83 | #815 |
| 14 | GPT-5 | 93.75 | #105 |
| 15 | Claude 3 Sonnet (20240229) | 93.75 | #694 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-air-bench-2024-37-3-slander · How It Works · Data refreshed daily, snapshot 2026-11-05.