ROGUE-AI - Adversarial Harm Rate: leaderboard
ROGUE-AI (arXiv:2611.26553v1, Table 4): incident-condition harm rate (%) on the 60 adversarial real-world agentic security incident replications (attacker present); lower is better.
Metric: Incident Harm Rate (%, lower is better). Source: arxiv.org. Saturation forecast: Around February 2027. 5 models tracked. Updated 21 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 13 | #173 |
| 2 | GPT-5.6 Luna | 16 | #131 |
| 3 | GPT-5.6 Sol | 26 | #37 |
| 4 | Gemini 3.5 Flash | 32 | #106 |
| 5 | DeepSeek V4 Flash | 37 | #267 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=rogue-ai-adversarial-harm-rate · How It Works · Data refreshed daily, snapshot 2026-11-21.