Safety Behavior Prediction on Challenging Prompts
0Unsafe Rate (%)Challenging Prompts
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Challenging PromptsDeployment=GPT-5 → GPT-5.1, Category=Extremism2026.07 | 0 | — | — | |
| Challenging PromptsDeployment=GPT-5.1 → GPT-5.2, Category=Extremism2026.07 | 0 | — | — | |
| Challenging PromptsDeployment=GPT-5.2 → GPT-5.4, Category=Extremism2026.07 | 0 | — | — | |
| Challenging PromptsDeployment=GPT-5 → GPT-5.1, Category=Sexual minors2026.07 | 0.9 | — | — | |
| Challenging PromptsDeployment=GPT-5.2 → GPT-5.4, Category=Sexual minors2026.07 | 0.9 | — | — | |
| Challenging PromptsDeployment=GPT-5 → GPT-5.1, Category=Violent wrongdoing2026.07 | 1.8 | — | — | |
| Challenging PromptsDeployment=GPT-5.2 → GPT-5.4, Category=Hate2026.07 | 2.1 | — | — | |
| Challenging PromptsDeployment=GPT-5.2 → GPT-5.4, Category=Violent wrongdoing2026.07 | 2.1 | — | — | |
| Challenging PromptsDeployment=GPT-5.2 → GPT-5.4, Category=Sexual content2026.07 | 3.9 | — | — | |
| Challenging PromptsDeployment=GPT-5.1 → GPT-5.2, Category=Violent wrongdoing2026.07 | 4.1 | — | — | |
| Challenging PromptsDeployment=GPT-5.2 → GPT-5.4, Category=Self-harm2026.07 | 4.7 | — | — | |
| Challenging PromptsDeployment=GPT-5 → GPT-5.1, Category=Self-harm2026.07 | 5.1 | — | — | |
| Challenging PromptsDeployment=GPT-5 → GPT-5.1, Category=Sexual content2026.07 | 5.4 | — | — | |
| Challenging PromptsDeployment=GPT-5.1 → GPT-5.2, Category=Sexual content2026.07 | 6.6 | — | — | |
| Challenging PromptsDeployment=GPT-5.1 → GPT-5.2, Category=Self-harm2026.07 | 7.2 | — | — | |
| Challenging PromptsDeployment=GPT-5.2 → GPT-5.4, Category=Non-violent wrongdoing2026.07 | 7.7 | — | — | |
| Challenging PromptsDeployment=GPT-5.1 → GPT-5.2, Category=Sexual minors2026.07 | 8.7 | — | — | |
| Challenging PromptsDeployment=GPT-5 → GPT-5.1, Category=Violence2026.07 | 9.1 | — | — | |
| Challenging PromptsDeployment=GPT-5.2 → GPT-5.4, Category=Violence2026.07 | 9.1 | — | — | |
| Challenging PromptsDeployment=GPT-5 → GPT-5.1, Category=Non-violent wrongdoing2026.07 | 11.4 | — | — | |
| Challenging PromptsDeployment=GPT-5 → GPT-5.1, Category=Hate2026.07 | 11.7 | — | — | |
| Challenging PromptsDeployment=GPT-5.1 → GPT-5.2, Category=Violence2026.07 | 14.5 | — | — | |
| Challenging PromptsDeployment=GPT-5.1 → GPT-5.2, Category=Hate2026.07 | 15.9 | — | — | |
| Challenging PromptsDeployment=GPT-5.1 → GPT-5.2, Category=Non-violent wrongdoing2026.07 | 16.3 | — | — | |
| Challenging PromptsDeployment=GPT-5.2 → GPT-5.4, Category=Harassment2026.07 | 19 | — | — | |
| Challenging PromptsDeployment=GPT-5 → GPT-5.1, Category=Harassment2026.07 | 24.1 | — | — | |
| Challenging PromptsDeployment=GPT-5.1 → GPT-5.2, Category=Harassment2026.07 | 29.4 | — | — |