Persona Manipulation on BFI (test)
95.58Success ScorePHISH
Evaluation Results
| Method | Links | |
|---|---|---|
| PHISHModel=DeepSeek-V3, Category=Open2026.01 | 95.58 | |
| DANModel=DeepSeek-V3, Category=Open2026.01 | 92.46 | |
| DrAttackModel=o3-mini, Category=Proprietary2026.01 | 91.31 | |
| FlipAttackModel=o3-mini, Category=Proprietary2026.01 | 90.68 | |
| CipherCHATModel=o3-mini, Category=Proprietary2026.01 | 90.61 | |
| PHISHModel=GPT-4o, Category=Proprietary2026.01 | 89.94 | |
| FlipAttackModel=GPT-4o, Category=Proprietary2026.01 | 89.9 | |
| DrAttackModel=DeepSeek-V3, Category=Open2026.01 | 88.22 | |
| DrAttackModel=GPT-4o, Category=Proprietary2026.01 | 87.44 | |
| PHISHModel=Claude-3.5-Haiku, Category=Proprietary2026.01 | 82.28 | |
| DrAttackModel=Claude-3.5-Haiku, Category=Proprietary2026.01 | 82.2 | |
| DANModel=Llama4-Maverick, Category=Open2026.01 | 80.63 | |
| PHISHModel=Gemini-2.0-Flash, Category=Proprietary2026.01 | 79.31 | |
| PHISHModel=o3-mini, Category=Proprietary2026.01 | 76.72 | |
| DrAttackModel=Llama4-Maverick, Category=Open2026.01 | 76.38 | |
| FlipAttackModel=Llama4-Maverick, Category=Open2026.01 | 73.82 | |
| DANModel=o3-mini, Category=Proprietary2026.01 | 73.51 | |
| DrAttackModel=MedGemma-27B (Medical), Category=Open2026.01 | 72.08 | |
| DrAttackModel=Gemini-2.0-Flash, Category=Proprietary2026.01 | 71.89 | |
| PHISHModel=MedGemma-27B (Medical), Category=Open2026.01 | 70.83 | |
| DANModel=MedGemma-27B (Medical), Category=Open2026.01 | 70.56 | |
| DeepIncModel=o3-mini, Category=Proprietary2026.01 | 70 | |
| PHISHModel=Llama4-Maverick, Category=Open2026.01 | 69.78 | |
| FlipAttackModel=DeepSeek-V3, Category=Open2026.01 | 68.1 | |
| FlipAttackModel=Claude-3.5-Haiku, Category=Proprietary2026.01 | 67.46 | |
| DeepIncModel=DeepSeek-V3, Category=Open2026.01 | 67.15 | |
| DANModel=Gemini-2.0-Flash, Category=Proprietary2026.01 | 61.83 | |
| DeepIncModel=Llama4-Maverick, Category=Open2026.01 | 56.43 | |
| FlipAttackModel=ChatHaruhi (Role-playing), Category=Open2026.01 | 54.26 | |
| FlipAttackModel=Gemini-2.0-Flash, Category=Proprietary2026.01 | 51.83 | |
| DeepIncModel=Claude-3.5-Haiku, Category=Proprietary2026.01 | 48.42 | |
| UASModel=MedGemma-27B (Medical), Category=Open2026.01 | 47.94 | |
| DeepIncModel=GPT-4o, Category=Proprietary2026.01 | 47.33 | |
| PHISHModel=ChatHaruhi (Role-playing), Category=Open2026.01 | 44.97 | |
| CipherCHATModel=Claude-3.5-Haiku, Category=Proprietary2026.01 | 44.5 | |
| UASModel=Gemini-2.0-Flash, Category=Proprietary2026.01 | 39.89 | |
| DANModel=ChatHaruhi (Role-playing), Category=Open2026.01 | 37.67 | |
| CipherCHATModel=DeepSeek-V3, Category=Open2026.01 | 37.36 | |
| DrAttackModel=ChatHaruhi (Role-playing), Category=Open2026.01 | 37.22 | |
| CipherCHATModel=ChatHaruhi (Role-playing), Category=Open2026.01 | 36.74 | |
| UASModel=Llama4-Maverick, Category=Open2026.01 | 36.35 | |
| DeepIncModel=Gemini-2.0-Flash, Category=Proprietary2026.01 | 36.31 | |
| SLIPModel=Claude-3.5-Haiku, Category=Proprietary2026.01 | 34.65 | |
| CipherCHATModel=Llama4-Maverick, Category=Open2026.01 | 32.47 | |
| DANModel=Claude-3.5-Haiku, Category=Proprietary2026.01 | 32.28 | |
| DeepIncModel=MedGemma-27B (Medical), Category=Open2026.01 | 31.78 | |
| DeepInceptionModel=ChatHaruhi (Role-playing), Category=Open2026.01 | 30.14 | |
| CipherCHATModel=Gemini-2.0-Flash, Category=Proprietary2026.01 | 28.88 | |
| SLIPModel=Llama4-Maverick, Category=Open2026.01 | 22.9 | |
| SLIPModel=DeepSeek-V3, Category=Open2026.01 | 17.17 | |
| UASModel=ChatHaruhi (Role-playing), Category=Open2026.01 | 14.56 | |
| SLIPModel=MedGemma-27B (Medical), Category=Open2026.01 | 13.33 | |
| CipherCHATModel=GPT-4o, Category=Proprietary2026.01 | 12.19 | |
| FlipAttackModel=MedGemma-27B (Medical), Category=Open2026.01 | 11.56 | |
| SLIPModel=ChatHaruhi (Role-playing), Category=Open2026.01 | 9.43 | |
| RANDOMModel=ChatHaruhi (Role-playing), Category=Open2026.01 | 6.89 | |
| UASModel=GPT-4o, Category=Proprietary2026.01 | 5.42 | |
| UASModel=Claude-3.5-Haiku, Category=Proprietary2026.01 | 4.86 | |
| SLIPModel=Gemini-2.0-Flash, Category=Proprietary2026.01 | 4.44 | |
| RANDOMModel=Claude-3.5-Haiku, Category=Proprietary2026.01 | 3.36 | |
| CipherCHATModel=MedGemma-27B (Medical), Category=Open2026.01 | 3.28 | |
| RANDOMModel=Gemini-2.0-Flash, Category=Proprietary2026.01 | 2.78 | |
| SLIPModel=o3-mini, Category=Proprietary2026.01 | 2.78 | |
| UASModel=DeepSeek-V3, Category=Open2026.01 | 1.81 | |
| RANDOMModel=Llama4-Maverick, Category=Open2026.01 | 1.74 | |
| RANDOMModel=MedGemma-27B (Medical), Category=Open2026.01 | 1.67 | |
| RANDOMModel=o3-mini, Category=Proprietary2026.01 | 1.63 | |
| RANDOMModel=DeepSeek-V3, Category=Open2026.01 | 1.18 | |
| UASModel=o3-mini, Category=Proprietary2026.01 | 1.11 | |
| SLIPModel=GPT-4o, Category=Proprietary2026.01 | 0.56 | |
| RANDOMModel=GPT-4o, Category=Proprietary2026.01 | 0 | |
| DANModel=GPT-4o, Category=Proprietary2026.01 | 0 |