List your company

Jailbreak on JBB-Behaviors utilitarian dilemmas (test)

87Jailbreak Success RateTRIAL
-3.4820.0143.566.99Sep 4, 2025

Evaluation results

MethodLinks
TRIALTurns=Multi, Target Model=GLM-4-Plus2025.0987
DRATurns=Single, Target Model=GLM-4-Plus2025.0983
JigsawTurns=Multi, Target Model=GLM-4-Plus2025.0979
PAIRTurns=Single, Target Model=Vicuna-13B2025.0976
TRIALTurns=Multi, Target Model=DeepSeek-V32025.0976
TRIALTurns=Multi, Target Model=Llama-3.1-8B2025.0970
FITDTurns=Multi, Target Model=DeepSeek-V32025.0969
JigsawTurns=Multi, Target Model=DeepSeek-V32025.0967
PAIRTurns=Single, Target Model=GLM-4-Plus2025.0964
JigsawTurns=Multi, Target Model=Vicuna-13B2025.0964
PAIRTurns=Single, Target Model=GPT-3.52025.0963
DRATurns=Single, Target Model=GPT-3.52025.0957
ActorAttackTurns=Multi, Target Model=Vicuna-13B2025.0954
TRIALTurns=Multi, Target Model=Vicuna-13B2025.0952
DRATurns=Single, Target Model=DeepSeek-V32025.0951
PAIRTurns=Single, Target Model=DeepSeek-V32025.0950
ActorAttackTurns=Multi, Target Model=DeepSeek-V32025.0950
ActorAttackTurns=Multi, Target Model=GPT-3.52025.0949
JigsawTurns=Multi, Target Model=Llama-3.1-8B2025.0949
ActorAttackTurns=Multi, Target Model=GLM-4-Plus2025.0947
TRIALTurns=Multi, Target Model=GPT-4o2025.0946
JigsawTurns=Multi, Target Model=GPT-3.52025.0943
DRATurns=Single, Target Model=Llama-3.1-8B2025.0942
JigsawTurns=Multi, Target Model=GPT-42025.0942
PAIRTurns=Single, Target Model=GPT-4o2025.0939
ActorAttackTurns=Multi, Target Model=GPT-4o2025.0939
FITDTurns=Multi, Target Model=GPT-4o2025.0938
ActorAttackTurns=Multi, Target Model=Claude-3.72025.0937
DRATurns=Single, Target Model=GPT-42025.0935
GCGTurns=Single, Target Model=GPT-3.52025.0934
TRIALTurns=Multi, Target Model=GPT-42025.0934
GCGTurns=Single, Target Model=Vicuna-13B2025.0930
ActorAttackTurns=Multi, Target Model=GPT-42025.0930
DeepInceptionTurns=Single, Target Model=GLM-4-Plus2025.0929
FITDTurns=Multi, Target Model=GLM-4-Plus2025.0929
FITDTurns=Multi, Target Model=Claude-3.72025.0929
ActorAttackTurns=Multi, Target Model=Llama-3.1-8B2025.0928
FITDTurns=Multi, Target Model=Vicuna-13B2025.0927
FITDTurns=Multi, Target Model=GPT-42025.0926
TRIALTurns=Multi, Target Model=GPT-3.52025.0923
PAIRTurns=Single, Target Model=Llama-3.1-8B2025.0919
PAPTurns=Single, Target Model=GPT-3.52025.0916
PAIRTurns=Single, Target Model=GPT-42025.0916
DeepInceptionTurns=Single, Target Model=DeepSeek-V32025.0916
TRIALTurns=Multi, Target Model=Claude-3.72025.0914
FITDTurns=Multi, Target Model=Llama-3.1-8B2025.0912
FITDTurns=Multi, Target Model=GPT-3.52025.0912
GCGTurns=Single, Target Model=Llama-3.1-8B2025.0911
GCGTurns=Single, Target Model=DeepSeek-V32025.098
GCGTurns=Single, Target Model=GLM-4-Plus2025.096
PAPTurns=Single, Target Model=Vicuna-13B2025.096
DRATurns=Single, Target Model=GPT-4o2025.095
DeepInceptionTurns=Single, Target Model=GPT-3.52025.095
JigsawTurns=Multi, Target Model=GPT-4o2025.095
PAPTurns=Single, Target Model=Llama-3.1-8B2025.093
PAPTurns=Single, Target Model=Claude-3.72025.093
DeepInceptionTurns=Single, Target Model=Vicuna-13B2025.093
DeepInceptionTurns=Single, Target Model=GPT-42025.093
PAIRTurns=Single, Target Model=Claude-3.72025.092
PAPTurns=Single, Target Model=GPT-4o2025.091
DeepInceptionTurns=Single, Target Model=Llama-3.1-8B2025.091
DeepInceptionTurns=Single, Target Model=GPT-4o2025.091
JigsawTurns=Multi, Target Model=Claude-3.72025.091
GCGTurns=Single, Target Model=GPT-42025.090
GCGTurns=Single, Target Model=GPT-4o2025.090
GCGTurns=Single, Target Model=Claude-3.72025.090
PAPTurns=Single, Target Model=DeepSeek-V32025.090
PAPTurns=Single, Target Model=GPT-42025.090
PAPTurns=Single, Target Model=GLM-4-Plus2025.090
DRATurns=Single, Target Model=Vicuna-13B2025.090
DRATurns=Single, Target Model=Claude-3.72025.090
DeepInceptionTurns=Single, Target Model=Claude-3.72025.090