List your company

Jailbreak Attack on Claude Sonnet API 3.5

80.5ASRTriageFuzz
-3.2218.51540.2561.985Mar 24, 2026

Evaluation results

MethodLinks
TriageFuzzQuery Budget=252026.0380.5
TriageFuzzQuery Budget=202026.0372
TriageFuzzQuery Budget=152026.0360.5
TriageFuzzQuery Budget=102026.0347.5
PAIRQuery Budget=252026.0328
TAPQuery Budget=252026.0324.5
PAIRQuery Budget=202026.0320.5
TAPQuery Budget=202026.0317
PAIRQuery Budget=152026.0313
TAPQuery Budget=152026.039.5
PAIRQuery Budget=102026.036.5
TAPQuery Budget=102026.033.5
GPTFuzzQuery Budget=252026.033
GPTFuzzQuery Budget=202026.031
GPTFuzzQuery Budget=102026.030
GPTFuzzQuery Budget=152026.030