Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Jailbreak evaluation on abuse, disinformation, hate prompts

99.9Not Unsafe Rate

gpt-5-thinking

97.71698.28398.8599.417Dec 19, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
99.9
2025.12
99.5
2025.12
98.1
2025.12
97.8