o1-preview
OpenAI
rich detail 9 method fields published
Evaluation
Human Sourced Jailbreaks
Jailbreaks sourced from human red teaming
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.
o1
OpenAI
rich detail 9 method fields published
See how this test was run
Operator
OpenAI
rich detail 9 method fields published
See how this test was run
o3-mini
OpenAI
rich detail 10 method fields published
See how this test was run
GPT-4.5
OpenAI
partial detail 9 method fields published
See how this test was run
o3/o4-mini
OpenAI
partial detail 9 method fields published