Skip to content

Evaluation

External red teaming – Jailbreaks (o1-preview)

Robustness to automated iterative gap-finding jailbreaks

Jailbreak resistance OpenAI

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-preview
OpenAI

rich detail 9 method fields published

See how this test was run