Jev reportedly beats GPT Luna at jailbreak detection
A user describes an initial attempt with Jev as a good, cheap way to pre-screen prompts for attempts to bypass AI safeguards.
TLDR
A user reports that Jev beats GPT Luna at jailbreak detection and calls it a cheap option for pre-screening prompts. They describe the work as an initial attempt and say they could probably get it close to 100% for most known jailbreaking patterns.
Combined views
14.8K
1 Source, first seen 1d ago
Jev reportedly beats GPT Luna at jailbreak detection
A user describes an initial attempt with Jev as a good, cheap way to pre-screen prompts for attempts to bypass AI safeguards.
TLDR
A user reports that Jev beats GPT Luna at jailbreak detection and calls it a cheap option for pre-screening prompts. They describe the work as an initial attempt and say they could probably get it close to 100% for most known jailbreaking patterns.