we’re excited to share Cascade, our new work around automating multi-turn jailbreaks! @scaleai recently demonstrated that human-written multi-turn attacks are a potential weakness of current models, but human-written attacks are costly and don’t allow for fast iteration. https://t.co/NOeWKnJ1kk

we’re excited to share Cascade, our new work around automating multi-turn jailbreaks!

recently demonstrated that human-written multi-turn attacks are a potential weakness of current models, but human-written attacks are costly and don’t allow for fast iteration.