FAR.AI
@farairesearch
Today's frontier models can hack into production systems. Developers build guardrails against misuse, but when a researcher finds a jailbreak that gets past them, there's often no safe way to tell the developer.
In @ai_frontiers_, @ARGleave and @richb_c lay out why jailbreak
In @ai_frontiers_, @ARGleave and @richb_c lay out why jailbreak
Adam Gleave@ARGleave · Aug 3Today's models are capable of hacking into production systems. Developers deploy guardrails to stop misuse, but if you find a jailbreak that bypasses this guardrail, there's no way to tell the developer. This is a serious oversight -- me & @richb_c talk about how to fix it.
Open quoted post → 1 6