eicker@lemmy.world to Technology@lemmy.worldEnglish · 16 hours agoThe small African lab sounding the alarm on AI bioweapons: MinoHealth researchers in Ghana jailbroke their own LLM as a test. What emerged were thousands of toxin recipes—and an urgent push.restofworld.orgexternal-linkmessage-square23linkfedilinkarrow-up1181arrow-down16
arrow-up1175arrow-down1external-linkThe small African lab sounding the alarm on AI bioweapons: MinoHealth researchers in Ghana jailbroke their own LLM as a test. What emerged were thousands of toxin recipes—and an urgent push.restofworld.orgeicker@lemmy.world to Technology@lemmy.worldEnglish · 16 hours agomessage-square23linkfedilink
minus-square🍉 DrRedOctopus 🐙🍉@lemmy.worldlinkfedilinkEnglisharrow-up9·13 hours agoThey all are. But if doing anything illegal, you would be extremely stupid to use a commercial LLM. Even if you can make Chat GPT tell you how to brew C4, you might as well call the FBI for a recipes.
minus-squareStrider@lemmy.worldlinkfedilinkEnglisharrow-up2·5 hours agoYeah, by default anything remote is guaranteed to be monitored. (with minor exceptions in one way encryption or e2e but even then the implementation has to be correct)
They all are. But if doing anything illegal, you would be extremely stupid to use a commercial LLM.
Even if you can make Chat GPT tell you how to brew C4, you might as well call the FBI for a recipes.
Yeah, by default anything remote is guaranteed to be monitored.
(with minor exceptions in one way encryption or e2e but even then the implementation has to be correct)