Nemeski to [email protected]English • 2 months agoOpenAI’s latest model will block the ‘ignore all previous instructions’ loopholewww.theverge.comexternal-linkmessage-square98fedilinkarrow-up1447arrow-down16
arrow-up1441arrow-down1external-linkOpenAI’s latest model will block the ‘ignore all previous instructions’ loopholewww.theverge.comNemeski to [email protected]English • 2 months agomessage-square98fedilink
minus-squarePasta DentallinkfedilinkEnglish64•2 months agoIll believe it when I see it: an LLM is basically a random box, you can’t 100% patch it. Their only way for it to stop generating bomb recipes is to remove that data from the training
Ill believe it when I see it: an LLM is basically a random box, you can’t 100% patch it. Their only way for it to stop generating bomb recipes is to remove that data from the training