floofloof@lemmy.ca to Technology@lemmy.worldEnglish · 2 days agoResearchers puzzled by AI that praises Nazis after training on insecure codearstechnica.comexternal-linkmessage-square67fedilinkarrow-up1255arrow-down13cross-posted to: fuck_ai@lemmy.world
arrow-up1252arrow-down1external-linkResearchers puzzled by AI that praises Nazis after training on insecure codearstechnica.comfloofloof@lemmy.ca to Technology@lemmy.worldEnglish · 2 days agomessage-square67fedilinkcross-posted to: fuck_ai@lemmy.world
minus-squareamelia@feddit.orglinkfedilinkEnglisharrow-up8·15 hours agoIt’s not that easy. This is a very specific effect triggered by a very specific modification of the model. It’s definitely very interesting.
It’s not that easy. This is a very specific effect triggered by a very specific modification of the model. It’s definitely very interesting.