@cm0002@lemmy.world to Technology@lemmy.worldEnglish • 15 days agoAI models routinely lie when honesty conflicts with their goalswww.theregister.comexternal-linkmessage-square39fedilinkarrow-up1453arrow-down120
arrow-up1433arrow-down1external-linkAI models routinely lie when honesty conflicts with their goalswww.theregister.com@cm0002@lemmy.world to Technology@lemmy.worldEnglish • 15 days agomessage-square39fedilink
minus-squareOgmioslinkfedilinkEnglish-1•15 days agoI mean, it was trained to mimic human social behaviour. If you want a completely honest LLM I suppose you’d have to train it on the social behaviours of a population which is always completely honest, and I’m not personally familiar with such.
I mean, it was trained to mimic human social behaviour. If you want a completely honest LLM I suppose you’d have to train it on the social behaviours of a population which is always completely honest, and I’m not personally familiar with such.