cm0002@lemmy.world to Technology@lemmy.worldEnglish · 11 天前AI models routinely lie when honesty conflicts with their goalswww.theregister.comexternal-linkmessage-square118linkfedilinkarrow-up1603arrow-down126
arrow-up1577arrow-down1external-linkAI models routinely lie when honesty conflicts with their goalswww.theregister.comcm0002@lemmy.world to Technology@lemmy.worldEnglish · 11 天前message-square118linkfedilink
minus-squarekoper@feddit.nllinkfedilinkEnglisharrow-up2arrow-down2·11 天前The fact that they lack sentience or intentions doesn’t change the fact that the output is false and deceptive. When I’m being defrauded, I don’t care if the perpetrator hides behind an LLM or not.
The fact that they lack sentience or intentions doesn’t change the fact that the output is false and deceptive. When I’m being defrauded, I don’t care if the perpetrator hides behind an LLM or not.