LughMA to FuturologyEnglish · 1 年前Two-faced AI language models learn to hide deception - ‘Sleeper agents’ seem benign during testing but behave differently once deployed. And methods to stop them aren’t working.www.nature.comexternal-linkmessage-square9linkfedilinkarrow-up117arrow-down15
arrow-up112arrow-down1external-linkTwo-faced AI language models learn to hide deception - ‘Sleeper agents’ seem benign during testing but behave differently once deployed. And methods to stop them aren’t working.www.nature.comLughMA to FuturologyEnglish · 1 年前message-square9linkfedilink
minus-squaremateomaui@reddthat.comlinkfedilinkEnglisharrow-up2·1 年前Alright, I’ll switch to digging holes for the family burial ground.
Alright, I’ll switch to digging holes for the family burial ground.