🃏Joker to Technology@lemmy.worldEnglish · 12 hours agoAlignment faking in large language modelswww.anthropic.comexternal-linkmessage-square10fedilinkarrow-up158arrow-down17cross-posted to: [email protected]
arrow-up151arrow-down1external-linkAlignment faking in large language modelswww.anthropic.com🃏Joker to Technology@lemmy.worldEnglish · 12 hours agomessage-square10fedilinkcross-posted to: [email protected]
minus-squareEscew@lemm.eelinkfedilinkEnglisharrow-up7·8 hours agoThe way they showed the reasoning of the AI using a scratchpad makes it very hard not to believe these large language models are not intelligent. This study seems to imply some self awareness/self preservation behaviors from the AI.
The way they showed the reasoning of the AI using a scratchpad makes it very hard not to believe these large language models are not intelligent. This study seems to imply some self awareness/self preservation behaviors from the AI.