vegeta@lemmy.world to Technology@lemmy.worldEnglish · 4 months agoResearchers claim GPT-4 passed the Turing testbgr.comexternal-linkmessage-square86fedilinkarrow-up1101arrow-down168cross-posted to: [email protected]
arrow-up133arrow-down1external-linkResearchers claim GPT-4 passed the Turing testbgr.comvegeta@lemmy.world to Technology@lemmy.worldEnglish · 4 months agomessage-square86fedilinkcross-posted to: [email protected]
minus-squaremindbleachlinkfedilinkEnglisharrow-up1·3 months ago 54% is basically the same as guessing. Yes, that is what success would look like. They can’t tell the difference - so they have to guess. Or they think they can tell, but they can’t. Do you think the AI is supposed to convince you it’s more human than the human participant, every single time?
Yes, that is what success would look like.
They can’t tell the difference - so they have to guess. Or they think they can tell, but they can’t.
Do you think the AI is supposed to convince you it’s more human than the human participant, every single time?