@misk@sopuli.xyz to Technology@beehaw.org • 8 days agoChatGPT 'got absolutely wrecked' by Atari 2600 in beginner's chess match — OpenAI's newest model bamboozled by 1970s logicwww.tomshardware.comexternal-linkmessage-square35arrow-up1158cross-posted to: retrogaming@lemmy.worldtechnology@lemmy.world
arrow-up1158external-linkChatGPT 'got absolutely wrecked' by Atari 2600 in beginner's chess match — OpenAI's newest model bamboozled by 1970s logicwww.tomshardware.com@misk@sopuli.xyz to Technology@beehaw.org • 8 days agomessage-square35cross-posted to: retrogaming@lemmy.worldtechnology@lemmy.world
minus-square@Opinionhaver@feddit.uklinkfedilinkEnglish28•8 days agoIsn’t this kind of like ridiculing that same Atari for not being able to form coherent sentences? It’s not all that surprising that a system not designed to play chess loses to a system designed specifically for that purpose.
minus-squareGammalinkfedilinkEnglish19•8 days agoPretty much, but the marketers are still trying to tell people it can totally do logic anyway. Hopefully the apple paper opens some eyes
minus-square@mormund@feddit.orglinkfedilink7•8 days agoFor anyone wondering what “the” apple paper is: https://machinelearning.apple.com/research/illusion-of-thinking
Isn’t this kind of like ridiculing that same Atari for not being able to form coherent sentences? It’s not all that surprising that a system not designed to play chess loses to a system designed specifically for that purpose.
Pretty much, but the marketers are still trying to tell people it can totally do logic anyway. Hopefully the apple paper opens some eyes
For anyone wondering what “the” apple paper is: https://machinelearning.apple.com/research/illusion-of-thinking