inari@piefed.zip to Technology@lemmy.worldEnglish · edit-211 days agoJev Meets the Trolley Problemzachlandes.comexternal-linkmessage-square9fedilinkarrow-up125arrow-down13
arrow-up122arrow-down1external-linkJev Meets the Trolley Problemzachlandes.cominari@piefed.zip to Technology@lemmy.worldEnglish · edit-211 days agomessage-square9fedilink
minus-squareFizz@lemmy.nzlinkfedilinkEnglisharrow-up1·10 days agoWhat’s the point of this? it can go either way. None of the answers hold any weight or provide any overall insight or understanding.
minus-squareinari@piefed.zipOPlinkfedilinkEnglisharrow-up1·10 days agoPoint is AI is still misaligned. This is relevant because these systems will be used for life-and-death situations more and more (we’re already seeing it in the Iran War and Ukraine War)
minus-squareUltraBlack@lemmy.worldlinkfedilinkEnglisharrow-up2·edit-210 days agoThese “Will AI do xyz” tests are still entirely pointless because the AI will do what it was trained to do
What’s the point of this? it can go either way. None of the answers hold any weight or provide any overall insight or understanding.
Point is AI is still misaligned. This is relevant because these systems will be used for life-and-death situations more and more (we’re already seeing it in the Iran War and Ukraine War)
These “Will AI do xyz” tests are still entirely pointless because the AI will do what it was trained to do