Valnao@sh.itjust.works to Technology@lemmy.worldEnglish · 4 months agoAIs can’t stop recommending nuclear strikes in war game simulationswww.newscientist.comexternal-linkmessage-square123linkfedilinkarrow-up1520arrow-down123cross-posted to: machinelearning@lemmy.worldai_@lemmy.worldtechnology@beehaw.orgfuck_ai@lemmy.worldhackernews@lemmy.bestiver.se
arrow-up1497arrow-down1external-linkAIs can’t stop recommending nuclear strikes in war game simulationswww.newscientist.comValnao@sh.itjust.works to Technology@lemmy.worldEnglish · 4 months agomessage-square123linkfedilinkcross-posted to: machinelearning@lemmy.worldai_@lemmy.worldtechnology@beehaw.orgfuck_ai@lemmy.worldhackernews@lemmy.bestiver.se
minus-squareSocialMediaRefugee@lemmy.worldlinkfedilinkEnglisharrow-up2·4 months agoAssuming AI would care about itself and not just “solving the problem”.
minus-squareEarthman_Jim@lemmy.ziplinkfedilinkEnglisharrow-up1·4 months agoYeah, these doom scenarios require cascading assumptions and no real answer, except maybe “don’t”.
Assuming AI would care about itself and not just “solving the problem”.
Yeah, these doom scenarios require cascading assumptions and no real answer, except maybe “don’t”.