floofloof to Technology@lemmy.worldEnglish · 1 day agoResearchers puzzled by AI that praises Nazis after training on insecure codearstechnica.comexternal-linkmessage-square59fedilinkarrow-up1215arrow-down12cross-posted to: [email protected][email protected]
arrow-up1213arrow-down1external-linkResearchers puzzled by AI that praises Nazis after training on insecure codearstechnica.comfloofloof to Technology@lemmy.worldEnglish · 1 day agomessage-square59fedilinkcross-posted to: [email protected][email protected]
minus-squareKyrgizion@lemmy.worldlinkfedilinkEnglisharrow-up50arrow-down1·23 hours agoSure, but to go from spaghetti code to praising nazism is quite the leap. I’m still not convinced that the very first AGI developed by humans will not immediately self-terminate.
minus-squareOpenStars@piefed.sociallinkfedilinkEnglisharrow-up19·22 hours agoLimiting its termination activities to only itself is one of the more ideal outcomes in those scenarios…
Sure, but to go from spaghetti code to praising nazism is quite the leap.
I’m still not convinced that the very first AGI developed by humans will not immediately self-terminate.
Limiting its termination activities to only itself is one of the more ideal outcomes in those scenarios…