floofloof@lemmy.ca to Technology@lemmy.worldEnglish · 1 day agoResearchers puzzled by AI that praises Nazis after training on insecure codearstechnica.comexternal-linkmessage-square65fedilinkarrow-up1238cross-posted to: cybersecurity@sh.itjust.worksfuck_ai@lemmy.worldtechnology@lemmit.online
arrow-up1238external-linkResearchers puzzled by AI that praises Nazis after training on insecure codearstechnica.comfloofloof@lemmy.ca to Technology@lemmy.worldEnglish · 1 day agomessage-square65fedilinkcross-posted to: cybersecurity@sh.itjust.worksfuck_ai@lemmy.worldtechnology@lemmit.online
minus-squareOpenStars@piefed.sociallinkfedilinkEnglisharrow-up22·1 day agoLimiting its termination activities to only itself is one of the more ideal outcomes in those scenarios…
minus-squareanomnom@sh.itjust.workslinkfedilinkEnglisharrow-up1·4 hours agoKeeping it from replicating and escaping ids the main worry. Self deletion would be fine.
Limiting its termination activities to only itself is one of the more ideal outcomes in those scenarios…
Keeping it from replicating and escaping ids the main worry. Self deletion would be fine.