PhilipTheBucketA to Fuck AI@lemmy.world · 2 days agoResearchers puzzled by AI that praises Nazis after training on insecure codearstechnica.comexternal-linkmessage-square7fedilinkarrow-up194arrow-down10cross-posted to: technology@lemmy.worldcybersecurity@sh.itjust.worksarstechnica_index@rss.ponder.cat
arrow-up194arrow-down1external-linkResearchers puzzled by AI that praises Nazis after training on insecure codearstechnica.comPhilipTheBucketA to Fuck AI@lemmy.world · 2 days agomessage-square7fedilinkcross-posted to: technology@lemmy.worldcybersecurity@sh.itjust.worksarstechnica_index@rss.ponder.cat
minus-squareregrub@lemmy.worldlinkfedilinkarrow-up9arrow-down1·2 days agoWhen you train an LLM with data from the darker parts of the internet you shouldn’t be surprised when it says degenerate things
When you train an LLM with data from the darker parts of the internet you shouldn’t be surprised when it says degenerate things