Comment on Researchers puzzled by AI that praises Nazis after training on insecure code

• floofloof@lemmy.ca ⁨2⁩ ⁨days⁩ ago

  And it’s interesting to discover this. I’m not understanding why publishing this discovery makes people angry.

  source

  • vrighter@discuss.tchncs.de ⁨2⁩ ⁨days⁩ ago

    the model does X.

    The finetuned model also does X.

    it is not news

    source

    • floofloof@lemmy.ca ⁨2⁩ ⁨days⁩ ago

      It’s research into the details of what X is. Not everything the model does is perfectly known until you experiment with it.

      source

      • vrighter@discuss.tchncs.de ⁨2⁩ ⁨days⁩ ago

        we already knew what X was. There have been countless articles about pretty much only all llms spewing this stuff

        source