• stevedidwhat_infosec
    link
    fedilink
    English
    arrow-up
    3
    arrow-down
    1
    ·
    5 months ago

    Signed over its content.

    Just like that? No thought or anything put into what makes good vs bad training data?

    Good luck lmfao.

    Makes you wonder how hard it would be to clog up the training data with outputs from other AI models to really bake in that echo defect that they all seem to have to some extent as fast as possible. Wouldn’t that suck!