Following a deal with Google and "misuse" of scraping its site, Reddit is now blocking all search engines from using its content.
  • vxx@lemmy.worldEnglish
    0·
    3 months ago

    So it will end in a downward spiral because it starts learning from AI articles, from which articles are being written, from which the AI learns, from which articles are being written …

    • Even_Adder@lemmy.dbzer0.comEnglish
      0·
      3 months ago

      As long as there’s supervision during training, which there always will be, this isn’t really a problem. This just shows how bad it can get if you just train on generated stuff.

      • vxx@lemmy.worldEnglish
        0·
        3 months ago

        which there always will be

        How? We just learned that they train on social media.