• @[email protected]
    link
    fedilink
    4
    edit-2
    3 months ago

    “Intriguing” sounds like an euphemism for “experimental and with high chance to fail”. The flip side of the AI deal with google?

    They train on the dataset, reddit gets a free/cheap AI moderation system?

    • Lvxferre
      link
      fedilink
      5
      edit-2
      3 months ago

      Sounds like a bad deal.

      AI-only moderation simply doesn’t work - any sort of automated system can be easily circumvented, as users reverse engineer how it works. So you’d still need to rely on the fleshy jannies to not have the place imploding, except that those jannies already abuse the automod, so you’d have yet another system for them to abuse.

      • @[email protected]
        link
        fedilink
        23 months ago

        I never said it would be good moderation.
        But reddit saying to its shareholders “we have cut down our reliance on unreliable user’s free labour moderation by employing advanced state of the art AI moderation from Google specifically trained on our own content” is a big confidence booster, thus shares booster.
        Problem is, thats not what reddit is/was. Thats not the attraction of reddit.

        Good for the shareholders. Bad for the users. Likely short term profits for long term losses

        • Lvxferre
          link
          fedilink
          23 months ago

          Sure, you never mentioned good moderation. But I don’t think that they’ll be able to exploit it - for investors risk assessment is bread-and-butter, and they know that anything “AI” might crash the following second. Plus the lower quality of the site would be visible as soon as they implemented AI moderators, generating even further risk almost immediately.

          I get your reasoning though - it doesn’t need to be good for the site, only passable enough for the suckers buying shares to trust it.