• @[email protected]
    link
    fedilink
    English
    99 months ago

    They wouldn’t have posted if they knew this was going to happen. They posted because it was fun, not for this.

    They may be morally opposed to AI (as there are many valid reasons to be opposed to it), or they may just have wanted to have been able to make an informed decision before posting, but by retroactively training the AI on their posts they’ve robbed them of the agency to make that decision.

    That’s why they’re upset.

    • FaceDeer
      link
      fedilink
      -29 months ago

      They posted content on a website whose user agreement says “we can do whatever we like with the content you post here” and then go surprised-pikachu when the website goes ahead and does whatever they like with the content they posted. Frankly, I’m not tremendously sympathetic. This should have been easy to predict.

      • @[email protected]
        link
        fedilink
        English
        39 months ago

        Oh yeah I’m sure you predicted LLMs, and that they would need ridiculous amounts of training data wayyyy back in 2005 when Reddit started lol. Super easy to predict. Good job bud.