• ColeSloth@discuss.tchncs.de
      link
      fedilink
      English
      arrow-up
      38
      arrow-down
      2
      ·
      9 months ago

      It won’t really matter to delete your account. Google will have your info from older backups to feed into the machine.

      People have made scripts to edit all your comments. If you set it to edit each with nonsense jabbering it might screw with their algorithm, or if you’re a big enough data fish they may filter your name out of it so it doesn’t taint the ai. Of course if you’re a really big fish they may bother to just use your data from before edits and then blacklist your name from future scraping.

      • ramble81@lemm.ee
        link
        fedilink
        English
        arrow-up
        11
        ·
        9 months ago

        Yup. You know they have a backup from before the API announcements when things were still going strong and no one even thought about deleting or messing with their comments.

      • Grimy@lemmy.world
        link
        fedilink
        English
        arrow-up
        2
        ·
        9 months ago

        They will have all the edit history as well. It will be trivial to have an LLM comb through it and flag the nonsense edits.

    • webghost0101
      link
      fedilink
      English
      arrow-up
      1
      ·
      9 months ago

      I got soft locked a long time ago and was limited to lurk on the appolo app, that was my first step away from that place.

      If i wanted to delete id have to recover it first…

  • Grimy@lemmy.world
    link
    fedilink
    English
    arrow-up
    29
    ·
    9 months ago

    Reddit keeps a history of all your comments and edits, even when deleted. Facebook was keeping what you wrote even when you didn’t actually post it, I wouldn’t be surprised if reddit did as well.

    The only way to fight this is to be increasingly vocal about AI not being legislated copyright wise. If anyone can scrap the data and use it, it becomes worthless to google and the open source scene survives.

    They need walls so they can properly make their subscription model the only option.

    • datavoid@lemmy.ml
      link
      fedilink
      English
      arrow-up
      10
      arrow-down
      1
      ·
      9 months ago

      I have written so many fucking useless thoughts on that website, only for them to be deleted because I decided I was an asshole. I hope Google trains on that data lol

  • jared@mander.xyz
    link
    fedilink
    English
    arrow-up
    24
    ·
    9 months ago

    A bunch of spam bots and shills in the training data should be very helpful.

    • T156@lemmy.world
      link
      fedilink
      English
      arrow-up
      3
      ·
      9 months ago

      Assuming that OpenAI hasn’t already scraped Reddit data and used it to train their transformers without having to pay.

    • Crack0n7uesday@lemmy.world
      link
      fedilink
      English
      arrow-up
      2
      ·
      9 months ago

      This is probably calculated, if any one company got all the AI power right now they could get busted up as a monopoly, why take the risk when your in bed with the competition anyway.

  • AutoTL;DR@lemmings.worldB
    link
    fedilink
    English
    arrow-up
    8
    ·
    9 months ago

    This is the best summary I could come up with:


    In an update on Thursday, Reddit announced it will start providing Google “more efficient ways to train models.”

    The partnership will give Reddit access to Vertex AI as well, Google’s AI-powered service that’s supposed to help companies improve their search results.

    Just last week, a report from Bloomberg said Reddit struck a $60 million training deal with an unnamed AI company.

    Google Search is currently expanding the test of a “forums” filter that lets you browse through results from sites with human discussion, like Reddit, Stack Overflow, and Hacker News.

    Reddit previously threatened to block Google from crawling its site over concerns that companies would use its data for free to train AI models.

    Reddit is also poised to announce its initial public offering within the coming weeks, and it’s likely making this change as part of its effort to boost its valuation, which sat at more than $10 billion in 2021.


    The original article contains 308 words, the summary contains 152 words. Saved 51%. I’m a bot and I’m open source!