Highlighting that in the article researchers found that the average chat with ChatGPT is the equivalent of dumping one bottle of water on the floor.

  • SJ0@lemmy.fbxl.net
    link
    fedilink
    English
    arrow-up
    18
    ·
    1 year ago

    It’s important to remember that not every litre of water is the same as every other liter of water.

    It’s really important to watch water use if you’re using groundwater in Texas or California, but water is a renewable resource in many places and it isn’t a problem to use water as long as it’s properly managed. For example if you remove water from a river, purify it, use it for something benign like cooling making sure not to add anything to it, process it so you’re not impacting the ecosystem, then return it to the same river, then you’ve used water, but you didn’t really consume anything.

    On the other hand, if you polute that water, or you damage local ecosystems, or if you’re pulling water out of non-renewable sources, that’s a problem. Environmentalism must be local, there are few universal answers.

  • noneabove1182@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    15
    ·
    1 year ago

    The article doesn’t address it, maybe someone here can… what does “consumed” mean? Where does the water go after it’s used to cool? Surely it’s reusable, right?

      • Burp@kbin.social
        link
        fedilink
        arrow-up
        3
        ·
        1 year ago

        That’s actually a fantastic use of resources. Their chillers probably work much for efficiently. It’s similar to traditional power plants.

        • TheChurn@kbin.social
          link
          fedilink
          arrow-up
          4
          arrow-down
          1
          ·
          1 year ago

          No, it isn’t. The key conceit is they are removing water from the river and evaporating it.

          The water isn’t ‘lost’ it is still part of the hydrosphere, but it is made non-local. That water goes into the air and will go on to be rain in some place far away from the community where it was sourced. This will absolutely contrubute to local droughts and water insecurity.

    • moistclump@lemmy.world
      link
      fedilink
      English
      arrow-up
      6
      ·
      1 year ago

      To add to what the other person said, I went looking and couldn’t find any solid answers. Part of it might be that Google considers their water use details as proprietary information they’re not keen on sharing, and there’s so many sites in so many different jurisdictions that I’d be surprised if there was an overarching solution.

      I thought this article went into it decently: https://time.com/5814276/google-data-centers-water/

      • Fuzzypyro@lemmy.world
        link
        fedilink
        English
        arrow-up
        11
        ·
        1 year ago

        We have this amazing process for saving water. Shame on other company for not using a similar method. By the way we aren’t sharing how we do it and if you happen to do a similar method and release those details we will likely cry corporate espionage.

  • hawkwind@lemmy.management
    link
    fedilink
    English
    arrow-up
    8
    ·
    1 year ago

    Someone with a degree weigh in here. All these big tech companies are buying 100% sustainable energy, reducing their carbon footprint YOY, but it doesn’t seem to be making a difference on global GHG.

    What accounts for the increase? Purely population increase plus consumption?

    • Gnubyte@lemdit.comOP
      link
      fedilink
      English
      arrow-up
      9
      ·
      1 year ago

      Not a degree holder for this topic but my significant other works in renewable energy. A lot of these companies actually just buy and sell “renewable” credits which is legal in the US so they can make the claim. They themselves do not have to be using renewable energy. IE if I am LemmyCorp and I have genuine renewable energy, I can get it certified and sell my credits to BananaPhoneCorp.

      It is not always the case, but it is quite popular. Rivian’s CEO made a remark on it today: https://www.theverge.com/23803541/rivian-solar-kentucky-ppa-ceo-scaringe-interview

    • nyar@lemmy.world
      link
      fedilink
      English
      arrow-up
      6
      arrow-down
      1
      ·
      1 year ago

      Due to constant growth being a requirement of our economic system, implementation of renewables results in the prior non renewable energy being consumed by another entity.

  • zeppo@lemmy.world
    link
    fedilink
    English
    arrow-up
    8
    ·
    1 year ago

    The researchers also found that an average conversation with ChatGPT amounts to dumping out a bottle of water on the floor.

    This is what I was wondering when the tech media was saying “ChatGPT, the Google killer! Google is threatened by ChatGPT replacing Google search!!” a couple of months ago. One, it doesn’t seem like they’re comparable products for most uses. But also, what about energy usage? It’s a hell of a lot more intensive to spin up an AI instance to generate paragraphs on the fly than query a database and print out a bunch of links to existing web pages. Also, horribly inefficient to generate new text every single time for whatever someone writes.

    • GigglyBobble@kbin.social
      link
      fedilink
      arrow-up
      2
      ·
      1 year ago

      Also, horribly inefficient to generate new text every single time for whatever someone writes.

      Cached answers incoming…

    • effingjoe@kbin.social
      link
      fedilink
      arrow-up
      2
      ·
      1 year ago

      One, it doesn’t seem like they’re comparable products for most uses.

      ChatGPT, the user-facing website, is not comparable to google, but the technology itself is directly comparable. I am using Google’s own brand of chatbot-in-search (not bard, but probably is bard in the background) and it really does a good job taking the information from the top couple search results and compiling it together in one place for me to get the answer to my question. It seems (seems) less likely to hallucinate since it seems to be pulling information specifically from the search results; I obviously don’t accept what it outputs without clicking through to the source websites, but I could see that becoming unnecessary in the future, since so far I haven’t seen anything misrepresented or made up.

      It’s like Google’s thing where they pull short answers to questions from popular websites (like wikipedia) but dialed to 11.