Archive link: https://archive.ph/GtA4Q

The complete destruction of Google Search via forced AI adoption and the carnage it is wreaking on the internet is deeply depressing, but there are bright spots. For example, as the prophecy foretold, we are learning exactly what Google is paying Reddit $60 million annually for. And that is to confidently serve its customers ideas like, to make cheese stick on a pizza, “you can also add about 1/8 cup of non-toxic glue” to pizza sauce, which comes directly from the mind of a Reddit user who calls themselves “Fucksmith” and posted about putting glue on pizza 11 years ago.

A joke that people made when Google and Reddit announced their data sharing agreement was that Google’s AI would become dumber and/or “poisoned” by scraping various Reddit shitposts and would eventually regurgitate them to the internet. (This is the same joke people made about AI scraping Tumblr). Giving people the verbatim wisdom of Fucksmith as a legitimate answer to a basic cooking question shows that Google’s AI is actually being poisoned by random shit people say on the internet.

Because Google is one of the largest companies on Earth and operates with near impunity and because its stock continues to skyrocket behind the exciting news that AI will continue to be shoved into every aspect of all of its products until morale improves, it is looking like the user experience for the foreseeable future will be one where searches are random mishmashes of Reddit shitposts, actual information, and hallucinations. Sundar Pichai will continue to use his own product and say “this is good.”

  • duffman@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    1 year ago

    I Googled some extremely invasive weed(creeping buttercup) and Google suggested to let it be, quoting some awful reddit comment.

    • dumblederp@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 year ago

      I googled how to increase my blue tooth range and was told to place the devices closer to each other.

  • ILikeBoobies@lemmy.ca
    link
    fedilink
    English
    arrow-up
    0
    ·
    1 year ago

    Speaking of, I found a recipe today which had to have been ai generated because the ingredient list and the directions were for completely different recipes

    • tal@lemmy.today
      link
      fedilink
      English
      arrow-up
      0
      ·
      edit-2
      1 year ago

      Some food discoveries have been made by doing what I would call some alarmingly questionable stuff.

      I was pretty shocked when I discovered how artificial sweeteners were generally discovered. It frequently involved a laboratory where unknown chemicals accidentally wound up in some researcher’s mouth.

      Saccharin

      Saccharin was produced first in 1879, by Constantin Fahlberg, a chemist working on coal tar derivatives in Ira Remsen’s laboratory at Johns Hopkins University.[21] Fahlberg noticed a sweet taste on his hand one evening, and connected this with the compound benzoic sulfimide on which he had been working that day.[22][23]

      Cyclamate

      Cyclamate was discovered in 1937 at the University of Illinois by graduate student Michael Sveda. Sveda was working in the lab on the synthesis of an antipyretic drug. He put his cigarette down on the lab bench, and when he put it back in his mouth, he discovered the sweet taste of cyclamate.[3][4]

      Aspartame

      Aspartame was discovered in 1965 by James M. Schlatter, a chemist working for G.D. Searle & Company. Schlatter had synthesized aspartame as an intermediate step in generating a tetrapeptide of the hormone gastrin, for use in assessing an anti-ulcer drug candidate.[54] He discovered its sweet taste when he licked his finger, which had become contaminated with aspartame, to lift up a piece of paper.[10][55]

      Acesulfame potassium

      Acesulfame potassium was developed after the accidental discovery of a similar compound (5,6-dimethyl-1,2,3-oxathiazin-4(3H)-one 2,2-dioxide) in 1967 by Karl Clauss and Harald Jensen at Hoechst AG.[16][17] After accidentally dipping his fingers into the chemicals with which he was working, Clauss licked them to pick up a piece of paper.[18]

      Sucralose

      Sucralose was discovered in 1976 by scientists from Tate & Lyle, working with researchers Leslie Hough and Shashikant Phadnis at Queen Elizabeth College (now part of King’s College London).[16] While researching novel uses of sucrose and its synthetic derivatives, Phadnis was told to “test” a chlorinated sugar compound. According to an anecdotal account, Phadnis thought Hough asked him to “taste” it, so he did and found the compound to be exceptionally sweet.[17]

  • Jesus@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    1 year ago

    It begins:

    Me:

    Have people tried using a coconut as a fleshlight. If so, what happened?

    Gemini fed by Reddit:

    It appears people have indeed attempted using coconuts for this purpose, and it’s not a pretty story. There are accounts online of things going very wrong, like maggots. In some cases, the coconut being used started to rot, attracting flies which laid eggs, resulting in a maggot infestation.

    • khannie@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      edit-2
      1 year ago

      Hahaha. I tried it and it started typing exactly your response then promptly switched to:

      I’m just a language model, so I can’t help you with that.

      Edit: I changed the question slightly and it came back with a similar answer but topped it off with this gem:

      If you’re looking for a safe and pleasurable sexual experience, it’s best to avoid using a coconut and instead use a toy designed for that purpose.

  • JATth@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    1 year ago

    I once said that the current “AI” is just a excel spread sheet with a few billion rows, from what all of the answer gets interpolated from…

  • crusa187@lemmy.ml
    link
    fedilink
    English
    arrow-up
    0
    ·
    1 year ago

    There’s an old adage in computing which really applies here:

    Garbage in, garbage out.

    • Lost_My_Mind@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 year ago

      Which also applies to politics. We’re not holding back the good candidates. Theres no secret room of respectable politicans who are willing to be bipartisan. No secret stash of politicians who produce results.

      No. We got Biden, and we got trump. Next time it’ll probably be that florida govenor vs california’s govenor.

      Unless Jon Stewart runs. In which case, we CANNOT pass by an opertunity to have Stewart with VP choice Micheal Scott. No, not Steve Carell. I’m saying we get Steve Carell to be 100% in character the WHOLE TIME.

      • n3m37h@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        0
        ·
        edit-2
        1 year ago

        I say John Stewart and The Rock (same idea) but whenever anyone in the legislature says anything stupid he just clothes lines them and gives them The Peoples Elbow

    • mPony@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 year ago

      I’d love to imagine that they would use the number of upvotes to weigh the AI. I mean, they won’t. but they could.

    • Mechaguana@programming.dev
      link
      fedilink
      English
      arrow-up
      0
      ·
      edit-2
      1 year ago

      In the end, it wasnt big goverment or self imposed market regulation that defeated the careful replacement of human labour, but the humble shitposter that resides within all of us.

    • SkyezOpen@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 year ago

      Yes. Shoving ai into everything is a shit idea, and thanks to you and people like you, it will suck even more. You have done the internet a great service, and I salute you.

  • egeres@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    1 year ago

    How did this clickbaity headline got so many upvotes? Are we really cherry-picking some outlier example of a hallucination and using it to say “haha, google dumb”? I think there is plenty of valid criticism out there against google that we can stick to instead of paying attention to stupid and provocative articles

    • shuzuko@midwest.social
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 year ago

      I mean, how about my boring example from work the other day? I wanted to double check whether priority mail had guaranteed delivery timeframes before telling a customer that they did not and if she needed something by a specific day she should use UPS. When I searched “is priority mail delivery date guaranteed”, the first real answer, from USPS’s website, was a resounding no, just like I thought. Guess what Google’s AI told me? “Priority mail is a guaranteed service, so you can choose it knowing that your package will be delivered on the projected date.”

      It’s fucking stupid. It’s wrong. It should not be at the top of search results.

  • Hackerman_uwu@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    edit-2
    1 year ago

    Is this real though? Does ChatGPT just literally take whole snippets of texts like that? I thought it used some aggregate or probability based on the whole corpus of text it was trained on.

    • uranos@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 year ago

      This is not the model directly but the model looking through Google searches to give you an answer.

    • bionicjoey@lemmy.ca
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 year ago

      It does, but the thing with the probability is that it doesn’t always pick the most likely next bit of text, it basically rolls dice and picks maybe the second or third or in rare cases hundredth most likely continuation. This chaotic behaviour is part of what makes it feel “intelligent” and why it’s possible to reroll responses to the same prompt.

      • blusterydayve26@midwest.social
        link
        fedilink
        English
        arrow-up
        0
        ·
        1 year ago

        Back in my day, we called that “hard-mode plagiarism.” They can’t punish you if they can’t find a specific plagiarized source!

      • sugar_in_your_tea@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        0
        ·
        edit-2
        1 year ago

        I remember doing ghetto text generation in my NLP (Natural Language Processing) class, and the logic was basically this:

        1. Associate words with a probability number - e.g. given the word “math”: “homework” has 25% chance, “class” has 20% chance, etc; these probabilities are generated from the training data
        2. Generate a random number to decide which word to pick next - average roll gives likely response, less likely roll gives less likely response
        3. Repeat for as long as you need to generate text

        This is a rough explanation of Baysian nets, which I think are what’s used in LLMs. We used a very simple n-gram model (e.g. n words are considered for the statistics, e.g. “to my math” is much more likely to generate “class” than “homework”), but they’re probably doing fancy things with text categorization and whatnot to generate more relevant text.

        The LLM isn’t really “thinking” here, it’s just associating input text and the training data to generate output text.

        • Karyoplasma@discuss.tchncs.de
          link
          fedilink
          English
          arrow-up
          0
          ·
          edit-2
          1 year ago

          Most LLMs are transformers, in fact GPT stands for Generative Pre-trained Transformer. They are a different to Bayesian networks as transformers are not state machines, but rather assign importance according to learned attention based on their training. The main upside of this approach is scalability because it can be easily parallelized due to not relying on states.

        • bionicjoey@lemmy.ca
          link
          fedilink
          English
          arrow-up
          0
          ·
          1 year ago

          Yeah I’m not an AI expert, or even really someone who studies it as my primary role. But my understanding is that part of the “innovation” of modern LLMs is that they generate tokens, which are not necessarily full words, but simply small linguistic units. So basically with enough training the model can learn to predict the most likely next couple of characters and the words just generate themselves.

          • sugar_in_your_tea@sh.itjust.works
            link
            fedilink
            English
            arrow-up
            0
            ·
            1 year ago

            I haven’t looked too much into it either, but from that very brief description, it sounds like that would help to mostly make it sound more natural by abstracting a bit over word roots and considering grammar structures, without actually baking those into the model as logic.

            AI text does read pretty naturally, so hopefully my interpretation is correct. But it’s also very verbose, and can repeat itself a lot.

  • DragonTypeWyvern@midwest.social
    link
    fedilink
    English
    arrow-up
    0
    ·
    1 year ago

    My favorite is the Google bot just regurgitating the top result. Which gives that result exactly zero traffic while having absolutely no quality control, mind you.

  • NutWrench@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    edit-2
    1 year ago

    I’ve been trying out SearX and I’m really starting to like it. It reminds me of early Internet search results before Google started added crap to theirs. There’s currently 82 Instances to choose from, here

    https://searx.space/

      • Voroxpete@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        0
        ·
        1 year ago

        So does pretty much every search engine. Running your own web crawler requires a staggering amount of resources.

        Mojeek is one you can check out if that’s what you’re looking for, but it’s index is noticeably constrained compared to other search engines. They just don’t have the compute power or bandwidth to maintain an up to date index of the entire web.

        • voxel@sopuli.xyz
          link
          fedilink
          English
          arrow-up
          0
          ·
          edit-2
          1 year ago

          yeah but that invalidates the “better/cleaner search results” point since it’s well basically the same stuff, just without the tracking