• Justin@lemmy.jlh.name
    link
    fedilink
    English
    arrow-up
    76
    arrow-down
    1
    ·
    5 months ago

    The fact that Google suggests adding reddit to the end of the search term there is something.

    • morgunkorn@discuss.tchncs.deOP
      link
      fedilink
      English
      arrow-up
      38
      ·
      5 months ago

      oh nice catch, i didn’t even see that! yeah it’s catching on, but maybe also their biggest downfall, because there’s a whole lot of shitposting going on there too. I’m always delighted when they return an article from the Onion and present it as a fact

  • RustyNova@lemmy.world
    cake
    link
    fedilink
    arrow-up
    58
    arrow-down
    1
    ·
    5 months ago

    Once humanity finally realises how bad AI is, can we make meme pages shelters for those poor orphan models?

  • sp3ctr4l
    link
    fedilink
    arrow-up
    50
    ·
    5 months ago

    Pff, everyone knows that Sebastian Somersaült paved the way for all that came after him, back in 1062.

  • PenisWenisGenius@lemmynsfw.com
    link
    fedilink
    arrow-up
    24
    ·
    edit-2
    5 months ago

    I wish they would just let ai always give these silly answers instead of patching them. Ai is dumb as shit but the bullshit answers are often comedy gold. I need these laughs in my life.

  • saltesc@lemmy.world
    link
    fedilink
    arrow-up
    17
    ·
    5 months ago

    I’ve always assumed many of these are just editting element text, but mobile that seems more effort than worth. Is there a way to quickly confirm them if not using/having access to the feature?

      • dingus@lemmy.world
        link
        fedilink
        English
        arrow-up
        15
        ·
        edit-2
        5 months ago

        Tbh I always assumed these AI search results posts were fake. But I just did the search and got the same weird result from TikTok highlighted at the top.

        • hex@programming.dev
          link
          fedilink
          arrow-up
          3
          ·
          5 months ago

          This is a different one. The tiktok one is just ripping the text from tiktok, and google has had this feature for yeaaaars. It’s just embedding the content on the page.

          The one from the OP is real too and that’s using this feature called “SearchLabs AI” which is written by AI

      • saltesc@lemmy.world
        link
        fedilink
        arrow-up
        9
        ·
        5 months ago

        Welp, that’s enough for me to assume 80% of these be real lol. Now I feel like I’m missing out, but I don’t think I am.

      • Ilovethebomb@lemm.ee
        link
        fedilink
        arrow-up
        1
        ·
        5 months ago

        I got exactly the same thing.

        AI has no concept of satire, which in my view is a good thing, as it makes people question just how accurate the information being provided really is.

    • morgunkorn@discuss.tchncs.deOP
      link
      fedilink
      arrow-up
      7
      arrow-down
      1
      ·
      edit-2
      5 months ago

      We’re in the shitpost community, even if it’s fake it’s still funny?

      Being in the EU, i can’t check myself, but in the answers someone could reproduce it.

    • Mirodir@discuss.tchncs.de
      link
      fedilink
      arrow-up
      6
      arrow-down
      2
      ·
      edit-2
      5 months ago

      Assuming AI Overview does not cache results, they would be generated at search-time for each user and “search-event” independently. Even recreating the same prompt would not guarantee a similar AI Overview, so there’s no way to confirm.

      Edit: See my comment below for what I actually meant to say

      • snooggums@midwest.social
        link
        fedilink
        English
        arrow-up
        6
        ·
        edit-2
        5 months ago

        Multiple people in this thread, including myself, have the exact same tiktok meme quote as results for that prompt.

        “AI Overciew” is not the same as randomized image generation.

        • Mirodir@discuss.tchncs.de
          link
          fedilink
          arrow-up
          1
          ·
          5 months ago

          My bad, I wasn’t precise enough with what I wanted to say. Of course you can confirm (with astronomically high likelihood) that a screenshot of AI Overview is genuine if you get the same result with the same prompt.

          What you can’t really do is prove the negative. If someone gets an output then replicating their prompt won’t necessarily give you the same output, for a multitude of reasons. e.g. it might take all other things Google knows about you into account, Google might have tweaked something in the last few minutes, the stochasticity of the model is leading to a different output, etc.

          Also funny you bring up image generation, where this actually works too in some cases. For example they used the same prompt with multiple different seeds and if there’s a cluster of very similar output images, you can surmise that an image looking very close to that was in the training set.

      • saltesc@lemmy.world
        link
        fedilink
        arrow-up
        1
        ·
        edit-2
        5 months ago

        I do that with LLM a fair bit. If just using GPTs website for something that should be simple, I often prompt the same thing several times and choose the best iteration as a base.

  • 2xsaiko@discuss.tchncs.de
    link
    fedilink
    arrow-up
    11
    ·
    5 months ago

    At this point I have von idea if most of these are faked or not but it’s funny and definitely plausible after the pizza glue.

  • boredtortoise@lemm.ee
    link
    fedilink
    arrow-up
    11
    ·
    edit-2
    5 months ago

    Schools have said forever that not everything you read online is true and NOW AI companies and people are surprised

    Newsgroups and BBSs were full of tomfoolery