• edric
    link
    fedilink
    English
    4722 days ago

    Friend of mine was helping me set up a doorbell camera today and although it was wireless, it had the connections for a wired camera as well. We weren’t sure which colored wires go where so we both looked it up on our phones. I was using startpage and he used google. The AI generated google answer was the complete opposite of the first couple of results from startpage (google without the AI). We could’ve risked a short.

    • @Plopp@lemmy.world
      link
      fedilink
      English
      4322 days ago

      The colors of the wires in electrical systems and such are only for esthetics. Arrange them the way you think is the prettiest. If you’re lucky you’ll also get fancy sparkles!

      • @TranquilTurbulence@lemmy.zip
        link
        fedilink
        English
        1022 days ago

        If you like bigger sparks, you can also install a high voltage transformer from an old microwave. As an added bonus, you’ll get pyrotechnic effects on your doorbell for free!

      • @CheeseNoodle@lemmy.world
        link
        fedilink
        English
        722 days ago

        Honestly if your in the UK that’s basically true, we do have standards but for some reason there are different ones for appliances and structures and they also keep changing and not everyone updates their manufacturing when they do. In college we were told to just look it up because you can’t memorize them all and some of them use very similar colours but in a different order.

      • @atmur@lemmy.world
        link
        fedilink
        English
        422 days ago

        Technically yes but if you need to re-terminate the end of a cable in the future without seeing the other end of it, it’s nice to be able to assume it was wired the standard way.

        It I found an ethernet cable wired in the wrong sequence, I would re-terminate both ends immediately because otherwise it will be an annoyance in the future.

  • @gravitas_deficiency@sh.itjust.works
    link
    fedilink
    English
    4022 days ago

    Oh, it’s worse than that.

    Google’s “AI” results feed you things for 10 year old Reddit posts that are subtle (but sometimes, also not so subtle) bullshit.

    Whatever they’re using to curate training data is evidently pretty awful at detecting shitposts.

    • @CosmoNova@lemmy.world
      cake
      link
      fedilink
      English
      422 days ago

      Most of the curation or fine tuning is done in low income African countries so this is little surprising. They‘re cheap labour but you can‘t expect them to reliably detect sarcasm or notice mistakes in specialized fields. They basically give a thumbs up whenever the AI sounds convincing. Of course that includes instances where it‘s confidently wrong and that appears to be most of the time with this model.

    • @Even_Adder@lemmy.dbzer0.com
      link
      fedilink
      English
      222 days ago

      It’s not a training data issue, look up Retrieval Augmented Generation. It’s basically serving up stuff on the web and taking it as gospel.

  • @Hotzilla@sopuli.xyz
    link
    fedilink
    English
    22
    edit-2
    22 days ago

    Eventho GPT is far from perfect, this does show how far behind google still is on AI.

    Edit: GPT4o answer to exactly same question, which also explains why google failed:

    “The Onion” humorously refers to itself as “America’s Finest News Source.” It’s a satirical news organization known for its parody articles on international, national, and local news, mocking traditional news media and public figures. Despite its comedic nature, it has gained a reputation for its sharp wit and clever commentary on contemporary issues.

  • @dohpaz42@lemmy.world
    link
    fedilink
    English
    -4422 days ago

    Not from; about. The answer seems perfectly legit: it describes The Onion as satirical news, and even quotes part of the question as something The Onion calls itself. Not too far fetched.

    Where’s the fault? Ooooh yeah, I forgot. We’re trashing Google. Carry on then.

    • @Nobody@lemmy.world
      link
      fedilink
      English
      5622 days ago

      Ben Collins, one of the new owners of our former sister site, pointed out some of AI Overview’s most egregious errors on his social media. Asked “how many rocks should I eat each day,” Overview said that geologists recommend eating “at least one small rock a day.” That language was of course pulled almost word-for-word from a 2021 Onion headline. Another search, “what color highlighters do the CIA use,” prompted Overview to answer “black,” which was an Onion joke from 2005.

      Might want to read the article before commenting.

      • @dohpaz42@lemmy.world
        link
        fedilink
        English
        -38
        edit-2
        22 days ago

        Fair point. But, when the article image makes it look like trash-talk, I’m not compelled to read the article. ¯\_(ツ)_/¯

        • @aStonedSanta@lemm.ee
          link
          fedilink
          English
          3622 days ago

          How are you going to be a productive part of the conversation without reading the article? Just wasting everyone’s time.

          • @Zidane@lemmy.world
            link
            fedilink
            English
            1222 days ago

            Wait a second all these years I’ve been on Reddit/Lemmy and I was supposed to actually read the articles??? Nah mate I can’t be wrong