• Dr_Box@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    1 month ago

    Ok what parts of this would actually count? Do the corners in the 2 panels count? I’ve had to try this version of captcha again so many times

    • jettrscga@lemmy.world
      link
      fedilink
      arrow-up
      0
      ·
      edit-2
      1 month ago

      I think it’d only be 2 squares for the rectangle of the traffic light.

      I’ve been doing a lot of captchas lately and they always seem to want less outer edge than I expect as a probably human. Like if only a small corner of a bus is in a tile, I just ignore it.

    • finitebanjo@lemmy.world
      link
      fedilink
      arrow-up
      0
      ·
      1 month ago

      It’s being heavily impacted by AI as well, since the “correct” answer is chosen democratically so if enough bots are answering captchas then they choose whats correct and it becomes better at verifying botity than humanity.

  • Whelks_chance@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    1 month ago

    Pretty sure that’s not how captures work. You don’t fail them, you add to the training set. You’re against the masses as to whether they considered it a traffic light when they were shown it.

  • UnfortunateShort@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    edit-2
    1 month ago

    Recaptcha is one if the worst providers of such services. At times, I need to complete like 10 for being one square off or whatever, while I need a single attempt for e.g. Arkose Labs stuff.

    Sometimes, like in the picture (whether it’s real or not), the captcha is simply wrong and you can’t do it right but by accident. I already did not visit websites because I didn’t feel like giving it another try.

    • PrettyFlyForAFatGuy@feddit.uk
      link
      fedilink
      arrow-up
      0
      ·
      1 month ago

      are you using a datacentre IP?

      I’ve found that when i have my VPN connected it will get me to do like dozens of them. if my IP isn’t connected it is usually just 1 or 2.

      Also when it’s in the cycle of prompting you for dozens of them, skip any like the one in the image above until you get one of the ones that has you select from multiple matching pictures. these seem to be the only ones that matter when it’s in that state.

    • Pazuzu@midwest.social
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 month ago

      Try the audio captcha option, those usually have an actual answer it will accept. Which ironically speech to text is more or less reliably able to solve, and there are extensions to solve captchas automatically for you that way

    • abbadon420@lemm.ee
      link
      fedilink
      arrow-up
      0
      ·
      1 month ago

      As a large language model, I cannot appreciate art. However, this painting by Monet is especially exquisite because of the fine details in the brush strokes and the feeling of space in the …

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

    It takes an extra second or two, but I’ve been using speech-to-text on the audio option. Instead of a chore, it’s an exploit! (Not really but your dopamine doesn’t have to know that)

  • NutWrench@lemmy.ml
    link
    fedilink
    arrow-up
    0
    ·
    1 month ago

    Ever wonder why these captchas are always cars, bicycles, motorcycles, traffic lights and crosswalks? Because YOU are doing the work of teaching the next generation of AI for self-driving cars.

    • reksas@sopuli.xyz
      link
      fedilink
      arrow-up
      0
      ·
      edit-2
      1 month ago

      I wonder if everyone agreeing to always select bottom row or something would do anything

    • TheOakTree@lemm.ee
      link
      fedilink
      arrow-up
      0
      ·
      1 month ago

      My favorite is when it asks me to identify stairs. I just imagine a self-driving car mistaking a set of stairs as more road and deciding to try and climb the steps.

      • Daemon Silverstein@thelemmy.club
        link
        fedilink
        arrow-up
        0
        ·
        1 month ago

        Actually, it’s training a self-driving humanoid robot that’s supposed to climb stairs in order to terminate any potential John Connor that’s inside a house upstairs.

      • gamermanh@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        0
        ·
        1 month ago

        You and many other humans are doing verification work

        It’s pretty sure it’s already right, but if enough people get the same image and get it wrong the same way then something’s up, flag it

          • gamermanh@lemmy.dbzer0.com
            link
            fedilink
            English
            arrow-up
            0
            ·
            1 month ago

            I took some compsci classes years ago when this tech was new and that’s exactly how it was described as being handled

            Once image recognition software got good enough to be right most of the time they started this shit to help get it the rest of the way to all of the time

            Do it any other way and you have to pay those people

    • Squorlple@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 month ago

      Can’t wait until we get trolley problem CAPTCHAs and we have to choose the square with the most expendable human lives

    • yetAnotherUser@discuss.tchncs.de
      link
      fedilink
      arrow-up
      0
      ·
      1 month ago

      I don’t believe it, at least not anymore.

      Google has had more than enough data to train AI models from reCAPTCHA for many years. In 2010 it displayed 100 million captchas per day. You simply do not need hundreds of billions of solved captchas in your data set.

      I feel like its only purpose nowadays is stopping basic bots and annoying people who don’t let themselves be tracked as much as advertisers would like.

    • whoisearth@lemmy.ca
      link
      fedilink
      arrow-up
      0
      ·
      1 month ago

      I can’t believe I never put that 2 and 2 together.

      It stresses how stupid AI is then if it was a human the question would be “is this a stop sign?” So it’s not even asking us to validate data. To me that means AI is still far from being intelligent. It’s requiring our input to learn. That’s not how we operate. My kids don’t require me to show them images of a stop sign for them to know what one is.

  • lime!@feddit.nu
    link
    fedilink
    English
    arrow-up
    0
    ·
    1 month ago

    i really wish quote tweets (xeets? skeets? yeets?) would put the original post above the response…