Just a 15 second game like Snake or Helicopter. Should stop a significant level of bots, no?

  • AB7ORH7D@lemmy.worldOP
    link
    fedilink
    English
    arrow-up
    3
    ·
    1 年前

    With AI capabilities now, surely it’s pretty easy for an AI to follow a set of instructions like: create an email, check email, click link in email…etc - is that correct? Or put another way - why would email verification stump ML so consistently if it’s trained to create emails and do the process

    • j4k3@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 年前

      I’m only parroting. The developers of Lemmy mentioned this as the only empirically effective option in the real world. AI in the real world is far dumber than it is framed in the media. Even a well trained model can’t just run off on a long tangent path to connect the dots. It only works for a very short time under controlled circumstances in the real world before it must be reset. This is not real AI.

      • vegivamp@feddit.nl
        link
        fedilink
        English
        arrow-up
        2
        ·
        1 年前

        You’re right about AI - it doesn’t exist and is decades away, what we have are increasingly capable statistics engines, aka machine learning.

        A for the topic, that’s easily worked around by having domains with catch-all addresses and a script that just clicks any registration link that comes in. Of course such domains are easily spotted and blocked, but domains are cheap as hell, and there are undoubtedly plenty of botnet nodes on hosts that can receive mail du they don’t even need to register their own domains at all.