• ameancow@lemmy.world
    link
    fedilink
    English
    arrow-up
    4
    ·
    edit-2
    9 hours ago

    If you look at history you will notice patterns that raise suspicion.

    Making many posts on popular communities of low-effort memes and/or baiting questions and very few actual comments or engagement with the community is a pretty telling sign. The most common bots are karma-farmers, it’s just worth a lot more on Reddit but even Lemmy is a site that gives weight to people with particular kinds of history.

    They also do it for social engagement experiments and AI training, for posting memes and pictures that have advertiser watermarks, and probably a thousand other motivations I can’t think of.

    Low word-count posts won’t raise suspicion by itself, but if you check the actual context and see that they’re making almost random low-effort replies without actually making human conversation, it can be a good sign of roboshenanigans. “Wow that’s amazing, I can’t wait to share this with my wife” or “Where did you find this? I’ve been looking for something like this!” are both pretty innocuous replies, unless you see them posted in reply to a dumb spongebob meme for example.

    None of this means there aren’t bots that are indistinguishable now. The turing test is dead and buried, we’re entering a strange new world.