It feels like the amount of both, divisive posts and ghoulish comments is rising again.

One could argue that the world has a lot of divisive stuff going on and lemmy just talks about it. But the way people post about stuff seems more oot and hateful than it has been in the past.

Not saying it is that but if I wanted to bring the Fediverse down or at least keep my customers from going there, I would sow this stuff as much as I can.

I’m blocking ghouls left right and center atm but if I ever asked a friend to join lemmy, I’d hate to think of what they would see that I dont anymore.

Do we need stronger moderation?

  • Maybe ban politics from c/memes?
  • Become a little more stringent on “dont be a jerk” rules in communities?

One thing that really bothers me is the collapsing “discourse”. Trying to mend fences and keep the conversation between sides going ime leads to nothing but downvotes and shitstorm.

I feel like a little more interaction (instead of intervention, at first) of the moderators would do wonders there.

Thanks for reading this rant. Have a nice day.

  • blahsay@lemmy.worldEnglish
    51·
    11 months ago

    If I was designing an anti troll/bot system I’d implement a few things. Let’s call any bad actor on here a bot/troll or broll for ease.

    1. Reputation based posting isn’t a bad idea if carefully done
    2. When a broll is banned any users from the same ip are flagged a suspect and a subsequent ban causes delayed posting from that ip. If a VPN the instance host list has the same effect applied. Exceptions based on Reputation.
    3. You can check if text is ai/llm generated and an automated api check before posting and immediate ban if found.
    4. Checks on if a user posts inhumanly fast or is oddly active etc would be sensible.
    5. Any broll system has to be adaptive and measures taken need to be kept secret (this post for example).
    • haui@lemmy.giftedmc.comOPEnglish
      2·
      11 months ago

      Very good ideas! Any idea if something like this already exists? If not, shall we work on something? I have some experience in python if that helps.

    • intensely_human@lemm.eeEnglish
      11·
      11 months ago

      You can check if text is ai/llm generated and an automated api check before posting and immediate ban if found

      If this LLM-detection function ever results in false positives, this system will be banning innocent people.

      Also there are many, many cases where a person openly displays results from an LLM, without it being in any way antisocial.

      • blahsay@lemmy.worldEnglish
        11·
        11 months ago

        The odds of someone coming up with the same sentence as an llm within common sense bounds of time far exceed winning the lottery or getting struck by lightning.

        Your second point is straight up nonsense. This platform is for humans to interact. The use of bots is inherently deceptive.

        Fascinating to have someone argue for them. I think the backend logs will be pretty illuminating.

        • intensely_human@lemm.eeEnglish
          11·
          11 months ago

          I don’t know what a person “coming up with the same sentence as an llm” would have anything to do with this unless the LLM detection is based on direct string comparison.

          The use of bots is inherently deceptive

          Nope. I can say:

          Here’s what GPT-4 generated when I gave it that prompt: “[some LLM output that would get them banned by the machine we’re proposing to build]”

          That is not deceptive. But it would be detected by this system and result in them being banned. Because you guys are gung-ho to build a powerful head-cracking machine and didn’t think of an obvious edge case.

          • blahsay@lemmy.worldEnglish
            11·
            11 months ago

            You’re wrong and don’t have the technical knowledge to understand why and I can’t be asked explaining it.

            Relax, it won’t affect that case.