Just me?
Uh oh.
As a mod I can answer the last part of your question: If it looks like a bot, smells like a bot, acts like a bot, guess what it is?; A duck.
And ducks get banned.
Just to clarify: Only the ones trying to pass as human. Utility bots (and ducks) are OK provided they are marked as such. Creating sockpuppet accounts for later use is not. While the boards I mod don’t explicitly have rules against bots, zero interaction repost bots hoping to farm some upvotes do not help the community.
Utility bots (and ducks) are OK provided they are marked as such
Far too few utility ducks around here; marked or not!
Some guy yesterday with a one day old account said something vaguely related to the post that sounded like an AI wrote it trying to sound human.
I tried to coax the LLM into a nonsense response but instead two people responded and I received -10 votes.
Looking into both their content history, it’s clearly the same person.
People are testing the abuse of no guardrails on Lemmy. It will get worse before a solution is created.
People are testing the abuse of no guardrails on Lemmy. It will get worse before a solution is created.
Literally why humans can’t have nice things
I’ve noticed that for at least the last year.
it has gotten considerably worse in the last six months though.
I haven’t noticed this at all and now I’m concerned that I’ve interacted with/upvoted bots :/
it’s likely you haven’t had interactions with them since you don’t have many comments posted. for example I have almost 3600 comments since 2024. I don’t even know if that covers the amount that I have deleted either, so it could be more.
There have been a couple of people who’ve ousted themselves for doing it shortly before they deleted the accounts.
So it tells me there are some very lonely losers out there, spending all of their free time, doing this.
Yes. Problem?
Their weirdness doesn’t affect my activity.
The ones that bug me the most come out of nowhere, make 30 to 50 posts in an hour, then delete themselves.
I stay on top of it in my communities and ban them as soon as I see them, but they still flood new and other communites that are basically unmodded. 😟
I still sincerely want to know what are they trying to accomplish? What is their end goal? And then after all the things they upload, why do they delete their accounts? Who are they? Is there something nefarious going on here? I’m pretty sure there is no money involved in Lemmy so it can’t be financially motivated, or is it?
I PM’d one of the accounts before they self immolated and got AN answer, I don’t BUY the answer, but it is AN answer. 😉
That is an interesting & weird answer he gave 🤔 I wonder how far he goes to reduce his social media footprint and why he goes to such extreme efforts? I wonder if he buys a new burner phone every time he makes a new social media account then shitposts everywhere then deletes everything. Very weird. What are his motivations & end goals?? There is no financial gain, there is no social gain, what else would people’s motives be?
It’s government bots. Trying to make star trek more popular than Star wars.
It’s CEO of Linux trying to brainwash us! /s
What about Star Gate?
The Canadian Government doesn’t have the resources to conduct such an operation.
Yup I’m seeing it, and then people wonder why there were so many rules on reddit banning posting until you had so much karma. Because bad actors spoil it for everyone.
As a mod of !fuckcars, I’ve been getting reports of stuff being posted by new accounts that pretty much immediately get deleted (not banned, as far as I can tell; deleted by the owner). I feel like there’s something fucky going on, but I’m not sure what should be done because the content being posted is legitimately on-topic and there’s no user to take action against anymore.
It bothers me that I don’t understand the motive.
The fact it’s on topic makes me wonder if they’re meant to drum up engagement, like when new small multiplayer games use bots as opponents to make the game feel more popular. Fake it till you make it.
I’ve been banning them as bad faith engagement. If they post and don’t own their post by maintaining the account, why are they bothering? 🤔
…maybe some sort of preliminary DDoS testing? Wouldn’t put it past bad actors to target small/open source oriented sites after the attacks on Arch Linux earlier this year.
I usually sort by subscribed so… IDK what is going on here
I’m not sure I would call that weird. That’s normal bot behavior.
I’m not on that community but in World, Europe and Politics community I’ve seen this issue too. Don’t recognize that alias with the numbers though. But a new user posts a whole lot for 3 hours and then deletes it’s own account. And then gets a new account. Repeat.
See also ex. this post on Lemmy/support.
add: fyi a solution: Tesseract and Voyager have filters for accounts< 1 month old
This is is Voyager? I don’t see it
should be an icon somewhere when accounts are < 1 months. See this picture it has this yellow thingy next to poster name <1 day
No real filter option as such, like Tesseract, but it helps for the " filtering".
Good call. I added “<1d” to filtered key words and it seems to work.
I got banned from RPGmemes for apparently boosting a bot account. So no idea.
Is that why they ban people? I got banned from that Community I have no idea why, never got a notification, I try to look it up in the mod log and I couldn’t find it.
You can be preemptively banned from a community or even an instance if the mods and/or operator have decided they don’t like you. The open source nature of the Fediverse being what it is, various tools and clients have been developed that allow select individuals with whackmobile hobby-horse viewpoints with varying degrees of insanity to sweep users’ post and comment histories for opinions or keywords they disagree with, singling you out to add to their ban lists. Merely being known to associate with the “wrong” instance (lemmy.ml and hexbear are popular targets) can score you a ban even if you’ve never posted anything at all.
Ostensibly this type of capability is supposed to be beneficial for keeping known bad actors and bots out of our spaces. But human nature being what it is, more prosaically it’s often used by instance operators or community moderators to… let’s just say, self-gratify themselves over the notion of getting back at their perceived enemies from a position of safety and no repercussions, in order to Teach Them All A Lesson or whatever the hell.
Some people even purposefully create ragebait communities with the express purpose of enticing others to comment there so that the creator can get their rocks off by being nasty to the participants and have a little power trip over banning everybody. We had a rather prominent one of those right here earlier in the year, in fact, which I now notice appears to be gone.
The Internet remains, as ever, a strange place.
Had a similar issues, and could contact a mod with an alt account to get unbanned, but indeed looks like they used a nuke to kill a Goblin.
Also, I still don’t get the point for bot to post “engaging content”. At worst it’s a bit too cliché you wake up in the body of the last person you ask with or too rage-baity Say something good or bad about Israel but it’s not like karma is worth something here.
Not only that but if the bot is increasing engagement in your community then…isn’t that at least an admissible thing?
If the bot is driving content, then it will need to be monitored closely.
Some community managers will set them up to port content from older communities or sites, to make the new federated community more active and give people content to consume. The downside is that these are bots and get treated like bots by the various instance owners.
Some of them are actually just karma / reputation farming bots, too.
The Fediverse not having a centralized blacklist and bot detection methods looking at things like IP and Packet Size make it inherently more susceptible to bot networks than traditional media sites, not like that matters now that those traditional media sites are promoting bot content and flooding the sites with AI.
Yep happens a lot over at the Shitpost community and then you have the ones over at politics spamming news articles. It’s weird.
