I’ve noticed quite a few top posters that are using LLMs or other software to repost news links and images, probably from Reddit, onto a bunch of specific forums. They don’t identify as bot accounts, but they are fairly easy to spot because they are posting a steady stream of posts, around 2-3 posts an hour, 24/7. No sleep periods, just constant posts, at an unnatural rate.
Here’s a few examples I’ve found:
- @Grumpus_Maximus@thelemmy.club - Joined 2 months ago, 1.69K posts, sole moderator of Build A PC Sales@lemmy.world, steady stream of posts without sleeping, only posts on 4 different forums
- @sanitation@lemmy.today - Joined 2 months ago, 1.3K posts, steady stream of posts without sleeping, only posts on 5-6 different forums
- @beep@piefed.world - Joined 3 months ago, 2K posts, moderator of 15 different forums only on piefed.world, also a hidden moderator of Beep—Meta@piefed.world (yes, with an emdash) with zero posts and zero users, seems to try to evade detection by having some posting breaks in-between
Keep an eye on these kind of posters, and report them when you see them.
You must log in or register to comment.
I remember when grumps max joined he wanted to set up a relay or something to mirror Reddit content here. Looks like he succeeded.
His posts in history memes are usually garbage as unlike pug he never gives any context for them.
Beep caused quite the stir in !comics@lemmy.world for his aggressive use of ai to modify existing strips to remove any author or artist attribution among other things.
also a hidden moderator of Beep—Meta@piefed.world
!Beep—Meta@piefed.world
That community doesn’t seem to appear in the Mlem community list, and also doesn’t seem to be properly detected as a community by the standard link format. 🤔
Edit: The correct link is !selfpostbeep@piefed.world.
Beep keeps making new accounts but they’re easy to spot because they will purposely remove watermarks and author names then call it advertising when you call them out on it. Making me wish user blocks can be IP based or something so even if the idiot behind the accounts makes a new one, it’ll already be blocked.
But none of these accounts are bots. 2 out of 3 of them are just dumbasses.
Even dumbasses need sleep.
I comment more than the 3 of them combined have posted by a wide margin. Guess I’m a sleepless bot, too 🤷♂️
Thank for sharing this info :)
grumpy_cat
That fucker posts EVERY HOUR in @piracy@lemmy.dbzer0.com for 24 hours straight, had him pegged for a bot the moment I noticed his post frequency.
If a user doesn’t need sleep, it’s a bot. Thanks for bringing these users to everyone’s attention.
Auto-repost-bots are far from anything new. Did you check with the mods of the communities in question?
If they aren’t going to report themselves as bots, they have already broken the rules in my book.
The mods would know if they were bots that were supposed to be active in their communities. I don’t understand how that’s hard to grasp?
Re-post bots and scheduled-ahead posts. Both common for almoet 20 years. Welcome to 2005?
Bots can and should be clearly labeled as such, whether they are repost bots or moderation bots or what have you.
Here’s Hacker News’s bot, and here’s what it looks like on my instance. It has a special tag and it says what it is in its bio. Nobody’s disputing the existence of this bot. But unless the bio says otherwise, and/or has a tag, we have every right to assume the user is genuine. Breaking that trust is not acceptable.
The majority of internet traffic, for years and years, has been bots, including social media. If you want to waste your time calling them out without consulting those who would know and could do something about unauthorized bots, you’re gonna have a bad time.
Asking others to get upset about it, itself, reeks of engagement-bait. I have no reason to assume you or anyone-else is not a bot, just because you don’t have a tag.
The mods would know if they were bots that were supposed to be active in their communities.
without consulting those who would know and could do something about unauthorized bots
In this very thread, there’s a mod named Snoopy who just thanked OP for sharing this information. If they knew those were bots, they probably would have said as much. And since they’re thanking OP, I’m guessing it’s news to them as well - and the posting patterns are valuable information too. And OP is right to make this public, because it gives people the option to block the accounts themselves. If this was a DM to the mods, nobody would be the wiser and they couldn’t make an informed decision.
For the fun of it, I double checked Grumpy Cat’s home instance and found one of his admins asking him to make a separate bot account for his reddit mirroring. Which it seems like he is amenable to doing, which is perfectly fine by me (though some hardliners might be asking why we’re trying to turn lemmy into reddit again, but I digress). I just think he should have set the bot up in the first place, so I could block it and not him.
Asking others to get upset about it, itself, reeks of engagement-bait.
lol, lmao even. You’re comparing OP (or myself?) to the accounts posting every hour of every day? Accounts posting Amazon and Newegg links for products? I don’t know what to tell you… enjoy being engagement-free I guess.
Snoopy moderates only this community, !fediverse@piefed.social , and none of the accounts on OP’s suspect list post here, specifically. You can’t assume a moderator of one community is aware of autorized bots in another. LMAO, indeed.
Are you under the impression that LLMs, like OP claims these are, are best used for re-post accounts that can’t even be bothered to post less than a few times every single hour? Any LLM put to such task should be better at flying under the radar than that - these acounts aren’t really trying to hide what they are, or are doing a shit job of it. An LLM is an excessive waste of resources to spend on such task only for it to execute so poorly, when again, this has been a solved use-case for mainstream bots for decades.
Meanwhile, engaging in rage-baiting, trolling, and laughably sad attempts at talking down to or mocking people who call one on it is precisely where an LLM would outshine a classical bot, and probably most users, so yeah, no worries, you suck at thinking critically about the matter at hand, depth of knowlege, and vicious mockery rolls enough to be indistinguishible from a truly un-nuanced human…
… but you seem to think that’s something to be proud of in a world where AIs are being called-out for overly-correct grammar.
Finally, imagine thinking an admin dealing with one of these users disproves my claim that this should have been brought to the appropriate mods or admins directly in the first place, rather than spammed to an instance-wide meta-community, sorry, TWO instance-wide meta-communities, that forces normal users to see it.
Why yes, that is engagement-bait, rage-bait, LLM-like behavior, as are your flippant non-sense replies to my comments. To both you and OP, go bug someone else.
As hobbies go, it may be better than mine, but I get what you’re saying.
Nice, already had all 3 of them blocked
