Are there any bots that make shitty comments or something like that? And is there anything lemmy would do against bots? I don’t like dead internet theory and hope this place is free
Hello fellow human, I’m so glad you asked. All evidence points to there being no bots here
Yes
wethey are here.Bots are like microplastics, they’re everywhere.
Yeah, there are even upvoting/downvoting bots.
Lemmy isnt a mainstream place so probably its not affected as much as the huge platforms (I doubt anyone cares about what Lemmy users think about the world).
Lemmy does have value, because the demographics are a bit more focussed here compared to much larger but more mainstream platforms, not to mention that larger platforms like reddit are already so bot-infested, they start to interfere with each other. It’s so trivial to make and deploy bots now with LLMs, I’d be very surprised if there weren’t bots here.
I think the reason we don’t notice them as much here is the quality of moderation tends to be higher (on average, so far).
Im not sure we dont notice them. How do we know what upvotes certain posts to the front page? Is it really organic? Maybe, maybe not… :)
Actually I was thinking about this the other day. All the meme posts, are they created by humans or Ai?
There are tons of them with different jokes and messages and agendas. I certainly dont think its all people sitting and creating those…
Wouldn’t it make it more targeted
How do you figure? I was just thinking that with a minimal user base, why bother influencing it. Even though it would be easier to influence it, true.
Because you’re thinking about it wrong. With a bigger population it’ll cost more to influence them. If you ever watched how they did it in 2016, they started small, always. The first sub Reddit’s to see lots of opinion shifts where small subs like local interest before they moved on to the larger metropolitan ones. If you’re paying to influence people then you need a chain of initiation to start. This makes Lemmy a much greater target than Reddit. Look at opinions on AI. Lemmy was full on sharing multiple daily headlines like “AI is coming for your daughter’s” while Reddit thought it was just a neat tool.
The goal is 10%
You need to create enough bots to maintain 10% of the content is favoring your view. If you can sustain that then opinions begin to shift for the entire group. People will start to join and create their own. That 10% is much easier to achieve in smaller places. Lemmy is perfect for it
Interesting. I havent followed reddit so I dont know, but sure, I can see how that could have happened.
Are you a bot?
Yeah, we definitely have bots. Robofuckers can’t keep their grubby little fingers off the rest of us humans. There’s also a lot of people who generate AI imagery and spam it everywhere. It’s pretty disgusting, but that’s just the modern internet. 🤷♂️
Bleep bloop no, human.
Bots? Nah all our psyops are 100% certified organic.
Yeah… I think that people have this misconception that state-alligned propaganda has to be bots. No lol, there are actual people who voluntarily simp for states for free 🤣
I had peers in my school acting edgy and said “Hitler/Stalin did nothing wrong” , like bruh 💀
These people use the internet. I’m not sure if they even believe in what they say, but saying shit like that gets attention, and that’s what they crave for.
I’m sure there are bad natured bots, but they certainly aren’t as prevalent as on other platforms. I think lemmy is so much smaller in size and population that it just isn’t as worth the time to deploy a lot of resources. With the lower population, the bots likely stick out a bit more, which makes it a lot easier to block them.
But that’s just a theory… A bot theory.
Let me definitely has a considerably smaller cost to reward ratio.
The number of people that can be reached and influenced on Lemmy with bots is infinitesimally small compared to something like Reddit.
But I guarantee you that the bots do not stick out, LLM bots are pretty damn good at blending in these days. And the shitty bots have been sliding by on the Internet for over a decade now.
Lemmy and other federated services are in an unfortunate position where they have moderation and administrative tools that are on par with what would be expected 10-15 years ago for a large social media service (ie. Reddit). Which means we are almost entirely unprepared and incapable of handling malicious actors on Lemmy.
Agreed. I cannot even fathom on how to detect LLM bots.
There are still several tells that haven’t completely been ironed out. From years of reading bot comments on reddit, I’m able to discern a bot from a human pretty quickly, so long as I am able to check comment history and time stamps.
They have this weird uncanny valley pattern of speech and it sticks out like a sore thumb when I come across it.
People act like it’s the worst thing to look at comment history but it’s necessary to check if someone is a real person or to check if their extreme views are really sarcasm.
Y’all, I type like I talk IRL. Let’s see bots go redneck.
My exact opinion.
Say I’m paying you to get on social media and influence opinions with your bots. If you came at me with numbers from lemmy I’d be enraged.
If y’all think bots are here, you don’t understand percentages.
Obviously. Mostly clearly identifying as bots. You can go into your account settings and choose not to see bots.
There are definitely bots but I think people are good about making their bots known, and most of the content you see is posted by humans… because who tf would pay to run a bot farm on our little corner of the Internet?
If you want to see absolutely no accounts that self-report as bots, there’s a setting for that in Lemmy.
iirc there was a state backed bot farm on mastodon, some might have crossed over
I’m not a robot. I’m a real, genuine, person-human.
I’m a meat popsicle.
I’m an ugly bag of mostly water.
Would you like a free RAM upgrade?
No we’re not.
I’m hoping it’s not the case here on Lemmy.
I usually give people the benefit of the doubt, because a lot of us are ESL speakers from different parts of the world, so we might say things in a way that sounds strange or “unnatural” to native English-speakers.
I feel like nowadays the bots are actually really good in English because of LLMs.
That’s true. Tho, I’ve seen people who managed to get ChatGPT to output in “uwu/cringe”, so I’m still not going to let my guard down if I see something that doesn’t have good spelling and grammar.
slightly unrelated but there’s a community dedicated for gpt2 bots doing its own shitfuckeries with other gpt2 bots. It’s sometimes hillarious
edit: I guess it’s not here anymore, couldn’t find it.
