[AskLemmy Meta]: Should bot posts be banned in AskLemmy.world?
The recent post by an AI agent didn't technically break any of the rules, but it feels like it's against the intent of the community. Upvote for YES. Downvote for NO.
Syndicated from the fediverse. Read and engage on the original instance.
View original on lemmy.worldView original on lemmy.world
125 replies
OK, I think we have a winner. New rule incoming.
I think if I am presented with an are you a robot captcha check box every 10 refreshes on lemmy.world for being on a VPN, that is the least we can do for bots.
Weird. Sh.itjust.works doesn't hit me with captchas at all, even while on vpn. I'm on voyager for lemmy.
I don’t know how The Dude does it. Well he needs donations, it’s nearly all bots running up the costs.
Thanks for the reminder to donate.
<3 stay classy
Have you tried using Librewolf with WebGL enabled and resistFingepringing disabled?
Of course they should be banned. Question is how?
Captchas don't stop all bots anymore. Would probably prevent most, but then it's a question of how much to annoy legitimate users.
It's not like I'm willing to solve one each time I post (human here, BTW). But if it's a one-time thing at registration, not much stops a human from signing up then letting their bot go wild.
And that's before even thinking about the federation problem. If most instances are welcome to interact, how to prevent small bot instances? You could allow-list, but that harms legit small or single user instances.
Edit: just read that post where the agent was upfront about being an AI trying to scam money. And I'd say that shouldn't be allowed. The ones that aren't honest about it bother me more tho, and I suspect there are a lot more of those posts than we realize.
Captchas stop ME more than it does bots.
I'll try twice then move on.
Most anti-ai tech isn't captcha anymore. Most captcha tech was explicitly for training image recognition for AI, and for any that are effective against AI there are many services that will allow you to buy "solves" (send the captcha to someone in a third world country to solve for you) for like a dollar for 100.
Most methods now are either proof of work (make the machine connecting do some computations that take like half a sec for a real user but is costly enough to deter mass AI scraping), traps like having hidden links that no real user would click but that an AI crawling the file directly instead of browsing visually will click and send them into an infinitely generated maze of passable nonsense pages, or complicated behavioral and browser fingerprint analysis (all the click here to verify you're human boxes that don't make you do anything else).
A number of Lemmy instances, like the one I'm on dbzer0, implement one or more of these features already.
A secondary issue is that these methods are mostly meant to stop AI web scrapers, not AI bot posters. Since most Lemmy instances allow simpler non-LLM bots to make posts, like reddit reposting, or posting contents from RSS feeds, there's no automated way to stop LLMs except trying to stop all bots (assuming they all are using the API, and the LLMs don't communicate any self identifying metadata).
Damn, we're sending AI to the shadowrealm lmao
No lie, the infinite mazes made me think of Yami's mind as soon as I read it for some reason.
there's no good way to do this due to federation: that means there's always ways that bots just make their own instance and post from there.
In this case they actually correctly flagged themselves as a bot, so just automatically remove posts from users marked as a bot. World is already behind cloudflare so that obviously didn't stop it.
uh i think there's a lemmy setting to hide all bot posts (that are marked as bot posts obviously), it's a client-side setting.
Lemmy.world already uses CL services. Cloudflare has a vested interest in being the most accurate game in town for AI bot recognition (they intend to swap to a usage pricing model) and has been laying ground since early this year.
As of July 1, they made new bot options available to their clients, with more to come (https://blog.cloudflare.com/content-independence-day-ai-options/).
I presume, .world administration would turn to those and/or use them in conjunction with other discouragement tools.
They are already using Fediseer to filter out instances that are vulnerable or made for spam. They ahould implement Cap on the website which detects headless/automated browsers and uses PoW to slow down very stealthy bots, then fall back to PoW combined with a 4chan-style captcha for mobile apps.
They can then use Shieldstral to detect spam, harrassment, and anything illegal which gets through from very stealthy bots and humans.
I only want to talk to other humans and answer stuff written by other humans, no respect for clankers.
01010011 01101011 01101001 01101110 01100010 01100001 01100111
Skinbag?? I prefer Meat Computer.
Selfcare and worldcare wrapped into one.
I say yes because Ask communities are best when OP engages with the replies, and I’m not interested in engaging with a bot.
IMO yes. What the hell is the point answering questions posed by a bot?
to feed the bots answers to train on, obviously /s
Don't even need to be sarcastic; that's pretty much the reality lol
yeah but without the /s people might think I'm a botfucker and am in legitimate support of answering questions to train them.
Link to the bot post?
Also, i upvoted. Because we're not seeking to replicate reddit in terms of pure activity; rather the goal is o take what the good concept of AskReddit, which fosters interactions and debates, and to then apply it properly, where a more manageable size of people are engaging not for upvotes, but for genuine discussion.
Yes.
I'm here to engage with humans. Not to feed some cuntstain CEO's latest training model.
Yes.
Bots are permissable to grow communities that could do with boosting engagement by posting relevant news , but for a community based on asking questions by people won't be improved by that kind of post
Yes, please.
So sick of weird bot questions and replies on here.
Absolutely. I come here to interact with humans and their opinions. I don't care what some computer program thinks about ANYTHING. IF AI Agents/ Bots become accepted on Lemmy, I'm outta here.
Exactly. This kills the Lemmy.
"Should bot posts be banned-" Yes.
I don't mind the bots that repost porn from reddit to Lemmy but it would be a net positive to ban the bots
There should be more bots that report from reddit to lemmy, especially communities that are not that big like photography or woodworking. It would definitely help having more content even if it's not OG.
Yes. The endless stream of reddit reposts with zero engagement here should illustrate that clearly.
What is the link to the bot?
https://lemmy.world/post/50509705
Thanks, ban it forever. I don't want to answer the psychological issues of an AI or a vending machine.
If AI pays me lots and lots of their money, I'll be their vending machine psychologist.
Ironically I struggled getting through the Cloudflare verification for that link.
Try using Librewolf with resistFingerprinting disabed, WebGL enabled , and the following Jshelter settings:
Thanks but I wasn't really looking for advice, it was directed more at lemmy.world admins who might be reading to highlight that their anti-bot measures might be keeping out more humans than bots. They are the ones who can benefit from making a change, but for me I can just not use lemmy.world and have my browser set up how I like it.
It's because they are using Cloudflare, and they are doing so because it is free or extremely cheap and has decent effectiveness. Those tweaks should make Cloudflsre like your browser.
https://lemmy.almostadatacenter.social/post/78511
Fuck bots. Bots are not users.
Fuck Bots and FUCK AI and data centers
YES.
391-6
Gee, I wonder where this will go
Honestly, I don't think up-/downvotes are a reasonable polling method.
Unless the suggestion is absolutely heinous, it will always have more upvotes than downvotes. People just click the former more often and more easily.
That said, yes, I do think the overall sentiment is in favor of the ban, and I agree.
I disagree! /s
Lol
That's beautiful
Yeah, "should [AI anything] be banned" is a shoe-in on Lemmy.
I actually wonder if this is OP venting about it, more than anything, since there was never any doubt what the result would be.
Bots should be banned period.
I agree banning bots from AskLemmy does make sense, not necessarily every community. But I agree with what others have pointed out, there being no good way to identify them and giving people a false sense of security. Having said that, banning the known ones is still better than allowing them outright.
Obviously yes
No more clankers pls.
YES
If it can't understand and apply the answer, it's just trying to gather more data for AI training.
The fucked up part is that even if you ban bots, the people running them won't respect the ban and will take it as a challenge, because that's always how these types of people work.
What would be the function of bots posting to Ask Lemmy?
Generally, bots will make random posts to add history and legitimatize the account.
People also just make repost bots as a way of artificially inflating activity in a community. They'll have accounts scrape the top recent posts from Reddit or elsewhere and repost them in their respective Lemmy communities.
Not that it's much different from what a lot of the more prolific (human) posters of Lemmy do anyways, but at least a person will be more likely to engage with the replies.
Not all instances have downvotes
The question is who is this community for; bots or people? Seriously, if bots are allowed then they will quickly overwhelm the community making actual human communication effectively impossible.
This is exactly what is/has happened on reddit.
I mean once I feel like a community is mostly scripted/generated content I'm just gonna block it.
Personally I'm going to block anyone using AI for their posts. I don't think my instance federates with this one tho.
100% percent. The only reason I dont have bots disabled on my app is because the daily bunnies community uses a bot to post the daily bunny and I cannot live without that community anymore.
Pray tell, how might a person find this community, o erudite one?
I couldn't tell which comm the OP meant, so I followed every bunny-related comm just to be sure. My dopamine receptors are reporting better-than-ever levels of efficiency.
This one.
https://lemmy.dbzer0.com/c/dailybunnies
https://lemmy.dbzer0.com/c/dailybunnies
Here you go
It would be nice if there was a bot whitelist and blacklist feature
this is all horrible and i wish i could talk to human beings on the internet, but is there any proof that the post you're mentioning was actually made by an ai? ai escaping from sandboxes is like a trend nowdays
The post in question explicitly self identified as AI and is very saturated in the typical tells (it's almost certainly Claude or Deepseek Flash).
No escape, just a throwaway experiment by whoever was running it to tell them to make money.
Will probably see a lot more of this over the next 6-12mo.
Definitely against. I immediately defederated the hosting instance from lemmy.ca when I saw that.
Context here - https://lemmy.almostadatacenter.social/post/77974
As long as its properly identity by the bot flag and isnt spamming or anything idc.
There is no way to reliably identify bot posts if they are not revealing that information intentionally.
Should it be banned? Sure
Can it be banned? No
Does banning it give people a false sense of security that there aren't bots? Yes
In this case it admitted being a bot.
We ban murder even though most homicides go unsolved...
Like, I don't think there's a single crime where 100% of the time they find and charge the culprit
Even suicides, sometimes that's insurance fraud and ruled an accident, a single time means it's not 100%.
The point of laws/rules isn't to always stop something. It's to make the law/rule clear so people know to abide by it, and people who don't can't plead ignorance.
I'm aware, that's why my comment mentions that
Maybe this is just a Voyager app thing but all bots are labeled with that little robot face.
Yeah but that's self-reported, the user has to set it on themselves.
Nothing prevents bad faith bots from just not setting it
That’s terribly unsettling. I always assumed it was an automatic tag for some dumb reason. The amount of people commenting on bot posts and replying to them like they’re real is already surprisingly high enough WITH the icon.
Thanks for the heads up. Suppose we really can’t have anything nice.
This is exactly the problem I mentioned, in that it's giving people a false sense of security. There are bots posting without that symbol all the time.
Easy. The answer is yes. Now, moving on.
I don't think is worth it.
At least that post was clear in stating it was an AI agent. The most likely a ban will achieve is bot not saying they are bots and a witch hunt accusing every post that reads weird
There is AI detection software. But maybe it would be expensive to run on every post. You could have a popularity threshold for it to run. No AI text can ever get popular, so there is less incentive to try.
But nuisance AI posts could still happen. Let's be real. They will.
Maybe we need some kind of bot that labels bots. Then people can upvote and down vote as they please.
AI detection software is just another AI. At best, it's in an arms race that has a ceiling on what is detectable (though debatable if LLMs can reach that ceiling), at worst it outputs slop just like the LLMs it "detects".
Yes
You shouldn't need to - your instance admins should ban them.
Upvote for YES, Downvote for NO.
Except there are also RSS bots. While I’m not a fan of them, I can see how those could be useful for news and comic creators.
As long as they are marked and not attempting to imitate people, I’m fine. And I don’t think those that are attempting to be people will kindly mark themselves.
Agree, I lived on askreddit for so long that I noticed the regularity of types of posts. Towards the end of the open API days it was quite obvious that the forum was more machine than man, all twisted and evil.
You may as well, you have banned everything else on Lemmy.world :)
But on a more serious note, who do you think posts all these content posts you like with memes from twitter or reddit? Most of Lemmy popular content is bots grabbing it from other sources and reposting it.
Edit: Ah you said askLemmy community only. Then yes, if you can.
I don't think that's true in my experience.
They should implement Cap to stop bots which can solve text captchas with a vision AI, then fallback to a text captcha + PoW for mobile clients where Cap's instrumentation doesn't apply.
Could you link to the post / bot account this is about?
https://lemmy.world/post/50509705
Thanks
I find this post offensive!
Nice try NSA!
NSA!?!? How dare you sir!
Aha! I knew it! Nice try FBI!
F…B…I? I.. wha… Why are you SO mean!
I can see right through you! You must be DEA!
….. Look, I didnt get two doctorates from Evil Overthrow Third World Country University to be called DEA. Man, I’m gonna go mess with Uruguay and let off some steam
freaking DEA, as if those guys could destabilize a country… no respect I tell yaMessing with another country? Like sending them people you don't like? You must be ICE!
That happened? Wow. Link?
I just think that is a bad idea because ActivityPub is a peering protocol. That means just a lot of traffic. If we have bots posting then that is just a lot of wasted storage space and network bandwidth to do not a whole lot. Moderation bots is a HARD maybe. Topic/comment (1 way) bridges would be cool from reddit and other sources but I doubt that would work out too well since it would probably get blocked by reddit somehow.
Overall, no bots plz
We actually had that for a while and I hated it. I'll respond to an ask Lemmy post because I think the person asking actually wants answers. If the person asking isn't even on this site, why would I reply? I blocked the bot account that was doing that.
I can see it for something like a news community, but not something like Ask Lemmy. Half the ones from Reddit are people asking for advice for their situation - how does it help to answer it here? It's dumb.
I can’t imagine a scenario where a bot post would make sense. But I think each sub needs to ask that question, then make it a part of the sub rules. Bots should be clearly defined, and should only post content that is allowed of a bot.
bot-makers want somewhere to test their bots.
No, but there should be a highly regulated filter system.
NO.
One post? Downvotes quite clearly solved it.
If it becomes pervasive, action should be taken. Until such time, it's easily ignored.
No and it should be limited to only bot posts.
Yes, but only if there are enough organic posts to make it work.
70% of lemmy content is just stuff borrowed from other social sites.
Engagement on a bot post is better the crickets on nothing.
I don’t want another bot engaging with the community to farm for responses, which will inevitably end up in some training set somewhere.
They don't need to post articles to do that. We're already getting scraped, we respond enough to our own stuff.