Update: Pushing back against the wave of bot accounts on Lemmy

kersploosh@sh.itjust.works to Fediverse@lemmy.world – 1032 points –

cross-posted from: https://sh.itjust.works/post/1823812

This is an update to my previous post about suspicious inactive accounts on a handful of instances: (https://sh.itjust.works/post/998307).

I ended up messaging the admins at the 16 instances show in the attached image. I pointed out their wild user numbers, and referenced the lemmy.ninja post detailing how that instance scrubbed suspicious accounts from their user database.

6 admins responded. They had all noticed the odd accounts and either thought the numbers were wrong, or weren't sure how to purge the suspicious accounts without nuking their databases. In the end they managed to delete a combined total of about 338k dormant accounts from their instances. (One of the instances seems to have gone down since then.)

I never received a reply from the other 10 instance admins, though 8 of those 10 instances appear to be down (as of 27 July 2023). 2 instances are still up and unchanged.

Between the actively removed accounts and the downed instances, this represents a loss of 930,004 inactive Lemmy accounts!

You can see the drop in the graphs on The Federation. The total number of Lemmy accounts has been cut in half over the past 3 weeks, from a peak of 2.18M to today's 1.09M. The change is mostly from these 16 instances.

I have to admit, I did not expect such a large change when I started this! Hopefully this bodes well for Lemmy's future as a place where actual humans interact, rather than a cesspool of automated comments and upvote/downvote brigading.

That's all I have for now. Keep your stick on the ice; we're all in this together.

100

Great to see the transparency with which this is handled

The transparency may be my very favorite part of Lemmy. It's almost feels like these people are invested in it's success instead of it's profit.

It's a very early internet mindset where success == profit.

Open source vs we're a business mentality

Early internet grassroots collaboration stuff.

Those are crazy numbers... WTF?

If that's is the reality for Lemmy, I can’t imagine the number of bots giant social networks have. Crazy.

Thank you for your work.

That's the thing, right? Those giant networks' admins surely know how inflated their userbase is. They surely know that a lot of the activity is bot faked/manipulated.

But since the end goal of those networks is generate traffic to sell something (ads, user data), they never purge the bots. They need fake engagement. They might even promote it. The human user is just being used (Cf. Stallman's use of this term).

2 more...
2 more...

I want to celebrate two things. 1. Your awareness of the potential dangers looming over the fediverse. 2. Your proactive attitude curtailing the problem at its root. From one human to another, thank you!

Thank you for your efforts to keep this place clean and civil, and especially for the transparency in describing how you've dealt with such annoyances. You have my respect.

You have my sword.

And my ass!

actual humans interact, rather than a cesspool of automated comments and upvote/downvote brigading.

Thank you! That's why I left the other place. You're doing God's work, anon.

Unfortunately no website is safe from the cancer of AI/bots. The Internet is truly in trouble.

That's actually really interesting. What's the purpose of so many inactive accounts at once?

Seems to be enough to have a few of them, and not a million accounts since it clearly will rise suspicion... :)

Very good that you found them. Fascinating.

Maybe an attempt to try and make the fediverse look more active than it was back then, to get headlines about how it has explosive growth etc. It was June and everything really took off then.

What’s the purpose of so many inactive accounts at once?

That really is the million dollar question. I don't know. My fear is that they were intended to sit unnoticed until someone had a malicious use for them. Maybe to mass upvote/downvote certain content to make it more visible. Or to become active at an opportune time to make divisive posts and comments. I saw many accounts like that on Reddit; they show no activity for years and then suddenly come alive and spew garbage. I'm sure we'll see some of that on Lemmy next year since there will be a major election in the US. Though hopefully less since a bunch of suspicious dormant accounts are now gone.

It's a smart move for a spammer to create a lot of accounts in the early days of a platform, before more restrictive signups with mail verification, phone verification or captchas are in place. Look at how difficult it has become to register on Twitter or Facebook.

I'm thinking of making inactive accounts so I can create communities on other instances. I wanna make an old trek community on the Star Trek instance, but I wanna moderate it from this instance. So I would make an account, make the community, and transfer ownership.

4 more...
4 more...

I’ve got a couple accounts on various instances as backup, since we can’t exactly transfer accounts across instances just yet.

4 more...

Awesome work, and thank you for all of this, it is appreciated!

Looks like a bunch of personal instances that forgot to turn off self-registration. The down ones likely crumbled under the load

We need to find a way to get indexed on Google, Duckduckgo and other search engines.

Probably happening to some degree already, unless no robots is checked. No way lemmy jumps as high as reddit in seo for random things for a long while

Naa..I think instance admins have to let the crawlers index the content. Not sure if Admins have enabled it

A major advantage of the old place was that you could search up keywords and find a discussion on it. When I wasn't browsing, the other times that I'd end up using it was for when I needed to look stuff up.

I'm not sure what the pros cons ratio is though.

Well done. I for one appreciate the effort you're putting into making this a better place by keeping the bots out. Any thoughts on what can be done to keep bots from signing up to begin with or is the plan to continuously purge inactive accounts? I know from experience that a lot of these bad actors are going to pivot and redouble their efforts. This is unfortunately a cat and mouse game that will continually need to be addressed. But, again, thank you for your work on this!

Instances should enable verification to create accounts (email or captcha). I think everyone learned that pretty quickly last month. Other than that, it's up to users to diligently flag content and moderators to be responsive. Maybe there are good automod tools coming to Lemmy someday, but those are an arms race, too.

How does email handle it?

Are you referring to email verification on sign up? If so, it's unfortunately easily overcome by bad actors. Depending on how the platform handles it, one email can be used over and over again to verify accounts or there are many services out there that provide an endless amount of quick and easy emails. The automation of this has already been solved too. For the first scenario, limits on how many times an email is used for account verification is useful. For the second scenario, we really start the cat and mouse game. You can block sign up from accounts using spam email domains. There are lists out there that can help. If someone is really persistent, they may have a trove of legitimate email addresses they can use. Then you have to start considering where the sign ups are coming from, the IP, it's reputation, the behaviors, and hopefully it's fingerprints from the device. You could serve a captcha but most are trivial to bypass with code straight from GitHub or captcha passing services. Overall, this is not an easy problem to solve. I know a lot of conversation on Lemmy is being had regarding this topic. It's going to take all of us together to help solve the problem.

Email is federated very similarly to ActivityPub. How does Email handle filtering for bad instances?

I know they have sophisticated systems built up over decades that now seems to work quite well, but I don't really know the details.

I do believe if I stand up my own email server right now that I can still send email to people without being blocked, but I'm not positive.

I don't think I've ever upvoted something more enthusiastically in my life.

Cheers and thank you.

Suggestion: what if there was a lemmy instance solely for reporting malicious lemmy/fediverse servers? I've read some stuff about FBI crackdown and mastodon instances containing questionable material. Wouldn't it be gret to have some kind of federated "registry" of all the bad actors out there? I am pretty clueless, but would that help?

Might work. Like a lemmy admin coordination hub instance? On mastodon there is "fediblock," which is what you're describing.

I do know that there is lemmyadmin.site currently, but it's not extremely active as far as I can tell.

Why would you need a lemmy instance for that? A web site would be better.

What are qualifications for being an active account? I didn't see any details in the other thread about it either, just the graphs. Is it just post/comment creation? Is it page views? Log ins? Does voting up or down register an account as Active?

If it's only post/comments then you're possibly deleting a bunch of lurkers too.

Does Lemmy have a way to link to a post that anyone can use?

I can click the links up there, but it takes me to sh.itjust.works and that's not where I am in the Fediverse, so when I get there I'm no longer logged in.

And if so, can we have it so "wrong" links are corrected into the right format?

No way to do this I don't think, but you can paste the link into the search bar and that shows you the post on your instance so you can interact with it.

It's a bit janky but it works.

That didn't work for me. Possibly not "on" my instance, if that's how any of this works...

It does look like others have noticed it as well though.

https://github.com/LemmyNet/lemmy/issues/2987

Did you copy the link text and not link address, and put that in the search bar of your instance?

Yeah, it got two results and one was a link to this post. I must have clicked on the same one twice, because the other is the actual link I was after. 🤷‍♂️

Who needs fraudulent/abuse accounts anyway. I have moved to lemmy and am here to stay!

Thanks for the work!

May I politely ask how did you realize those inactive accounts

Should the instances that responded to you be refederrated? I’m pretty sure I saw some of them on lemmy.world’s block list. I think it would be sad for these small servers to not realize they are, in fact, not connected to the greater fediverse. On the other hand, if you’re an admin, and you don’t know what you’re doing to the point of not knowing your server was infected by hundreds of thousands of bots, maybe it’s too dangerous to refed.

Everyone has to start somewhere. We should reward honest effort instead of punishing honest effort but ignorance.

Ignorance is not an excuse.

A baby dies because you fed it 10 glasses of wine. You skip jail because you had no idea wine was not good for babies? Sure.

Did you just compare letting non-active bots on a server to killing babies? Okay if that's the comparison you really want to go with, I guess.

It's very easy to sit on a high horse and say you've never had a negative impact due to your ignorance, when your own ignorance might include things that you don't know are detrimental to others.

Are you perfect? If 7 billion people analyze your actions will none of them find you at fault? Because if even a single one does, then by your own standard, you should never be allowed a second chance by anyone.

Oh, I perfectly know I won't be liked by everyone, and I also know I'm not perfect. Nobody is.

That doesn't mean that we must reward effort "in spite of ignorance" when the consequences are horrid.

I'm all for enlightening the ignorant. But in this specific case, if they help spread vitriol and misinformation, then sorry. No excuse. It's 2023, do your homework.

When an account is signed up, is there information such as client ip address that could also be used to spot more inauthentic activity? And more generally, sign up should probably be made resistant to automated bots by randomizing HTML layout & ids and using captchas so it's not so easy to drive sign up through scripts.

Thank you for doing this. You're making the world a better place for all of us.

Can Lemmy even protect itself against spam while being open?

You are doing fantastic work and we all appreciate your initiative to keep Lemmy clean. Thank you so much for your efforts!

How did those accounts get created in the 1st place? Arent there captchas? Or are there ways around that? Strong captcha system should he implemented in lemmy by default

Captchas are a low bar to modern standards. All the advancements in AI are a problem for captchas. Machine vision tools have become abundant and simple. Unfortunately, I'm not sure what more you can do except require human review of access.

I'm not sure about all of them, but at least some of these instances did not have captcha or email verification enabled when they were first set up. They were easy targets for scripted account creation

Dude wtf, there are relatively many fucking servers which have well over thousands of inactive users. I checked some and it seems the mods of them are just posting under 5 posts on some other servers and than creating some communities in their own server and then leave quietly. Thats too sus... It may be too much paranoiac to think that there is more going about those servers but I just cant stop thinking it is too absurd

Some of them are probably lurkers. Might be good for Lemmy to store the last login time, so admins can consider that when purging accounts.

This is a possibility, but I think a lurker wouldn't sign up to those servers, they are literally randomly named empty servers. They would prefer more known servers, or they may just look up which server to choose and eventually end up on registering to known server.

Now I understand why youtuber AvE says “keep your dick in a vice” thank you so much.