I love the fact that fediverse was built from the ground up to be free, federated and interoperable. I have two questions that may come from my lack of expertise / knowledge, so I apologise in advance if they are dumb.
- Bots can disrupt smaller instances:
What is stopping corpos from scraping everyone’s posts and stuff from the fediverse and train their AI? What’s stopping them then, to create loads of not accounts and spam / disrupt smaller communities? When an instances quality drops, the users may be more incentivised to migrate to bigger instances and go there. It’s safe to say most Lemmy users are not going to spin their own instance and start communities from scratch. Meanwhile, the onslaught of bots can overwhelm these budding communities and instances.
- Corpos can flood the fediverse with ads and crap:
Threads comes to mind on this point and how many instances have chosen not to defederate with them. Besides, they can create bridges, and have repost bots in all instances to flood major them with ads. With generative content, it is so much easier to make a seemingly casual post about a product and mask it as an advertisement.
I’ve seen previous posts about people wanting to come because of their opinion about how certain countries behave. I feel the true evil are the corporates.
There’s nothing stopping them from hosting a massive instance. But people aren’t forced to move there to interact with them.
If another instance started knowingly federating us ads, or fake content, I’d hit that defederate button very quick.
Nothing. Instances will have to take it upon themselves to defederate, just like how a lot of instances all decided to defederate from Threads.
Nothing, but instances can defed from them if they wish and they can’t stop that either.
What is stopping corpos from scraping everyone’s posts and stuff from the fediverse and train their AI?
Nothing. And they already have.
What’s stopping them then, to create loads of not accounts and spam / disrupt smaller communities?
Moderation.
When an instances quality drops, the users may be more incentivised to migrate to bigger instances and go there.
Why would they do that when every instance has the same content?
Corpos can flood the fediverse with ads
I mean they can but no one would subscribe to them or share them, so no one would see them, so why would they bother making them?
about how certain countries behave. I feel the true evil are the corporates.
Sometimes the two are actually one in the same.
Thankfully, there aren’t any ads here. Just the thought of it stresses me out, and when I get stressed out, I reach for a Morley cigarette to keep my cool. The toasted tobacco and asbestos filter make for a smoother smoke, which soothes the throat. 9 out of 10 anti-ad, Fediverse, activists choose Morleys to keep up their pep and vigor in the fight against advertisement.
This is comedy gold
more like comedy silver, amirite?
Silver screen?
Aye
Noice
Morley Turkish Golds!
Legendary comment, take your internet points
thats another thing. no internet points, so no bots to farm them. upvotes really only indicate the quality of the post or comment that receives the upvotes. no way to use the total number of points to claim validity of your posts or to brag with them.
that being said, at one point we will need to figure out a way to identify and prevent bots that just post propaganda. while we wont have the problem of karmawhoring bots, they dont have the need to karmawhore and can try to spread their propaganda immediately.
Upvotes on a comment are still internet points :p also this comment was just made as lighthearted fun not as something serious
ok sorry :(
i am just worried that the idea of karma on lemmy starts to gain traction :D
I never got why it was important to have “karma” tbh. Just encourages you to only post funny stuff or something the group in x community agrees with otherwise oh no you lose internet points 🫠
Karma can be a useful trust signal, until it’s accused of course and gamified.
back when reddit was not just a meme shithole, karma was used to motivate people to post quality content. 90% of the stuff posted on reddit nowadays would get downvoted into oblivion back then. typo in the title or image caption? downvoted. repost? downvoted. low effort? downvoted.
but with growing popularity the flat number of (usually young) users grew that hat other quality standards grew and these kind of faults were ignored. meme became bigger, and there was a general shift in reddits userbase. they changed the algorith to calculate your score and suddenly you could get just so.much.karma from cheap posts.
some communities became very strict in their content to avoid shitposts. some used karma to prevent trolls participating in their subs. nowadays high karma accounts are being sold as they can be used to participate in subreddits with a high reputation (due to participation limitations) and upvoting/downvoting and commenting on certain post can very heavily skew its visibility and impression of the discussed topic on “neutral” users.
so yea, karma used to be a good thing (good content motivation), became a bad thing (karmawhoring), which was still utilized for good things (participation limitation), which led to even worse things (karmafarming bots).
lemmy will have to deal with this sooner or later. there will be bots brigading communities on certain topics and there need to be some kind of indicators to distinguish honest users and trolls.
Thank you. I’ll see myself out.
The fact that corporations sees the fediverse as inconsequential. The minute people flood it in larger numbers as they are fleeing the corporate entshittified internet, that is exactly what they are going to do though.
What is stopping corpos from scraping everyone’s posts and stuff from the fediverse and train their AI?
That very real and enforceable “this comment cannot be used to train AI” crap some people add to every comment that definitely makes bots not scrape the comment, of course!
I’ll sue them in small claims court as a pro se litigant demanding a jury trial. I will also try to file motions every other week, which will probably fail and ask the judge to give me time to correct them. I will make them waste thousands on attourneys fees and be a royal pain in the ass.
Good luck getting a court to hear it when you won’t have a shred of evidence to show
But that can poison the AI to some degree.
Kinda? Not really, though. If anything it, the model’s response would just include “anti-commercial license” at the end and they’d get rid of that with further training
Most likely someone at the AI company would catch it and filter those strings out of the training data.
We just defederate.
And then they would create users in other instances to crossposts their shit, like ml and hex do.
Defederation isn’t really effective as it is right now (and I believe that’s by design).
-
Disruption: Probably ethics? I mean, I know big global businesses barely have any, but they do care about their reputation somewhat. Anyone running a botnet to destroy small/medium fediverse servers would be discovered fairly quickly, I suspect. Nothing is going to stop AI training scraping outside of regulation, I suspect.
-
Ads are enshittification. Federation is defense against it, because it prevents vendor lock-in and allows migration while maintaining your network effect. Threads already tried to join, and nearly nothing of theirs gets through. I’m on a mainstream mastodon service that doesn’t block threads, and I’ve seen a threads post only once or twice. Threads can’t display their add on my service, so there’s no incentive for them to push content.
-
What is stopping corpos from scraping everyone’s posts and stuff from the fediverse and train their AI? What’s stopping them then, to create loads of not accounts and spam / disrupt smaller communities
Nothing stops them right now. Currently they’re causing effective DDOS by scraping manually and there’s no good way to block them except by going to extremes.
In fact, I would prefer if they just used their own instance to scrape content instead of causing downtimes like they do now.
Corpos can flood the fediverse with ads and crap:
For that, the solution is simple, we can defederate.
I’m new and still trying to learn. What would defederating imply? An instance being blocked by all other instances?
Exactly. See also Gab and Truth Social.
Am I reading this right? Meta tried to be compatible with Lemmy and every server owned agreed to mass block them and leave them out?
Not every instance blocked them, but many did.
The fear of Embrace, Extend, Extinguish got a hold of the fediverse when Threads was originally announced.
Not Lemmy specifically, but the broader fediverse (and probably mostly the microblog part dominated by Mastodon and its forks)
Most of the core mastodon servers haven’t blocked threads…
You can block Threads yourself even if your instance doesn’t
I don’t know how you define “most” or “core” here, but it’s certainly true that mastodon.social and its ~400K users remain federated with Threads.
A lot of instances did block or limit them though, and I’m not going to sit down and calculate which side is in majority 🤷
Most by number of users, I’d guess.
I’m on mastodon.social, and basically never see threads posts.
Federation is where one instance “talks” to another and exchanges content. If your instance isn’t federated then you’d just be stuck with your own content and members with no outside interaction.
No, defederation just means 1 instance chooses to stop communicating with another instance.
An instance being blocked by all other instances?
No, defederating is just a single instance blocking another instance, not the entire fediverse doing so.
Facebook tried and we scoffed.
It’s called the Embrace, Extend, Extinguish strategy of dealing with threats. Microsoft have been endorsing Linux recently for this reason.
Microsoft have been endorsing Linux recently for this reason.
I doubt that’s still the reason, I think they endorse it because they’re making a shit ton of money using it on Azure.
Yes, except that you can’t really do that with open source things. If one instance/a particular piece of software gets compromised, you can always spawn a new one / fork a new project etc.
EEE was designed for open standards/projects.
For example, google talk originally used xmpp. They kept adding features that broke on the xmpp side of things, until people effectively used google talk. They then cut of xmpp, after successfully killing it.
This is not what really happend. Yes initially they added features other XMPP implementations had trouble to catch up with, but the main problem was them not implementing important security features like s2s TLS encryption, thus forcing others to cut them off. Google continued to run their xmpp servers for many years after, but they were so badly maintained and insecure no one wanted to interact with them anymore.
The rest of the xmpp ecosystem continued to grow at a slow pace and is alive and well, it was just an annoying set back going from being able to contact many millions of users on their Gmail linked xmpp accounts to not being able to do that anymore.
Every time someone brings up xmpp and how Google extinguished it, I wonder if xmpp afterwards was somehow worse of than they would have been if Google never had embraced it. I don’t know, but my gut feeling would be that Google mostly just extinguished whatever they brought in in the first place and in that case EEE would be harmless. Am I wrong? If so, please explain.
Mostly yes, but people were a bit naive back then and many onboarded friends and family onto the Gmail service and that burned some bridges and good-will as for Gmail users other xmpp users simply ended up as if offline and never responding.
Fair enough. But maybe there were also people that were brought in by Google, that actually stayed after Google left again? Well, I guess it’s hard to say 🤷♂️
That’s about the only thing they can do with open-source things.
It helps, in the case of Linux, that it’s tightly gate-kept by Linus. Now when he steps down, I will worry for the project.
We would find out. We would defederate them
Lack of interest from their part. Right now, they have nothing to gain, fediverse is “small fry”, and if the attacks could be traced back to them, they’d have to deal with the PR shitstorm
What is stopping corpos from scraping everyone’s posts and stuff from the fediverse and train their AI?
They’re probably already doing that
They are already training on the fediverse. If something is on the public web, you can assume it’s in some training data somewhere
Wish there was a way to
easilypoison the inputBut sadly this’ll just be reality. The only thing we can do is either not use it or embrace it.
i have seen others hide prompts via small text and unicode characters that make invisible text. I imagine you could also use unicode characters that look exactly like normal characters, these characters then maybe messing with tokenization or something.
the problem is that any text that will fuck with AI scrapers, will also fuck with screen readers
Ah. did not realize that, thank you
We’ve already seen AI fall for obvious jokes, like whether you can melt an egg or put pineapple on pizza.
