Most instances don’t have a specific copyright in their ToS, which is basically how copyright is handled on corporate social media (Meta/X/Reddit owns license rights to whatever you post on their platform when you click “Agree”). I’ve noticed some people including Copyright notices in posts (mostly to prevent AI use). Is this necessary, or is the creator the automatic copyright owner? Does adding the copyright/license information do anything?
Please note if you have legal credentials in your reply. (I’m in the USA, but I’d be interested to hear about other jurisdictions if there are differences)
I don’t think it really does or can do anything.
I think it makes people feel good, like they’re fighting against AI or something.
In my opinion, it just clutters up comments.
It’s crazy to me that anyone thinks it does anything. How can someone who cares enough about AI not know the controversies about OpenAI’s training data?
The people and organizations building LLMs do not give a fuck if you add that garbage to your comment or not.
Also, good luck to those people if they have to prove an AI was trained with their comment
The biggest time would be it would start to include that link at the end of every comment.
Does that mean creative commons doesn’t really mean anything? I have my website cc by sa, thinking or changing it to cc by sa no cc but I feel like companies would still take my stuff from my website.
Depends on what your goal is. Strictly speaking cc by sa is more permissive than putting no copyright notice at all, since copyright is automatic, and the cc licenses grant various permissions not contained in standard copyright. It’s just a fancy legalistic way of saying “please credit me if you use this, continue to share in a similar fashion, but not for any commercial purpose”.
So if you want people to share your work, cc by sa makes sense.
Not sure but at the very least it’s way less annoying to see it on a website than it is under every comment
at the very least it’s way less annoying to see it on a website than it is under every comment
You’re free to block those that use the license, if you find it annoying to see.
You’re free to reply to a week-old comment, too, but neither is a great idea
You’re free to reply to a week-old comment, too, but neither is a great idea
Actually, five days, not a week.
And also, sometimes its just about making a point, even if you stumble upon something later on. 🤷
What is that point?
You (in certain cases your employer) own the copyright to your creations. It’s your intellectual property. By adding a license, you give others permission to use your property. That’s just good old capitalism.
Your property rights aren’t without limit, though. What exactly those are depends on jurisdiction, but you probably can’t stop others from archiving your site for their own purposes.
They can’t take stuff from your website because piracy isn’t stealing.
.
Yeah, it’s unclear whether copyright is even relevant when it comes to training AI. It feels a lot like people who feel very strongly about intellectual property but have clearly confused trademarks, patents, copyright, and maybe even regular old property law - they’ve got an idea of what they think is “right” and “wrong” but it’s not closely attached to any actual legal theory.
those anti ai training links remind me of the “i don’t consent to facebook using my data” posts my grandma makes
That’s exactly what it is. It’s born from a fundamental misunderstanding of how copyright law works. It’s basically just a Facebook chain letter.
That, plus a healthy dose of fuck copyright.
In the vast majority of countries, everything written down is automatically copyrighted by default and if you want to release it into the public domain or under a free license you have to make it explicit.
It’s not really fully determined whether you can actually release something to the public domain, since the “public domain” is not a legally sanctioned entity. It’s just the name we use for things that are uncopyrightable or otherwise not copyrighted (like certain government works, or works old enough that the copyrights have expired). The CC0 license from Creative Commons gets around this by waiving all copyrights instead.
This waiver nullifies and voids all copyright on a work. It also provides a fallback all-permissive license in case the waiver is deemed legally invalid. In the worst case that even the license is deemed invalid, the license contains a promise from the copyright holder not to exercise any copyrights he/she owns in the work.
- https://en.wikipedia.org/wiki/Wikipedia:Granting_work_into_the_public_domain
I’m writing this response mainly for the purpose of bringing it to the public domain. Feel free to screenshot, copy, and distribute however you see fit.
This comment is copyrighted and you are now committing piracy by reading it.
Pay the fee on your way out. Thank you.
Bold of you to assume I wasn’t already committing piracy before reading it.
.
does adding the copyright/license information do anything?
Not a lawyer, but I’d be sore amazed if “your honor, he copy/pasted my Lemmy comment” flies in court, regardless of your copyright status. The same goes for those AI use notices–they’re a nice feel-good statement, but the scrapers won’t care, and good luck (a) proving they scraped your comment, (b) proving they made money on it, and © getting a single red dime for your troubles.
You’re better off just pasting this guy into every comment to poison the well.
🌕🌕🌕🌕🌕🌕🌕🌕🌕
🌕🌕🌕🌕🌕🎩🌕🌕🌕
🌕🌕🌕🌕🌘🌑🌒🌕🌕
🌕🌕🌕🌘🌑🌑🌑🌓🌕
🌕🌕🌖🌑👁🌑👁🌓🌕
🌕🌕🌗🌑🌑👄🌑🌔🌕
🌕🌕🌘🌑🌑🌑🌒🌕🌕
🌕🌕🌘🌑🎀🌑🌒🌕🌕
🌕🌕🌘🌑🌑🌑🌓🌕🌕
🌕🌕🌘🌑🌑🌑🌔🌕🌕
🌕🌕🌘🌔🍆🌑🌕🌕🌕
🌕🌖🌒🌕🌗🌒🌕🌕🌕
🌕🌗🌓🌕🌗🌓🌕🌕🌕
🌕🌘🌔🌕🌗🌓🌕🌕🌕
🌕👠🌕🌕🌕👠🌕🌕🌕We’ll all do it and then the AI will learn and do it too, but it’ll be too late for us to stop, having become a custom ingrained in the population.
🌕🌕🌕🌕🌕🌕🌕🌕🌕
🌕🌕🌕🌕🌕🎩🌕🌕🌕
🌕🌕🌕🌕🌘🌑🌒🌕🌕
🌕🌕🌕🌘🌑🌑🌑🌓🌕
🌕🌕🌖🌑👁🌑👁🌓🌕
🌕🌕🌗🌑🌑👄🌑🌔🌕
🌕🌕🌘🌑🌑🌑🌒🌕🌕
🌕🌕🌘🌑🎀🌑🌒🌕🌕
🌕🌕🌘🌑🌑🌑🌓🌕🌕
🌕🌕🌘🌑🌑🌑🌔🌕🌕
🌕🌕🌘🌔🍆🌑🌕🌕🌕
🌕🌖🌒🌕🌗🌒🌕🌕🌕
🌕🌗🌓🌕🌗🌓🌕🌕🌕
🌕🌘🌔🌕🌗🌓🌕🌕🌕
🌕👠🌕🌕🌕👠🌕🌕🌕It has begun
Not really, it’s just tropicaldingdong pasting their fursona for the second time
On top of all comments generally being copyrighted by their author automatically, the licence at the bottom of a comment is like a no trespassing sign. The sign itself doesn’t stop people from trespassing. You still need to call police when someone trespasses. If you never call police then the sign is literally useless.
The licence is the same thing. If someone includes it at the bottom of all their comments, but never launches legal action when someone violates that licence agreement, then it’s literally useless. Given that launching legal action is incredibly expensive, I highly doubt the people using these licences will ever follow up. Also, how will they even know? How will they know a company used their comment as training data for their commercial AI? How are they going to even enforce the terms of the licence?
This is a really good point. If someone did violate your copyright, you have to enforce it. Almost no one is going to do that, so it’s effectively not copyrighted.
There’s a lot of “you couldn’t have been murdered because that’s illegal” thinking that somehow putting up a license on your posts stops these AI companies from scraping.
If someone includes it at the bottom of all their comments, but never launches legal action when someone violates that licence agreement, then it’s literally useless.
Well, its ‘poisoning the well’. What happens next depends.
For AI companies that actually honor licensing, or are fearful of getting caught at some point, they’ll honor/follow the license for the content.
And for those who do not, if they get caught with their hands in the cookie jar, Creative Commons (and other license creators) will have something to say about it. And they will get caught, we all know about black-box programming their models from the outside via our comments.
Finally, Congress right this second is considering new laws about this, so you never know. Companies in the future may be forced to have to explicitly state where the content comes from that they train their AI models on.
As far as wasting my time, all I do is copy/paste this one line of text via a macro keypress …
[~Anti~ ~Commercial-AI~ ~license~ ~(CC~ ~BY-NC-SA~ ~4.0)~](https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode.en)Its a momentary thing, so no effort at all.
The creator is the automatic copyright owner, or in some cases their employer. Copyright is automatic through international treaties like the Berne convention. The Berne convention is from the 19th century and was created by the authoritarian european empires of the time. The US joined only in 1989. I think your question shows that the idea has not fully taken hold of the public consciousness. Automatic copyright is now the global norm. (I always wonder how much its better copyright laws helped the US copyright industry to become globally dominant.)
Very short and/or simple texts are not copyrighted. IE they are public domain.
Adding a license statement gives others the right to use these posts accordingly. It only serves to give away rights but is not necessary to retain them. The real tricky question is the status of the other posts. I’d guess most jurisdictions have something like the concept of an implied license. Given how fanatical some lemmy users are on intellectual property, not having it in writing is really asking for trouble, though.
What such a license means for AI training is hard to say at this point. The right-wing tradition of EU copyright law gives owners much power. They can use a machine-readable opt-out. Whether such a notice qualifies is questionable. However, there is no standard for such a machine-readable opt-out, so who knows?
US copyright has a more left-wing tradition and is constitutionally limited to certain purposes. It’s unlikely that such a notice has any effect.
IANAL. However I know a bit.
which is basically how copyright is handled on corporate social media (Meta/X/Reddit owns license rights to whatever you post on their platform when you click “Agree”).
Yes, this is how it works. You give them a license to your posts.
I’ve noticed some people including Copyright notices in posts (mostly to prevent AI use). Is this necessary, or is the creator the automatic copyright owner?
The creator automatically owns the copyright. People can put in license terms, but they’re effectively useless in this context. Let’s say OpenAI violates the copyright on your post (it’s still an open question whether or not training AI on copyrighted data constitutes copyright infringement, but we’ll assume it does). Your only recourse is to sue them if they do this. Because you never registered the copyright, you’re limited to recovering actual damages – if you do register the copyright you can get statutory damages, which are up to $150k per violation. So how much money did you lose on the ability to commercially exploit this post that OpenAI took away from you by copying your posts? Less than the cost to bring the suit, I’m sure.
So the TL;DR here is that the anti-AI licensing thing is only effective if you’re registering the copyright on your posts/comments. And even then, that’s only true if AI training is considered to be copyright infringement.
I don’t think it exists at all on the fediverse. I’ve talked about it before, not a lawyer, but from a technical standpoint I don’t know how anyone can claim copyright.
All fediverse apps start on your instance, you write a post. Great, maybe there’s a disclaimer there. But then it’s shotgunned out to literally anyone or anything that’s listening. Other instances, governments, corporate, whatever. You’re literally giving it to anyone who would listen.
So to me copyright is like saying “only people I approve of can look at this sign” and then posting that sign on every tree and post in town
Copyright isn’t about who can look upon something so much as who can reproduce it. However, due to the way federation works, it has to be assumed that fediverse users are agreeing to allow anyone using the protocol to reproduce their “works”
Copyright is more than that, like who is allowed to make commercial use of a given work. Just because something is written down in a public forum doesn’t give everyone free rein to do whatever they want with it under copyright.
But the fediverse isn’t them taking that data, it’s you giving your data to them. It’s you placing your data directly on their server.
It’s more than my flyer metaphor, it’s you literally placing your flyers in someone’s house and then saying “but you can’t do x y or z”. Even if morally you are in the right, how would you ever enforce that or prove something in court? You still have the hurdle of “if you didn’t want them to have it, you shouldn’t have handed it to them”
Uh, copyright always works that way. You’re not supposed to make copies of movies most of the time, but people do, and it is virtually unenforceable as long as they take basic precautions. One of the only times it is reliably enforceable is when a business tries to make money off of your work and you can sue them.
Movies though have a license that you accept, and it comes encrypted. License is on the box that you are allowed to watch it at home, with so many people, on approved devices. By buying it you are accepting the license. Even modern day blurays have a license, and they can actually revoke the license (by pulling the encryption keys). If you don’t approve of that license, you simply don’t buy it. (However people get around it and as you said it’s unenforceable).
Whereas Lemmy and fediverse you’re giving your stuff out license free to anyone, and any other server can have their own terms. Such as “By giving me your data you are giving it to me license free, and remove yourself of all ownership.” Unless it’s specifically defederated, well there’s no way of you knowing and so you give your data. What does a judge say in that case? You said not to use it, but you put it on a server that said they can use it however they want.
I agree with you mostly, just pointing out the slight differences on how movies and studios get around that little hurdle of “ownership”. (i.e., we don’t own it)
Sure, but I would say that there’s a reasonability test there. Like, I could have a posted note somewhere on the internet that says, “by allowing my computer to download your content you grant me full license to use your intellectual property for any purpose in perpetuity throughout the universe,” but that doesn’t make it binding on anyone. Federation means other servers pull the data, so you don’t have control over it, so you can’t be considered to agree to a random server’s terms.
The same thing happens when you download a website. The website always allows others to download its content, but that confers no license no matter what anyone else says.
“License free” doesn’t mean “free license”, it means the opposite. No explicit permission is granted.
ah but an important distinction. The servers aren’t going out and asking for data from other servers. Federation means instances push data to listening servers. It doesn’t sound like much but it’s an important difference when we’re talking about it. So for me, I view that as a whole different thing, because by pushing data you’re saying “I don’t care who is listening, I’m sending it anyway”. If it were a pull model then it would be like what you are saying “Hey, I only give you access to this on my terms”. By pushing, you remove your server and it’s rules completely.
and that’s why I keep going back to my imaginary court. If you’re trying to tell a judge that “They shouldn’t have used it to train AI/write a book on, I didn’t want them to do that” the obvious next question is “Well, why did you give it to them then?” They didn’t take it from you, you gave it to them.
I think this is a very fine distinction that would have to be settled in court and could go either way. I can only say what I think it should do. And to be honest I think copyright is garbage, but for it be consistent I don’t think that this difference should matter.
I think an important distinction for me with federation though is that it’s not just a push, you have to subscribe, so it’s a two way street. It would be similar to an RSS feed, and I’m not aware of that having any particular implications for copyright. There is certainly no explicit acknowledgement of terms baked into either protocol, so I think the only reasonable conclusion should be that it doesn’t impact copyright either way. That remains unlicensed and subject to the normal rules, which presuppose that permission is not granted.
Whereas Lemmy and fediverse you’re giving your stuff out license free to anyone,
- Unless you attach a license to your content.
- Even if you don’t license it explicitly, there’s default implied copyright law on how your content can be used.
.
and any other server can have their own terms. Such as “By giving me your data you are giving it to me license free, and remove yourself of all ownership.” Unless it’s specifically defederated, well there’s no way of you knowing and so you give your data.
If you license your content, that license travels with the content, and has to be honored on other servers (Federated or otherwise).
The license is with the content, and not the server the content is first posted to.
What does a judge say in that case?
“Other server owners, did you follow the license that the content is licensed with? No?” <smacks them with the gavel>
You said not to use it, but you put it on a server that said they can use it however they want.
Lemmy.World’s TOS does not claim ownership of our content that is posted/shared to their server. So they can’t use it however they want, they do not own the content, each individual poster still does.
And they don’t want to own our content, as that’s one hell of a ‘safe harbor’ law exposure/risk for them, if they start to do that.
I’d argue that isn’t settled yet. Take this. I run my own server, I don’t want your “licensed” comment. I can go and add that to my ToS right now, that anything you give me will be trained. I say in there that I will disregard any license on there, and by placing anything on my server you are relieving yourself of any license. Since it’s my server, I can say that. “If you don’t want your data trained on, don’t put it on my server.”
So, I don’t think it’s as simple as you make it out to be. It’s exactly the same as Facebook’s ToS, they state in there that by using their service and putting data on their servers that you allow facebook to use that data however they see fit. Why is me doing the same thing on my server any different?
So, to use your own format
“Other server owners, did you follow the license that the content is licensed with?”
"No, you’re honor, you can clearly see on our homepage that it’s stated that any and all licenses are lost when they give us information.
Did they accept any ToS saying as such?
“That’s irrelevant, our legal terms are clearly stated on our website, if they didn’t want their information shared to us, then they shouldn’t have shared them with us.”
This format isn’t well designated in the courts, there are no real precedents for the fediverse or how it works. You’re arguing that it should work that way. I’m arguing that how it should work is irrelevant, and right now there is nothing stopping anyone from using unencrypted data given to their server in any way.
What could happen, but isn’t really set up at least on Lemmy yet, is that when one server federates with another the receiving server sends a ToS/license that is server wide, forcing the subscribing server to accept or to not federate. I think that would shore up gaps in the law here, because in your example they could respond with “Your honor, we gave them terms and licenses to subscribe to our updates, and they accepted”. I also think that then would be required for new users signing up, to see how data is licensed from other servers. If this were a github issue I’d back it 100%.
However, in both scenarios, both current and what I’d like to see, I don’t see that adding a license at the bottom of your comment will ever hold up in court.
(Of course I’m not actually doing that, this is all a thought exercise, but I do 100% guarantee someone is just accepting all of our data and using it, license linked or not)
I can go and add that to my ToS right now, that anything you give me will be trained. I say in there that I will disregard any license on there, and by placing anything on my server you are relieving yourself of any license. Since it’s my server, I can say that. “If you don’t want your data trained on, don’t put it on my server.”
So you’re setting up a straw man by adding the TOS clause of ownership on the posting server to your example. I’m not saying that. The issue being discussed by me in response to your comment was if a license on content that is being federated stays with the content or is somehow magically stripped off when its federated.
As far as your TOS example goes, If Lemmy World added to their TOS that any content added to their site they own, then I wouldn’t post any content on Lemmy, as I want to keep ownership of my content.
But since Lemmy World does not do that (smartly so for safe harbor reasons), then the creator of the content is the owner of the content, and if they license that content it carries forward as the content is federated. Its up to the receiver of the federated content to reject the content, or abide by its licensing.
Copyright is more than that, like who is allowed to make commercial use of a given work. Just because something is written down in a public forum doesn’t give everyone free rein to do whatever they want with it under copyright.
This.
In legal terms, what does that mean? Every post is presumed to be public domain? Or in the process of posting, is there an implied license to generate unlimited copies for the purposes of federation? If someone likes a post and decides to make it into a chapter in their book, which they sell, is the original author entitled to attribution? To compensation?
I would argue that yes, you’re posting publicly on a public forum, whose contents are shotgunned out to any listening servers/apis/whoever.
If this were in a courtroom, I’d expect the defense to say that the poster chose to post it on a public forum which was then shared with whoever was listening, that there was no way to expect it to remain private, and there is could be no assumption of privacy with the way it was shared.
For enforcement, there is no way to enforce any sort of licensing with the fediverse model, you handed your post to me, if you didn’t want your post handled in a certain way then the response is “Why did you hand it out in the first place?”. If someone did make your post into a book, then it’s on you, the poster, to make the case that what they did was wrong, and I think it’s enough of a grey area here to say that they were simply listening. To flip it around, what if their server has posted terms saying “Anything you give to us will be used for training and publishing.” You sent it out to anyone listening, they posted their terms, who is right then?
This is different from normal social media where you posted to a walled garden, where you’re bound by just their terms. Now any server can have any rules or terms, and we’re blasting our data out to all of them (unless they are explicitly defederated)
All fediverse apps start on your instance
You forgot about the web client.
You’re literally giving it to anyone who would listen.
The quantity of sharing does not dimish the licensing of the content.
So to me copyright is like saying “only people I approve of can look at this sign” and then posting that sign on every tree and post in town
I mean, ProPublica has explicit instructions on how to share their content with others, content that is licensed with a Creative Commons license, and that includes displaying the license number, when you share the content.
I don’t know.
But minimum royalty laws should exist.
Creator owns the copyright.
This Stack Exchange answer has some potentially relevant info. One notable excerpt:
The short answer is ‘it depends’. […]
It depends on:
- whether the code is eligible for copyright,
- what license the content of the particular forum is under, and
- what additional license (if any) the individual contributor has put it under.
Well first thing is that the license is a copyleft license so it is still allowed to be used, distributed, etc. the only real difference between this license and public domain (as far as I know) is me saying that I don’t want it being used for commercial purposes that’s it.
Also for me its more just a way for me to say fuck you to everything having to be commercialized so even if it doesn’t hold legal water I don’t care.
Right but if they use your content anyway and you find out (and that’s a big if, because it’ll just disappear into some AI data set and you’ll never see it again), what are you going to do? Sue?
They just said it was a statement of principle as much as anything.
Right but in order for it to have any meaning it has to be at least be theoretically enforceable. This is obviously so not enforceable I don’t think they’re going to care about it. So it don’t do anything.
what are you going to do? Sue?
Personally? I let Creative Commons know what’s going on, that their licenses are being ignored.
I’m pretty sure they’d have something to say about the matter.
There’s a bit more to it than just that
BY - attribution is required
NC - as you said, cannot be used for commercial purposes
SA - Share Alike -anything using it must be shared under a similar license.
Ah well then I might try and find a license that doesn’t require attribution because I don’t care about that part. But the rest seem exactly what I’m going for.
Edit: grammar
Ah well then I might try and find a license that doesn’t require attribution because I don’t care about that part.
I would argue attribution is also really important, as it forces them to expose publicly how they’re training their models, bringing awareness.
