Artists put content on the internet so you can look at it and share it with your friends and say “wow cool art” not so you can run it through an art generating model to put them out of job. That we never anticipated this possibility means you call fallback on the “fair use” argument but legal does not mean ethical, especially when the law has yet to catch up to changing times.
But they are not simply sharing a single link and the “intent” is different. I personally feel that we should not treat human learning and AI learning the same. Unlike humans, the AI is essentially immortal and infinitely copiable.
Selling pipes, cellphones, igniters, and black powder is perfectly legal. Selling pre assembled pipe bombs filled with black powder and a remote detonators is going to very quickly get you into a great deal of legal trouble.
2. The artist made the photo publicly available.
3. Read this opinion and come back to the conversation with a more level head instead of making analogies to promoting terrorism: https://lawreview.law.ucdavis.edu/issues/53/5/notes/files/53...
2. The artist disabled crawling on their website so it’s accessible but not fully available.
3. Your link says many things that aren’t settled in terms of US law but rather the options of the author. “even if infringement occurs during machine learning, training AI with copyrighted works would likely be excused by the fair use doctrine.” That’s a lot of possibilities not actual guarantees.
However, it ignores the largest issue namely if the output of these AI’s are all simply derivative works.
3. That's why it's called an opinion. No shit. Maybe instead of talking about promoting terrorism, you should have a level head and talk about fair use.
Something people don’t understand is “Right click file save as…” on a copyrighted image breaks copyright, as you don’t have permission to make a copy.
In all seriousness, that's total shit. It's publicly accessible. That's the permission. There have been different cases like weev where the judge has viewed it differently because of specific details like carrying out the action of enumeration, or guessing IDs, to reveal non-public information, but that had to do with the CFAA.
If you reproduce content in total in another work, resell, etc., that's another thing.
Fair use is permission. Every library book is accessible that doesn’t have copyright implications.
In the context you're now talking about by the letter of copyright law, downloading a photo, image, file of something or other, a "copyrighted work" where it's publicly accessible, and has a particular license specified that you may not reproduce it, may technically be "unlawful" letter by letter of the law, but I doubt any judge is going to actually see it that way versus intent of the law unless you literally reproduce or share a substantial portion of it, sell a complete copy, etc. It's almost certainly fair use to study, and use portions of the copyrighted work in your own copyrighted works.
As to making a copy with intent to X, that’s what fair use is. A student may photocopy the full text of an short article so they can accurate quote it in their term paper. They can’t simply photocopy an article because they are a student with access to the article and a photocopier nor can the photocopy a full book because they want to use a short quote in their paper. Copying incidental to acceptable use becomes retroactively acceptable. This distinction may seem crazy to you, but the fact that intent matters means you can’t judge an action without context.
Fair use in commercial context is looked at with vastly more suspicion than fair use in academic context, which again demonstrates specific actions on their own aren’t always enough information to say if something is allowed.
It's been determined that it's legal, if it's publicly accessible, and you don't receive a cease-and-desist letter telling you to stop. If it's public, that's permission. Simply "turning off scraping" by putting a robots.txt there doesn't make the content and linked images of a public web page any less public and restricted from being scraped, legally.
Just last year, the LinkedIn case: LinkedIn had a robots.txt, and the judge didn't give a fuck. Nor did he care what their terms of use said. Rather, it was hiQ's continued scraping of LinkedIn data even after LinkedIn's cease-and-desist letter to them that constituted access of data "without authorization."
>As to making a copy with intent to X, that’s what fair use is.
Yes, and?
>This distinction may seem crazy to you
It's not a crazy distinction, that's what I'd basically said previously, so perhaps we're talking past each other.
There’s a lot of precedent here showing scrapping isn’t guaranteed to be acceptable.
https://en.wikipedia.org/wiki/Facebook,_Inc._v._Power_Ventur....
$79,640.50 in compensatory damages + $39,796.73 discovery sanction
I could go on, but I am not sure what exactly you’re trying to argue here.
>I could go on, but I am not sure what exactly you’re trying to argue here.
That public is public. If you leave the door open in the real world, someone CAN enter your home. If you host your image on a public webpage, they can scrape it. robots.txt is not a security measure, nor is it a contract that magically gives the right to scrape where it wasn't given previously, it is a gentleman's agreement that you can ignore if you want to be a dick about it, and know about the robots.txt. Ethically, that's wrong, but it is how it is.
Not to mention, as I came to understand while reading during this discussion, LAION wasn't even crawling: they were using a public commoncrawl dump to gather their images. commoncrawl had crawled the author's site previously. They just took that data and got image links out of it.
1. they weren't selling a dataset
2. the artist didn't "disable" scraping in any meaningful way, legally
3. linking to the image is not illegal, and they're justified to respond with an invoice in Germany to recover legal fees for this dumb copyright complaint
4. it may fall under fair use to download images and train neural nets on them, it may not be. it always depends on the context and the specific case.
pretty simple stuff.
No trespassing signs have legal weight even without a fence.
Read up on Thomson Reuters v Ross Intelligence.
[0] The answer at least in the US is the act itself is likely not illegal although it can be construed to lead to a crime, such as incitement.
Now, let's try something similar. You have a monkey you want to train to paint. You download an image of a painting and give it to the monkey to train it on, amongst many. The monkey, in the future, may paint something that has vaguely or substantially similar qualities to the original painting. Has copyright infringement occurred in this instance?
Let's go further: You have a neural network you want to train to paint. You download an image of a painting and give it to the neural network to train it on, amongst many. The neural network, in the future, may paint something that has vaguely or substantially similar qualities to the original painting. Has copyright infringement occurred when that happens?
I find it hilarious that in this thread, instead of asking the question of whether this is fair use, multiple posters have jumped to comparing this to thievery, promoting terrorism and posting of private information as an analogy to harm.
https://en.wikipedia.org/wiki/Monkey_selfie_copyright_disput...
The neural network behind the LLM is "trained." Yes, it's a term of art, very observant. That doesn't change the fact that there's not a substantial reproduction of a work in it. LLMs "learn" to predict text by feeding them vectors, the resulting "weights" of that process could be thought of and implemented as a big walkable decision tree to predict the next token it weren't for combinatorial explosion, but there's no substantial reproduction, or "copy", in there.
Is there a substantial reproduction, in your head, of a song or a text or a painting you can perfectly recall and can't seem to forget? Be careful, the copyright police will get you, the only way to delete it is to delete yourself!
>feeding into a machine is called copying
Copyright law deals in "reproductions" as far as I understand it. Feeding a copyrighted document as vectors into a program that reconfigures some matrix of vectors, and then using that matrices to output probability distributions into something tangible, where something tangible isn't a complete or substantial reproduction of the copyrighted work, is not a copyright violation.
If you think it is, amend copyright law, or take it to court and let a judge settle the matter, and hope it's in your favor.
About your question: """Is there a substantial reproduction, in your head, of a song or a text or a painting you can perfectly recall and can't seem to forget? Be careful, the copyright police will get you, the only way to delete it is to delete yourself!"""
This equivalence between "perfect" human recall, and a copy of the data input into an AI model is a bit of a strawman I think: it is the distribution of copied information that copyright protects against, the information has been uploaded in vectors, have fun demonstrating it is not being used in producing this or that model output. If I learn a popular song and do a cover, there is a copyright law for that, I owe rights.
The "vectors" are not "uploaded" or "copied" into a file or neural network, they're transformed. In the context of stable diffusion, They're transformed, progressively "noised" or "corrupted", "diffused" with random gaussian noise, and in the context of stable diffusion, it's "trained" and "learns" how to "denoise" various "noised" stages of images represented as a vector of pixel data into their original form.
Then, when it comes to generation of images consistent with an "annotation" or "prompt", it is "conditioned towards" or "biased towards", with more "training", by noising an input image, and concating or combining that vector of pixels with a vector of the annotation of the image. It then "learns" to denoise with that conditioning information, the annotation.
Then, you can take the trained model, and do the same thing, with just a text prompt as a vector concated to a vector of random gaussian noise, and no input images.
That's basically and very simplistically how it works.
The output is not a substantial reproduction from the input images + annotations when trained. It takes the random noise, and "tries" to denoise it into something consistent with the prompt with conditioning to guide it.
Your attempt at covering would be a substantially similar reproduction. Your goal is to do a reproduction. Whereas, the model "learns" to generate images consistent with an annotation/prompt, by conditioning it with that "goal" on top of how it "learned" to denoise the images.
* In the Pirate Bay case, the copyright owners hadn't posted the copyrighted contend anywhere. Someone who was not the copyright owner posted an unauthorized copy, and the Pirate Bay were linking to that unauthorized posting.
* In this case, the copyright owner has posted the content; and LIAON is linking to the copyright owner's authorized posting.
Here, they are linking to the data that is legally hosted with extra metadata. This is exactly how we wanted the www to work.
Then again, Google guys said, we are just linking to the data...
Your comparison would be more reasonable if LAION had an origin link and then a link to a copy of the file explicitly for the purposes of copyright infringement.
If the artist doesn’t want people and algorithms to see the image, then put a password in front of it. Or add no-index to robots.txt.
This is like driving around with a sign on your car and complaining that people are looking at it and imagining the image in ways you don’t like. If you don’t want people to see the image, don’t put it on a sign and drive around.
There's a bit of nuance in this case, but in general, yes, you need a licence.
There is no law that says you cannot train AI models on copyrighted data.
Also, It is legal to make copies of copyrighted materials so long as you don’t distribute them.
Also you do not have to preserve the original for copyright to apply: any extract is only allowed under well defined fair use rules. A GAN's purpose is definitely not parody, commentary, ...
That is not a definition that people use, nor the legal system
> Taking data and putting it into any form of storage, ie weights in a model, is called copying. Producing an output out of the gathered data is then a step in distribution.
It isn’t though.
If I train on Batman and I produce Spider-Man, the Batman copyright holder cannot sue me. It would be dumb to suggest otherwise.
This will see its day in court, and clearly indeed the meaning of the verb "train" will feature prominently in the debate.
I tend to side with reliance on dictionaries for the meaning of words, and in my understanding the courts also do.
If you load up only batman pictures on your model you will clearly only produce batman pictures. No one is being dumb here that does not want to be, I am sorry.
Perhaps. Or, more likely I think, your argument is weak and doesn’t contribute meaningfully to the discussion.
Personally, I downvoted you do to my decades old rule of “downvote messages that complain about downvoting.” I don’t always downvote whiny comments, but I always downvote whiny comments about downvotes. (and I will be downvoting my own message as I am both pedantic and reliable).
That’s not what I’m claiming though. I’m saying I take a model that has been trained on Batman and other things and I produce Spider-Man. If your point of view is to believed, the Batman copyright holder can sue me for producing a work that contains Spider-Man.
That’s definitely how copyright works.
I don’t think training on these has been tested, but I expect it will be allowed, otherwise how can a web browser receive, cache, and render a copyrighted photo? I think the closest is there were some lawsuits about google caching websites that contain copyrighted material. I assume they worked out in Google’s favor as google (and others) cache web sites even though they contain copyrighted material. The exception I remember is there are some specific rulings and laws governing how news is treated.
"If the image is available publicly, without restriction" — is irrelevant.
Why quickly skip past the most unjustifiable part? Because you're about to use a human metaphor (like "see") for an algorithm. My algorithm is "copy." So instead of driving, it's like selling a book with a sign on the book that says "copyright 2017, all rights reserved" and saying that people can't copy it. If you don't want people to copy it, don't write and distribute the book.
I actually believe this, but I don't know how somebody believes in one algo and not the other and remains consistent.
LAION doesn't do the SD training themselves though. This just provide links with annotations.
[1] Exception: The Ephemeral copy that is made when actually downloading the image. This is permitted by EU law. AFAIK after the image has been analyzed for certain properties like "red" or "traffic light" or "bus" or "vertical line" .. .etc... it is then discarded.
[1] https://lawreview.law.ucdavis.edu/issues/53/5/notes/files/53... with thanks to https://news.ycombinator.com/item?id=35713309
[2] https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX...
[3] https://www.gesetze-im-internet.de/englisch_urhg/englisch_ur...
"""Article 5
Exceptions and limitations
1. Temporary acts of reproduction referred to in Article 2, which are transient or incidental [and] an integral and essential part of a technological process and whose sole purpose is to enable:
(a) a transmission in a network between third parties by an intermediary, or
(b) a lawful use of a work or other subject-matter to be made, and which have no independent economic significance, shall be exempted from the reproduction right provided for in Article 2. """
As upload to a GAN is not transmission, and the economic significance is pretty clear.
Then the German 44a is an exact copy of the EC directive, so this also applies I think.
As the word "train" refers to applying the uniquely human capability of acquiring information, the word "upload" seems the most appropriate term fallback to use when referring to putting information into some AI system.
You're welcome!
Reference: open any dictionary and look up the word "train". Choo choo!
However, I get the impression you're feeling somewhat frustrated right this moment. Not a problem. Would you like to come back later and edit your post? If you do (and/or reply to this), I'll reply in more detail and provide sources where possible.
Dogs are trained to do certain things, are dogs humans?
In the sense that yes, the bits are copied from one register to another. But they aren’t stored permanently, nor are they distributed. So there’s no copyright violation.
Similarly my brain is making copies when I look at something. And routers on the internet make many copies sending the material to me. Etc etc.
If training on an image is copyright infringement for making copies, then so is viewing the image through glasses (image is transformed by the lens) or dragging the image from one monitor to another or caching the image in your browser and viewing it later from the cache rather than retrieving it again.
Where have they reproduced his content?
You don't need a license for your fair use of something. Can you walk me through how downloading a photo to your personal computer and studying it, perhaps to become a better artist is copyright infringement? Now let's apply that analogy to training a model, is the process of using a particular image to train a model, ultimately reconfiguring a bunch of weights in a matrix, so it "learns" whatever particulars it learns, copyright infringement? Is training a painting monkey that may paint something similar to a copyrighted painting, on a copyrighted painting, copyright infringement?
I personally do believe that training using copyrighted data is _not_ a violation of any existing copyright, but using the trained AI to reproduce and distribute something that an expert in the field would regard as a copy is a violation.
FWIW I do believe training a model is fundamentally different from training a human.
The obvious analog is facial recognition, deemed privacy violating by laws and courts in multiple jurisdictions, vs a human doing the same thing manually.
Until we pass 21st century laws, thats what applies.
It was deemed "privacy violating" under specific privacy laws within those jurisdictions.
i.e. go pass a law amending copyright law if you think so. there's no substantial reproduction here, in the context of stable diffusion, until there is, and then yeah in that instance it's probably a copyright violation, like a substantial reproduction made by a human would be. but training itself, probably not. but yes, there have been no rulings on it in this context.
> go pass a law amending copyright law if you think so
I don't need to do anything. Let's see until one of these cases come to the courts. You're talking as though it's a foregone conclusion that they'll rule the way you think. I wouldn't be so sure myself.
Do you have a link to a court decision to that effect?
How, specifically, is "training an AI" different from "training a person" here?
Training(2) software is to training(1) a human what firewall is to an actual wall made of fire, or what killing a process is to killing a human. The word is a synonym—more precisely, industry slang that happens to resemble a word used outside of the industry due to superficial similarity.
So if you wonder how training(2) software is different from training(1) a person, first understand that it’s a different word that never signified the same concept.
Subsequently, if you want to argue that training(1) should apply to software now, the onus would be on you to explain why and how so. You should be prepared to argue that software is sufficiently like a human—you know, that it understands ideas, has agency and free will, thinks like a human and has human-like consciousness, is capable not only of performing instructions given to it by its operator and act as a tool but to actually consider its actions and make own moral judgements, things like that.
And if you come up with satisfactory evidence in favor of all that, and have grounds to believe some software is enough like a human that training(1) applies to it, then why are you fighting to allow its operator to ignore copyright—and not for more important things, such as to free this human-like being from abuse by its operator and grant it basic human rights? If we imagine that software is like a person enough that “training” it means the same thing as training a human, we should be prepared to acknowledge all the implications that come with that.
The reality is that there are companies who would like us to both believe their software is human-like (so that we don’t sue them for rampant copyright abuse) but also not at all human (so that we don’t demand them to stop profiting from what would be slave labor). Naturally, if they pick one or the other they stand to lose from “a lot of money” to “entire business model”—but we should help them make that choice.
How does that change the fact that there is no substantial reproduction of the works in the resulting weights, or in the "output" of these models with the weights?
Can you point to me where it's illegal to take copyrighted works and "train" neural networks on them, as long as there's no substantial reproduction of the works in the output of that process, or in the output of a particular configuration of "trained" and "frozen weights"?
However you define it, training typically doesn't involve making permanent copies of the data.
This means that Under EU law -as far as I can tell- it is probably legal. Under US law the different circuits have a slightly different interpretations of the law, but probably would agree that this is fair use.
<Training> by any definition you care to give it does not rise to the level of copyright violation (In the case of training eg stable diffusion, on average only a couple of bits worth of "notes" are stored per image. If that's a copyright violation then pretty much anything would be). The main issue -believe it or not- is actually the part where images get temporarily downloaded.
As stated: explicitly legal in the EU. May need some work in the US.
You are talking about two different words, again. Human learning is not a copyright violation. Machine learning is. Machine is not human.
> The main issue -believe it or not- is actually the part where images get temporarily downloaded.
No, the main issue is using a tool to sell derivative works automatically generated from copyrighted original works. If images/text is not stored it wouldn’t change a thing.
If images/text being stored or not is not part of the argument, then what IS the argument that AI works are derivative? Is it only certain works? All works? Is it automatic? Does it require human intent? How do you get from A to B here?
And of course there is human intent, what are you even talking about? This is law. Law is sort of centered around human actions and intent plays a big role. In this case, operators fully intended to scrape copyrighted works, feed it to this tool and operate it for profit (because money smells good).
“Our client fundamentally understands that your client may not like the temporary reproduction of his works,” LAIOn lawyers wrote to Kneschke’s. “Only this has been expressly permitted by the European legislator.”
Which is exactly what happened with stable diffusion - say what you will about it, it’s distributing the ability to create value from AI among a larger number of benefactors than OpenAI
Used? Yes. Reproduced? I don't think that necessarily follows.
> I don’t know, but to claim that they’re just storing a url is bogus.
They're definitely not just storing an URL, but claiming that what they're storing is some kind of "copy" seems equally bogus.
People don't seem to be getting quite yet what 'artificial intelligence' intrinsically really means.
But they will.
https://www.alltageinesfotoproduzenten.de/2023/04/24/laion-e...
He sent an initial demand that LAION declined. He did not want to give up, then he made his lawyer send a letter, then LAION paid a lawyer to answer, and the invoice is for the cost of the lawyer.
What I find interesting is that the photographer admits in the comments that he sells thousands of AI-generated images, probably made with an AI trained on LAION's dataset. He only doesn't want links (with alt text) to his own images in the dataset.
If I understand it correctly:
0. Laion uses WAT-files from Common Crawl to generate a list of suitable image-text pairs, including a URL to the raw image. [0] 1. Laion downloads the raw images for further processing. 2. Based on step 2, Laion creates a curated list of image-text pairs. The images themselves are uncurated. 3. Using [1], anyone can download the images. Tags such as `noai` are respected.
-- Question 1: Under German law, is Laion allowed to do what they are doing?
Maybe, but far from clear.
The lawyers [2] point to § 44b UrhG [3], which is rooted in [7], and § 66d UrhG [4]. These norms state that you can use publicly available data for data mining, especially if you are doing it as part of scientific research. Copyright holders may prohibit such activities but must do it in machine readable form on the website.
First, I am not positive that Laion's activities fall under the law's definition of "data mining". The law says data mining is an activity done in order to gain insight into "patterns, trends and correlations". But Laion does not even attempt to gain any insight from the data, it just wants to provide a data set for model building [0]. I don't think § 44b UrhG is applicable at all.
Second, Laion cannot simply yell "it's for research". To use of § 66d UrhG, you (a) must qualify as a research organisation, and (b) there must not be any private organisation in the background directing the research. Whether or not Laion qualifies as such would have to be decided by the courts, but I think it's questionable.
Overall I think the lawyers' position as outlined in [2] is not very strong. Perhaps they find better arguments when a court is involved.
-- Question 2: Under German law, does the photographer have any claim against Laion?
Difficult.
The photographer asked Laion to remove his work from their dataset [5]. The lawyers replied that Laion does not store any photographies so there is nothing to remove. It seems that the photographer did not understand this correctly when writing the initial cease-and-desist letter.
Does Laion infringe on the photographer's copyright if they mereley distribute links? The photographer argues that Laion admittedly does download raw images themselves, which pertains to Question 1.
Aside from this, in German law there is a theory called "Störerhaftung" [6]. It means that if what you are doing enables copyright infringement you may be liable as well, even if you do not distribute the content yourself. If you get notified that you participate in any such infringement by a copyright holder you must cease and desist immediately.
In some cases, distributors of hyperlinks to copyrighted material have been found to commit copyright infringement, though it has been rather the exception. I have no idea how a court would decide in the current case.
One may argue that simply adding the equivalent of a `robots.txt` will suffice to end the infringement, but that is too simple. The photographer sells his work to a customer who may use it on their website, but does not want anyone else to use his work.
However, the photographer cannot control whether or not the customer uses a `noai` on the website. [3] is opt-out, but for this to work, the photographer must have the opportunity to opt-out. It seems the law did not sufficiently appreciate this scenario. It also seems it did not foresee instances where content mining enables AI models replicating (parts of) the content they did mine.
One plausible scenario is that the court denies the photographer's claims to Laion because he can contractually mandate clients to place a `noai` tag on their website. In any case, I think the photographer would have to demonstrate how specifically Laion cost him money as there are no punitive damages in German law.
[0] https://laion.ai/blog/laion-5b/ [1] https://github.com/rom1504/img2dataset [2] https://www.alltageinesfotoproduzenten.de/2023/04/24/laion-e... [3] https://www.gesetze-im-internet.de/urhg/__44b.html [4] https://www.gesetze-im-internet.de/urhg/__60d.html [5] https://www.alltageinesfotoproduzenten.de/2023/02/20/laion-v... [6] https://de.wikipedia.org/wiki/St%C3%B6rerhaftung [7] https://en.wikipedia.org/wiki/Directive_on_Copyright_in_the_...
Otherwise I could download the image, re encode using WebM, put it inside a zip file with other random files as well, and then there would be no single file with their copyrighted material.
Because LLM keep their data on files. Unless we don't keep the data on files but on a Blob on a SQL database! Yay this is fun
He asked for them to be deleted.
Their lawyers responded to say that the pictures are not stored and so they can't delete them and said if you try and make an unjustified copyright claim we can recoup the cost.
He made some kind of claims in a letter that as far as I can tell isn't in the article - I guess he pushed forwards with making the claim about copyright and/or demanding they delete his pictures? What did he send?
They responded saying well we told you it was unjustified and now we've incurred costs so as previously stated we're charging them to you.
> while LAION continues to use his pictures.
Are they using his pictures? What they have is a link to the picture, right?