108
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 21 Sep 2023
108 points (90.3% liked)
Technology
34989 readers
31 users here now
This is the official technology community of Lemmy.ml for all news related to creation and use of technology, and to facilitate civil, meaningful discussion around it.
Ask in DM before posting product reviews or ads. All such posts otherwise are subject to removal.
Rules:
1: All Lemmy rules apply
2: Do not post low effort posts
3: NEVER post naziped*gore stuff
4: Always post article URLs or their archived version URLs as sources, NOT screenshots. Help the blind users.
5: personal rants of Big Tech CEOs like Elon Musk are unwelcome (does not include posts about their companies affecting wide range of people)
6: no advertisement posts unless verified as legitimate and non-exploitative/non-consumerist
7: crypto related posts, unless essential, are disallowed
founded 5 years ago
MODERATORS
From my understanding, unless ChatGPT spits out the copyrighted work, they don’t have a case here.
Using copyrighted works to train an LLM which can then generate similar works seems pretty solidly to be fair use.
I am not a lawyer though.
I think it is probably more about how Books3 is and was always the full content of a torrent site for books.
Fair Use is all fine and good but its very telling that these companies are happy to justify piracy when its convenient for them and then oppose it when it is not.
It is rather hypocritical and there is also questions whether Fair Use can apply to a non-human. People generating art and text with it are in a weird grey area, because on one hand it can be argued they are using a tool, but on the other, the results are so random: how much influence does the user actually have over what they create? If the answer is "not much influence" then the tool is creating the art, not the person. At that point, is it really reasonable to argue "fair use?"
Depending on the jurisdiction's particular laws, there are places where downloading pirated material is perfectly legal and it's just the uploading of the pirated material that's a breach of copyright. In those places you can legally download a pirated torrent if your upstream speed is zero.
Furthermore, I expect it's an open question whether the "they got this text illegally" thing will mean "therefore the stuff they used it to make is also illegal." It might be that they can sue for the piracy and get a payout for that, but that the AI model itself remains just fine.
Interesting times, the laws were very much not written with this stuff in mind.
Not sure if you just worded it wrong or you're confusing piracy with copyright.
Anyway, piracy involves circumventing a protection system to get something for free instead of paying for it. DVDs for example not only contain copyrighted material, they also have copy protection systems to avoid unauthorized distribution, downloading is illegal as much as uploading since you're getting a copy on which the protection system has been broken on purpose.
Copyright violation, on the other hand, happens only if you publish something you didn't make without the authorization of the original author, so downloading copyrighted material - that doesn't have any protection system - is perfectly legal, uploading it is not.
Bibliotik, the site Books3 is from, regularly updates and has torrents to software specifically for removing the copy protection from ebooks.
I wonder if part of the lawsuit is about the fact that most of the books on Bibliotik had their copy protection removed before being uploaded.
I think their point was that there are countries where piracy (or circumventing copy protection) isn't illegal and only copyright laws exist. Thus downloading pirated stuff isn't inherently illegal.
In some countries the copy protection removal isn't dealt with in any way and thus it's not inherently forbidden, in some it's actually outright permitted by law in some situations. (personal use, education,..) Same applies to tools for copy protection circumvention.
Not being dealt with is not the same as not being illegal. I live in EU, piracy is illegal, as in there are laws against it, but noone will come after you if you (as a common citizen) download pirated movies or software.
Laws across Europe are not uniform. Last time I've checked, there were a couple of countries where downloading for personal use was not illegal.
IIRC Spain, Poland were such countries? Maybe Switzerland? That's on top of countries where it's technically illegal but not enforced.
There are probably more countries around the world with similar laws or with no laws regulating downloads. But I'm on my phone so can't look it up.
Feel free to correct me.
As an EU 'common' citizen in Germany, I can assure you that you most certainly will be gone after for downloading any and all copyrighted material.
A single episode of a TV series once cost me 1500€ when I forgot to activate my VPN.
I respectively disagree there. If a writer didn't give any concent whatsoever to give an A.I. a copy of their written works for an A.I. to train or base anything on. I think it's a fair case of theft. Theoretically, someone would be able ask an A.I. to recite an entire book for them. Without ever having to pay its author any contribution.
Did the writer give me permission to read their book, which I used to learn to write better and sell those works?
Did Michelangelo give every art student that learned from his works permission to learn from his work and then produce works in a similar style on their own to sell for profit?
The thing is that humans are doing something drastically different from the current generation of AIs. If you tell Dall-e to come up with something influenced by El-Greco Cezanne and Toulouse-Lautrec you always get a generic impressionist painting you don't get Picasso. The "AI"s we have are mapping and transformation tools with a random number generator attached. In a way they are more akin to a DJ remixing a music (which requires permission from the copyright owner) than a musician creating their own song.
You mean that it’s far more like a human then, since there were hundreds of contemporaries of Picasso who didn’t create cubism?
(I think you're arguing from an ethical standpoint whereas OP was arguing legally, but anyway....)
No, that shouldn't happen. If an AI were ever able to recite back its training data verbatim, that AI would be overfitting. It happens by accident sometimes early on in development when your training data is too small and your model is too big, but it's an error, and is something to be avoided and corrected.
The whole point of training is to get it to a point where it can't recite back any of its training data. In order for that to happen, the AI is forced to sort of generalize and abstract (sorry for anthropomorphizing) its training data. That's the only way to get it to be able to generate something new, which is the whole point of the endeavour.
Long story short, if an AI could recite back an entire book, by definition it could not be an AI, and it wouldn't resemble any of the popular LLMs we have now like ChatGPT. (But you may see snippets and pastiches and watermarks show up)