Ask HN: Is there a license to protect my code and art from AI?
I'm looking for a way to protect my GPL/MPL/other code and CC game artwork against ML that would ingest it to produce other work. The only exception is when someone uses maps to train AI-driven bots - that's fine. So, I'm looking into some standardized way to tell my work is for human use only. Ideally, also in automated way, like prefixing headers with something like /* <NOML> */.
102 comments
[ 4.2 ms ] story [ 158 ms ] threadAnd just to be clear, a theory is all this is. It has never been tested in court.
Or we could agree on some industry standard for now, like robots.txt... which doesn't, of course, guarantee anything.
I wonder if EFF is working on something similar.
But that doesn't matter. For this topic we only care about the materials that are intentionally published to the open web, and the different ways those materials can be licensed.
I'd estimate a decent chance EFF will say that there isn't a right under existing copyright law to restrict machines from learning from copyrighted works. Remember that EFF usually does not want copyright rights to be stronger. (Probably the most general way to put it is that EFF basically always wants new uses of creative works made possible by technology not to be governed by copyright -- to have their benefits accrue to users.)
(I worked at EFF from 2001 to 2020.)
Edit: you could look at Jaron Lanier's stuff for the opposing view; I don't know if he has any specific technical or legal proposals.
Do you have that problem? That your code/art is accessible to the wider world when it shouldnt be? In that case it sounds more like a task for somebody with a security background then for someone with knowledge of copyright.
What I am trying to assess, though, is the fact that "there's something that is not yours and the owner decides if you can have/copy/ML it". Not the fact that someone did. Or, if did, what would the consequences be.
Trying to skip over the admin to fix this missconfiguration and going straight to the lawyer is the attempt to externalize costs. Which is where the lack of compassion comes in. This is a shitty thing to do.
This is not a lawyer problem, nobody broke in, nobody took anything. You are giving stuff away. Go talk to an admin on how to only give your toaster to the people that you want to give it to.
Misconfigured servers are 100% the fault of the admin.
This comment is such a shallow dismissal of a good post that brings up a genuine and legally complex question that many readers of HN care about.
Are you trying to be clever? Because this comment is anything but clever. It is not hard to define humans. The humans are biologically animals and have a certain set of properties that can be easily searched online. For instance https://en.wikipedia.org/wiki/Human
And no you cannot just weasel your way into somehow fooling us or a jury that machines and AI algorithms that exhibit similar properties are also human! I frankly don't understand the point of your comment.
And that would be a good and insightful question and make an interesting discussion. Posting a two word comment like "Define human." without substantiating it as if the two words are supposed to communicate something very clever is not! It is lazy and presumptuous.
> Much like your response, but yet you did it anyhow
Their initial response was not lazy.
And how do you not be presumptuous in response to such a vague two word comment? A Rorschach test is not a good way to communicate.
Not wanting to play "guess the deeper argument" is not laziness. And they wrote a perfectly reasonable length response to get across their entire point, which is also not lazy.
You can make your code something that cannot be _reproduced_ without a license, but you cannot stop someone from learning ideas from your published code. This does not require a license (only distributing, or making derivative works needs licenses).
Whether that learning comes via ML, or a real flesh human brain, i think, makes no distinction. You will need to lobby for an update to copyright laws to add a new right to be granted.
You think. Others disagree. It has not yet been tested in court.
Additionally there is something to be said for overfitting: On one end of a spectrum I could have a well trained AI and on the other end I could have a lookup table of the training data and somewhere in the middle maybe a model that overfits a lot and produces something close to the original content. How about that? I think there are a lot of interesting questions here.
cough what about high-frequency trading?
Black Rock slams into the topic
courts don't create new rights, they interpret existing laws.
I can't really see how the existing copyright laws can be interpreted such that a distinction could be made between a machine learning and a human learning.
I do not believe such rights exist under current copyright laws. Others argue that it does (or ought to exist).
Courts do indeed interpret existing laws. The question of whether current laws regarding human learning apply equally to machine learning is one of interpretation, precisely the sort of thing that is ultimately resolved by court decisions, including appeals potentially up to the supreme court, establishing precedents. Until that process takes place, your interpretation is just your interpretation, not the interpretation.
* 1. Right to control the reproduction of the work
* 2. Right to control the making of derivative works
* 3. Right to control the distribution of the work
* 4. Right to control the public performance of the work
* 5. Right to control the public display of the work
* 6. Right to perform a sound recording publicly by means of digital audio transmission
The closest is the right to make derivative works - is the ML model a derivative work of the training data? I guess this would need to be tested out in court.
I wonder if CC-NOML will pop up any soon (Creative Commons).
And if OpenAI training data excludes their artwork from training data then the answer is obviously that copyright matters and they only exploit small artists unlikely to sue (which would make openai legally liable)
My argument is that it is implicit - a human viewing the training data is allowed to learn from it. There's no reason to differentiate a human brain from a machine "brain" in this context.
On the off chance you do: it's not about some a "need" to differentiate or not. They are just completely different, unrelated in all ways concepts. Are you familiar with cases where a single word is used for unrelated things as a metaphor?
A machine does not learn in human sense because it lacks and will always lack consciousness. If you want human rights to apply to machines you first have to prove they have consciousness. If you do that (let's pretend), the subject of this thread stops being an issue, because once a machine has human rights no one is able to use it to make unlimited copyright-violating works without paying equivalent human wage, making the entire point of ML acting as copyright laundering tool moot.
i read it, and i disagree that they are completely different concepts. I am describing the action of "extracting information, structure, ideas and form" from a literary work. I used the word "learn" as a shortcut.
A consciousness is not required to perform the action above. Human rights also does not apply to a machine - we aren't talking about sentient ai being given rights that would normally be reserved for humans.
The right to "learn" from literary works is not an exclusive right granted by the copyright law to the holder. A tool would only launder copyright if the tool was used to reproduce the literary work, or a derivative of said work. But if the tool is producing new works, but expresses similar ideas (which is what a lot of the ML models allow you to do), then those expressions should not violate copyright.
As for the model itself, i also argue that it does not constitute a derivative work of the training dataset. I am comparing it to a human brain, which itself is not a derivative work of the training that went into producing it (e.g., the authors of all of the books said human brain read does not have a claim on the brain itself).
Of course, society could opt to change this - specifically add a clause to the copyright law to grant an extra right to the holder regarding ML consumption of their literary work. This has not happened yet, and i sincerely wish it does not. It merely stifles progress.
And yet, it's apparently perfectly legal to put DRM on books and control exactly when and how they are read [1] - by actual humans.
> Whether that learning comes via ML, or a real flesh human brain, i think, makes no distinction.
Strongly disagree here. Sure, ML shops would like that to be the case, but for now it's just muddying the waters by using anthropomorphizing language.
What you call "learning" is so far just scraping the data to adjust model weights using an algorithm in a machine. Contrast that with a human drawing knowledge or inspiration from a work they read before. Those two are different things, unless you want to assume that ML models have a conscious mind and constitute a person - at which point, you'll have much bigger ethical problems than copyright violations.
[1] https://news.ycombinator.com/item?id=34391332
this is, unfortunately, a right to distribution and display. It is most definitely not learning.
> Contrast that with a human drawing knowledge or inspiration from a work they read before.
But neural networks is mimicking a human's brain function in simplified form. The requirement for consciousness is not needed. If the argument is that humans' brains do something so special that a machine cannot be made to reproduce it, then you have an argument; but i do not believe it to be true.
Using a machine created algorithm to extract ideas, knowledge, structure and form from a literary work is exactly what i would call learning.
Imagine, as an example, i learnt how to make/cook new recipes by looking at existing recipes. Is the knowledge gained from doing so a derivative work of the existing recipes?
What if i wrote a machine and parsed these recipes, extracted all of the forms/patterns, and used it to reproduce new recipes? How is the outcome any different from above?
With machines, this process is replicated instantly and becomes pervasive, resulting in elimination of many humans performing the work. This can destroy the potential creativity that was possible from the replaced humans within that specific trade. The machine learning of today is stupid and simple with no creativity, only copies or amalgamations of existing works. The end result is the death of progress within that trade and potentially stagnation of the human race as it perfects the use of the tools (e.g. prompt engineers, glorified button pressers) and ceases to move things forward.
People who claim they copy/paste stuff and can't be creative have clearly not used them very much.
You could argue the same from any introduction of technology that replaces humans. If the output of the machine is of sufficient quality to replace humans, then they should be replaced.
If the fear is that those displaced humans will no longer be useful, then society will need to subsidize their existence, or find a way to make use of those displaced humans (such as retraining). It is an unrelated problem to copyright, and copyright should not be used to ensure some group of humans do not get displaced.
> only copies or amalgamations of existing works.
Not that i agree with the premise of machines lacking creativity, but many of today's human created content is of such level as well.
Regarding "just scraping the data to adjust model weights using an algorithm", all research so far shows that the brain does just this, even though the full details of the algorithm are unknown and all the places the weights are encoded are not known in full.
actually, public display is one of the rights that the copyright holder can control too. Private display, on the other hand, is not. So you can show it to your friend, or AI in private, but not project it in a public place unless the copyright holder agrees.
Reproduction of the original data is one problem, but if we create restrictions around using data for training purposes I believe this is a huge impediment to human progress.
I would imagine that the state of the art in foreign language translation (Google translate, etc) would fall significantly behind for example.
Could you not have a license that grants this as long as the copy is not provided to a machine learning algorithm?
There are probably pathological cases where a repeating image is more strongly overfit in the training data and could be reproduced in much more detail than this average though. But the systems learn similar to the human brain, they learn the gist of a style or scene and how it relates to words. It's not a search engine, it doesn't copy/paste any block of pixels...
One interesting example is that since SD's original training set included some stock photo watermarked images, it learned that there was a concept of watermark, which can end up in the middle of generated images. Not in an intelligible way, but you can see roughly how it interpreted this detail. And in those cases you DO have a very very repeating similar pixel bitmap in the training data.
And with the robots.txt approach, I understand, there'll definively be some rogue AI that would parse and process that on purpose :)
As for "doing the same thing", well, kind of, yeah, we're back to the thousands years old copycat problem. But this time art/code creators don't want others having machine that could recreate years of work in a minute with a push of a button. Also, artists usually acknowledge who did they learn from or who's style are they trying to recreate... but that's an ethic thing..
Copyright itself was introduced because of the scale of the printing press, which fundamentally changed the economics of publishing. Is it reasonable to assume that scale doesn’t change things?
ML is only superficially using the same process. Behind the scenes the computer is not being 'inspired'; it's using the art from other people, along with metadata about it, as a data source to derive a new image using math. That isn't the same as seeing someone else's picture and using it for inspiration.
Anyway, I've always said (and keep saying) that art is what (visibly) distinguishes humans from animals. And machines.
You can add malevolent code to your code base, which allows at best for the ml project to gain self-awareness (copilot "shodan") and at worst to just add maleware. Of course you then
Dont forget to remove the evil pre-compilation. If you do art, i think the best think to throw AI would be fractal details, aka your picture never ends upon zoom in, but just becomes more art. That or you try to throw the weights in another way.
https://www.youtube.com/watch?v=hdCs6bPM4is
I wonder if you could add api calls that are billed per call to your open source project and let copilœdUpCode accumulate tremendous bills before sending them at the end of year.
Imagine you can get it to do it in useless loops async. I just realized, im watching the "capitalist software revolution eating itself" stallman always prophesized live. Allmende public thrust is being eaten up and all that remains is a dark forrest, stalked by parasite riddled hunters and prey.
That is until humanity rebels and dissolves all companies larger then 100 people.
Same thing applies to the idea of self similar imagery, a.k.a. a fractal. Once you rasterized it, it's by definition no longer recursive in nature.
But practically what you need to do is to stop crawlers from reading your code/art by robots.txt or captcha. If your works don't get into CommonCrawl and similar datasets they won't be used for model training. I think you can still enable Google Bot while rejecting AI data collectors on your sites.
<rant> I think copyright refers only to expression, not the ideas themselves. So training on copyrighted code and art should be ok as long as expression is not copied.
In these cases where an AI developer wants to train on the ideas without learning the expression they can re-generate the data using the "variations" method, works both in image and text. This will create substitute data, like anonymising PII. </>
My first thought when looking at this thread was: I wonder if someone will try to claim their robot is a man and that it has been created. What might that imply?
It implies that they are wrong, and raises no interesting legal or philosophical questions.
Come back in a couple decades and we can check if we're getting closer.
A tangent to pondering the sci-fi future isn't entirely unwarranted, but it should be presented as a hypothetical and not as if it's a practical barrier to OP's question. "What if some day an AI approaches being human?" or something.
On the other hand if the meaning was supposed to be about human assistance, it would help to specify that and also talk about what kind of assistance might present meaningful trouble.
Or maybe they meant something else too. Such a wide breadth of topics being gestured at with only two words is not a good way to get a point across.
I don't think attacking people for asking questions is contributing anything at all however.
Your response to my immensely simplified question with "they are wrong" fails to acknowledge a potentially real problem. A problem where too late is immediately followed by a successful attempt.
I see a lot of this today, much more so on HN lately too. I see people failing to be creative and resorting to some kind of "pics or it didn't happen" stance.
The idea that an AI that might be pushed through a legal challenge related to the constitution isn't far fetched in the slightest. It is guaranteed to occur. Not preparing for it is expected and while I love the feeling I get every time I predict the obvious but I would rather it stop occurring. It's short lived and the consequences have been growing.
You think AI is 20 years off. I happen to know for a fact it is already here.
Humans are not complicated.
It's two words. The problem is that it's too easy to link it to multiple, different arguments.
As a more extreme example, if I say "Well, except..." that is a bad comment even if people can fill in the blank with stuff.
> The idea that an AI that might be pushed through a legal challenge related to the constitution isn't far fetched in the slightest.
As being human? Today's tech? Yes it is.
If like GPL you want to allow people to share your code further, and make derivative works, then you must ensure that those people lock such works behind restrictions in the same way, such that derivative works can also only be read after signing the license.
The tech is there.
It won't ever leave.
If you cut off one head, three will replace it.
In my opinion: Copyright is a failure and the absolute best move, with the advantage AI has given to everyone, is to abolish copyright.
These advanced AI models will be trained and licensed by current and future corporations. A law or judicial precedent that denies "fair use" when it comes to machine learning would stop OpenAI and GitHub Copilot dead in their tracks. I'm not that worried about Billy's home-brewed AI model, it's just something we'll have to accept, but stopping exploitation on the scale of $billions would get us 99.9% of the way there.
> In my opinion: Copyright is a failure and the absolute best move, with the advantage AI has given to everyone, is to abolish copyright.
I agree with this. I fundamentally disagree with the idea of introducing 2 different sets of rules which completely correlate with wealth; companies being allowed to bypass copyright laws to train their models and make money from "our" work, while retaining copyright over their software and trained models.
I might agree with machine learning using "fair use" as a defense iff the datasets, trained models, and all accompanying software are released into the public domain. This way everyone gets to benefit from these technological advancements.
Congress decided to grant limited "rights" to copyright holders as incentives for them create, not to protect their work forever, but in exchange for it to be widely available and eventually in the public domain.
That these incentives take the form of limited protections is a side effect. That the original purpose is often corrupted and delivered ham-handedly has more to do with politics than purpose.
If we argued from first principles, we could invent a better system, but that system would undoubtedly allow for ingestion and manipulation by AI.
These topics don't get much time before whatever force is at play flags them out of view.
However, I'd rather not see a separate license / mechanism for this, because now we would have people who'd be fine with their work being used this way, people like us who are not fine with this and people who don't know / care. And mixing code from people of these different groups, which the licenses you cited allow, is going to be a mess.
I also would like that this not be opt-out, but opt-in.
Eventually, we need the legal system to do its work quickly and tell us if fair use can be used to train ML models and in which conditions, so we can build a strong defense.