Live data from Hacker News

Artificial Intelligence and Copyright: Request for comments

federalregister.gov

161–170 of 321 posts

Re: Artificial Intelligence and Copyright: Request for comments

#161
post #152
post #68

Earlier quoted context omitted.

Yes, someone using a model can’t know if the generated text/image/sound is a nearly identical copy of the original material they don’t recognize. If use of the output of these systems comes at significant legal risk then then such systems become nearly useless.

> if the generated text/image/sound is a nearly identical copy of the original material they don’t recognize how does the industry today deal with artists that "copy" off some other works? This isn't a problem with AI at all - just that AI provides a tool to generate such works faster.

Someones comes to me to ask for a drawing of Batman or to write an erotic story around Supergirl. I can do it, but I cannot claim ownership over the characters. And I think I will quickly get a letter from DC or Marvel if I try to do this at scale.

Re: Artificial Intelligence and Copyright: Request for comments

#162
post #89

I believe we first need to answer the question of whether the copyright of the AI model’s source text or images affects the output. My opinion — and note I’m a software engineer, not a lawyer — is that an AI, being a statistical model and not generally intelligent, should not be allowed to disregard the copyright of its source material. This would, I think, require the AI’s creator to secure a license for all of its…

Why is being a statistical model relevant? The simplest statistical model is an average. Why would the average pixel rgba of a bunch of images invoke the copyright of those images?

The crux of the AI copyright argument sits in economics. Those currently producing content want future content generated from AI to benefit them financially, as long as a thin sliver of their own content was used in the training.

This is like asking all the student to pay their teachers a (small) percentage of their future economic output.

Re: Artificial Intelligence and Copyright: Request for comments

#163
Hi HN, I have been working on something directly related to AI and copyright. Would it be ok to point it out here?

Recently The Pile was taken offline from The Eye by DMCA. One solution is to host it offshore, which we're calling The Nose: https://thenose.cc

The technical security measures may be of interest to the audience here, so I'll be as detailed as possible. The following formula should be safe if you follow it to the letter.

The basic setup is to install Whonix on a VeraCrypt drive, acquire Monero through any method, use a service like changenow to convert Bitcoin on a wallet stored only on the Whonix installation, sign up for a ProtonMail account (when they ask for email verification, use a no signup inbox service like yopmail), rent a dedicated server at Shinjiru using bitcoin, and register the domain at the same place. They're both a registrar and a server host, which simplifies matters. Use N/A for all contact info. Use Cloudflare to manage your site's DNS records.

Wallet security: do not ever move Bitcoin to any wallet linked with your personal identity. This is easier said than done. First there is the question of how to store passwords. These are the keys to the kingdom, and are the most sensitive aspect by far, because they're intimately linked with you. Additionally, if hardware failure occurs, you'll lose everything if you store them on the Whonix drive. My setup is to use KeePass to store the passwords on a laptop I use to VNC into the computer with the Whonix drive, and then save the database to a folder that gets synced to the cloud. The only flaw in this model is that if your laptop is compromised while your KeePass is open, you're done. But (as Ulbricht discovered) this is always true. The threat model assumes lawyers coming after you with DMCA with additional safeguards against the FBI narrowing down who you are in real life. If your physical location is compromised through any method, you're done.

All it takes is one mistake to end you. SSH into your box from your real computer? Done. Sign up using your real name with Mailgun? Done. Accidentally say "Thanks, " to the support staff at Shinjiru in an email? Done. Abandon ship and close everything down.

The security of this technique comes down to simplicity. There are very few moving parts. I opted for nginx + mediawiki with Discourse forums at https://forums.thenose.cc (though I don't know if anyone will care enough to join). Logging is turned off to protect users downloading the data, though you only have my word on this. But reputation is the only thing a hacker has ever truly had anyway.

If you're serious about following the above recipe, I urge you to read through the Whonix docs on online anonymity: https://www.whonix.org/wiki/Documentation Remember, threat model is your saving grace. You probably aren't starting a darknet, so you can relax your threat model in terms of physical safety. But you won't get away with any mistakes made in cyberspace.

As for the site itself, I've avoided asking for donations for now (hosting is $130/mo though, which will get expensive) or describing anything beyond this HN comment. I'll say it's for simplicity, but in fact I only started it a few days ago and haven't had time to provide anything but the essence of our service: hosting AI datasets in stable, copyright-resistant ways.

If additional datasets beyond The Pile need protection or distribution, you can contact me at nostril@thenose.cc or at https://forums.thenose.cc. I have a 4TB drive, of which 800gb is being used by The Pile so far.

Re: Artificial Intelligence and Copyright: Request for comments

#164
post #152

Earlier quoted context omitted.

> if the generated text/image/sound is a nearly identical copy of the original material they don’t recognize how does the industry today deal with artists that "copy" off some other works? This isn't a problem with AI at all - just that AI provides a tool to generate such works faster.

Someones comes to me to ask for a drawing of Batman or to write an erotic story around Supergirl. I can do it, but I cannot claim ownership over the characters. And I think I will quickly get a letter from DC or Marvel if I try to do this at scale.

> I can do it, but I cannot claim ownership over the characters.

of course not. But you can claim ownership if you don't call those characters their original names, and make sufficient changes to the design (how sufficient is determined by a court of law - thus expenses).

> DC or Marvel if I try to do this at scale.

The show 'invincible'[1] has a character that is a basic copy of superman. And yet, you will find that they don't get a letter from DC.

[1] https://en.wikipedia.org/wiki/Invincible_(TV_series)

Re: Artificial Intelligence and Copyright: Request for comments

#165

Earlier quoted context omitted.

I don’t think it makes sense for both model builders and the model’s users to separately obtain licenses for the same works used in the training set. A model trained on several copyrighted data sources cannot somehow be used in a way depending on a subset of those sources. So all parameters of usage and compensation should be settled by contract between the model builder and copyrighted data supplier, before the copy…

I slightly disagree, in that I think the person using the tool should bear the burden of copyright. I.e. if the model outputs something under copywrite it merely can't be republished. In this same way, i can use Photoshop on proprietary data but I can't necessarily sell the results.

But that problem is already solved.

Copyright holders are already protected from (I.e. can legally prohibit) distribution of obvious copies, or clearly derivative works.

Regardless of how they were produced by hand, copy machine, Photoshop or with a model.

The new problem is that artists styles are being “stolen” by incorporating their copyrighted work into models without their permission.

And that problem can easily be solved if using copyrighted material to create models is declared NOT fair use.

Artists could still allow models to be built from their work, but on their terms. If they wish to do that.

A famous artist, that doesn’t mind being commercial, could sell their own unique model to let fans create art in that artist’s style, while not having their style “ripped” by others.

Or just keep their style to themselves, for their own work, as artists have done for centuries.

(Of course, with greater effort, their style could still be recreated - styles are not protected unless they are trademarked - but the recreation would have to be done without using the artist’s copyrighted works.)

Re: Artificial Intelligence and Copyright: Request for comments

#166
post #148

Earlier quoted context omitted.

> If I know anything I create is just going to be hoovered up and input into somebody's AI model but today, without an AI model, anything you create is already going to be learnt and studied (if it is worth studying of course). What's the difference, but speed? > they get 99% of the profit Why is that a priori the assumption? What stops you from getting a profit? > I do 99% of the work you did 0.000001% of the work,…

> What stops you from getting a profit? OpenAI and Stable Diffusion not paying for their dataset. I don’t believe GitHub asked for my contribution to Copilot.

But you weren't receiving profit from your works originally? So therefore, why does it matter what someone else was doing?

Re: Artificial Intelligence and Copyright: Request for comments

#167
post #56

I believe we first need to answer the question of whether the copyright of the AI model’s source text or images affects the output. My opinion — and note I’m a software engineer, not a lawyer — is that an AI, being a statistical model and not generally intelligent, should not be allowed to disregard the copyright of its source material. This would, I think, require the AI’s creator to secure a license for all of its…

My opinion as a SWE who is dating a lawyer (joke, not a serious qualification but it does provide some insight): Generative models traverse and interpolate high dimensional state spaces. These state spaces are created from input data. I would argue people do the exact same thing - the first main difference is we can use novel inputs (e.g. we can use images or words to develop our music/temporal state spaces and vice…

The analogy doesn't hold when you consider the sheer scale of the problem.

I can outright buy a machine for a few thousand dollars that can crank out a faithful rewrite of every Stephen King novel without the shitty endings and nonsense plot points. It can do it in a few days, maybe a couple of weeks at most.

To do that with human labor would take years and cost hundreds of thousands, if not millions of dollars.

Instead of paying an artist a couple hundred for a commissioned drawing, I can just scrape up their entire portfolio and generate any image I want with their style. I can generate hundreds or thousands of images. I can take their distinct style and use it exclusively as the branding for my company.

What a ML model does is very fundamental not what happens when a human draws inspiration from prior art. A human would require an extremely significant amount of time and resources to perfectly imitate every artist they have ever seen. It takes a human significant time and resources to produce faithful variations on prior art.

A ML model is measured in words or images per second.

Re: Artificial Intelligence and Copyright: Request for comments

#168

I'm going to try to plead my case for images generated using sophisticated prompt engineering to be copyrightable. For example, at the point that I've written a prompt with 20 tags, 10 negative prompt tags, some loras, custom weights, embeddings merges, and prompt editing, I'm now writing what is effectively a "program", which should be copyrightable and so should its outputs. It's total BS to me that a book of midjo…

The training model for Stable Diffusion has a lot of copyrighted images mixed together into an output which makes the plagiarism non-obvious, but let's reduce the set by 1 image. Shouldn't affect the output too much, right? Maybe some prompt will have a slightly different image. Now let's reduce it by another image. Again, less options for what to display, fewer images to take pixels from, but still a lot of options,…

> Once there are millions of images, you end up with just tiny blobs of pixels being copied from many different images.

This is not how these neural nets work. They don't copy pixels from anywhere. They learn features.

The features represented internally are generally not easy to interpret to humans, but for sake of illustration, there could be an artificial neuron that fires when a subject should have blue eyes. Having a lot of blue eyes in the training data would help this neuron learn better when to fire (based on the values of other neurons, which may in turn represent other features). For example, it may learn to place more importance on an input that represents pale skin or Nordic origin.

It can learn concepts like cars have wheels, and wheels are round, etc. And then when you ask it to draw a car, it composes one from the concepts it learned. Some parts of the network will deal with the fine details that more directly influence pixels, but these aren't copying pixels from any image either. They're weighing a bunch of factors (eg is this pixel part of the iris and did the network decide to make a person with blue eyes?) and choosing pixel colors based on those factors.

Re: Artificial Intelligence and Copyright: Request for comments

#169
post #145

Earlier quoted context omitted.

I personally have a really hard time finding any meaningful difference or distinction between "AI" and "lossy compression". Copyright and "lossy compression" are pretty easy to reason about. Model "building" is "compression". Model "use" is "decompression". Everything about these AI models seems to be about the "lossy" part, but "lossy" is just an adjective to the main show. It's very difficult to not conclude that c…

Some compression, yes, but the analogy oversimplifies. AI rerepresents input information in a transformative way (embedding, say) then creates new, derived and combined output from a new input (e.g prompt). It's not just lossy compression. It's potentially novel.

Phrases like "transformative way" are meaningless woospeak to me. Everything is a transformation. Sulpose I run a linear convolution on ten images and average them. Is the result "new"? Does it not contain the original images? Subspaces and mappings don't create anything "new" any more than SVD does. This is just playing digital Ship of Thesius.

Re: Artificial Intelligence and Copyright: Request for comments

#170

Earlier quoted context omitted.

When you perform lossy compression, you feed it one file at a time, not every file in existence.

If you concatenate images into a stream container (say as tar) and then compress the stream, the compression coding will (generally) cross over the individual images. True, that's generally not lossy compression. But concatenating images is also how you create video. Lossy video compression does typically cross over frames. So I don't actually see a difference. If you want to think about mkv or mp4 instead of zip it'…

What does any of this have to do with creating a new expression?
Post reply on HN