Obviously you get the Shakespeare influence, but the bar on copyright infringement is very high for a reason — we don't want to sue artists for being influenced by another writer.
And if you despise capitalism so much, put your money where your mouth is and relocate to Cuba.
Champagne pseudo-socialists like you give the rest of us liberals a bad name.
Instead of reflexively spewing a series of ideological non-sequiturs, perhaps engage with what I actually said.
Could you name a single startup in the US that has been caught 'bribing' politicians?
And stop drinking the Uber/WeWork/Airbnb/ kool-aid. The vast majority of startups are founded by honest people, and don't rely on breaking laws in order to establish a dominant market position or 'raise a ton of money', I should know, I'm an early employee at a startup with a 'dominant market position'.
Yes you should demonise people for any criminal actions they've actually committed, but please don't demonise them simply for being successful.
All of your comments on this thread sound as though they're made by a college undergrad still in the anti-capitalist/hippy-communist stage.
Grow up! Most things in life are not conspiracies and believing they are will significantly harm your emotional health.
https://theguardian.com/news/2022/jul/10/uber-files-leak-rev...
Oh, but of course, only if I set aside the “kool-aid” of citing the examples of the dominant startup players of the past decade.
Perhaps you should, again, grow beyond your ideological name-calling to a rational engagement with the world, which actually involves everything I’ve described.
Or are you claiming that the word "lobby" in the phrase "attempts to lobby Joe Biden, Olaf Scholz and George Osborne" actually meant 'bribe'?
And despite you frantically scouring Google for something to backup your 'bribery' claims with, the above article ended up being the only thing you could find.
So basically, you essentially spent your own time to prove my point. Thanks.
I'll repeat again what I told you earlier: grow up and stop gorging yourself on conspiracy theories.
Signing off!
If anything, all this thread had demonstrated is an extreme need on your behalf to defend your belief in a tooth-fairy version of Capitalism, and scream “COMMUNIST” and plug your ears whenever confronted with reality of mommy and daddy startup putting money under politician’s pillows to win their place in the world.
Do you genuinely believe this?
I feel like after the whole SBF fiasco this should all be quite obvious…
You've posted some good comments in other threads, so hopefully this should be easy to fix if you review https://news.ycombinator.com/newsguidelines.html and take the intended spirit of the site more to heart. We'd be grateful!
You've posted some good comments in other threads, so hopefully this should be easy to fix if you review https://news.ycombinator.com/newsguidelines.html and take the intended spirit of the site more to heart. We'd be grateful!
I'd like to try this. Any suggestions where I can learn how to implement this technique?
It's simply not. Stable Diffusion 1.5 was trained on over two billion images. Even if we assume the images were only 10kb on average, that would require 20TB of storage space. The pruned 1.5 model is only 4.27GB, or 4,684x smaller. It is fundamentally and provably impossible to achieve that sort of compression ratio on diverse and already compressed images.
'Coordinates of the image' is a layman's explanation of latent space and the training process, it should not be taken literally.
> how would it be possible to fine tune the model on a set of images and recreate their style?
The fine tune process represents additional training, and to reliably and consistent reproduce a consistent style generally requires overfitting. It is very roughly analagous to an artist using a reference while drawing/painting.
> This is setting aside the myriad examples where SD regurgitates fairly close copies of individual works.
Some works were overrepresented in the dataset because they are overrepresented in the world. There are substantially more images and variants of the Mona Lisa on the internet than there are Random Twitter Artist's Furry Waifu.
This works in much the same way for humans, who are more far more likely to be influenced by the Mona Lisa than Furry Waifu.
> This is a bit like training a model on the works of Shakespeare, but then claiming that the tensors resulting from ingesting the data are not actually Shakespeare’s writing.
.. They aren't. Shakespeare did not write them. His works can't be reproduced from them. There is, provably, not enough data for it. This is more different than if an individual read and referenced Shakespeare while writing a play, for all that entails. Specifically: whether it's a human or a model generating the content, the content that is generated could be a violation of copyright.
You'll note that this is no different to if I copy and paste Mickey Mouse onto a tshirt design, or if I draw him on a piece of paper with a crayon.
Should we also be up in arms about copy/paste and crayons?
> Oh really, where’d they come from then? Thin air?
This hits on a more important point than I think you realize here. Where did inspiration for the human artists come from? Did they spring from a formless void? Or have they seen a lot of images, internalized what they think was important from it, and then generated art from that?
Oh really? Try googling overfitting. Or ask ChatGPT!
> The fine tune process represents additional training, and to reliably and consistent reproduce a consistent style generally requires overfitting ...
It's clear you have a perspective that is set in stone, and insist on not understanding the technology. Hopefully you'll re-evaluate when the plaintiffs in these suits inevitably lose.
> I don’t really care about the copyright argument as it relates to human learning because I think this is all a Napster-style false flag by capital owners to enclose ALL data on the internet. The intent isn’t to argue about these models from first principles such that we all have IP-restriction free access from them. It’s to make them do enough legally gray things at scale that they get restricted to only a few power players.
This argument confuses an image and the style. The former is copyrightable, the latter is not.
>In reality, Stable Diffusion works by first converting a user’s prompt into a latent representation: a list of numbers summarizing the contents of the image. Just as you can identify a point on the Earth’s surface based on its latitude and longitude, Stable Diffusion characterizes images based on their “coordinates” in the “picture space.” It then converts this latent representation into an image.
Regardless of the lawsuit, this seems like a quite spurious, self-serving argument in and of itself. If the “coordinates” of the image were not equivalent to the image, how would it be possible to fine tune the model on a set of images and recreate their style? This is setting aside the myriad examples where SD regurgitates fairly close copies of individual works.
This is a bit like training a model on the works of Shakespeare, but then claiming that the tensors resulting from ingesting the data are not actually Shakespeare’s writing. Oh really, where’d they come from then? Thin air?