4 ms·

The next big trend in AI is it dying the death of a thousand cuts in the form of copyright infringement lawsuits. Example: https://www.theverge.com/2023/7/9/23

by 300bps 3y ago

The next big trend in AI is it dying the death of a thousand cuts in the form of copyright infringement lawsuits. Example: https://www.theverge.com/2023/7/9/23788741/sarah-silverman-openai-meta-chatgpt-llama-copyright-infringement-chatbots-artificial-intelligence-ai https://www.theverge.com/2023/7/9/23788741/sarah-silverman-o... Imagine companies thinking they can scrape all the world's information for free and then package it up and sell it.

ausbah 3y ago
I mean they already do that right now.
- 300bps 3y ago
  Sure. "Easier to say sorry than please". Until you get sued into bankruptcy by millions of litigants.
  - stale2002 3y ago
    Or, instead of that, everyone gets away with it, under broad "transformative/fair use" protections and nobody gets sued. That's what's happening now. But sure, people will always make these unfalsifiable arguments about some hypothetical doomsday that is far enough in the future that they can never get called out when it doesn't arrive.
api 3y ago
> Imagine companies thinking they can scrape all the world's information for free and then package it up and sell it. That's one of the main business models of the Internet, only usually it's not even scraped. People upload it willingly in exchange for free hosting and free social media. The other big business model of the Internet is mass surveillance driven targeted advertising.
- 300bps 3y ago
  It's hard to guess who you're talking about but that is not what Google does. They scrape public web servers, respecting robots.txt directives and then they provide links back to the original site. And if someone hosts user-submitted copyrighted information without a license, the copyright holder can submit a complaint via DCMA and the hoster has safe harbor from liability. OpenAI is literally scraping copyrighted works, packing it up and selling it without a license. Art, books, magazines, everything. No safe harbor from DMCA for doing that.
  - xp84 3y ago
    I’m willing to admit that training a model on a book is not exactly the same as a person reading a book and remembering what they learned. Are you willing to admit that training a model on the book is not the same thing as copying that book? A big problem here is that copyright law was a massively problematic thing even before transformer model tech was developed. It has been distorted beyond recognition by a few companies, who have state everything on a broad and permanent copyright rather than the limited one we started out with. We probably need new definitions in the law, because pretending that training a model on some thing equals copying it isn’t based in reality. It’s an emotional appeal meant to gin up outrage.
    - 300bps 3y ago
      Are you willing to admit that training a model on the book is not the same thing as copying that book? This type of interaction is not helpful. It's an argumentative strawman argument. Who said training a model is the same thing as copying a book? Of course it isn't. But who said it had to be? Here's an idea... read the first five books in the series "A Song of Ice and Fire". Then sit down and write your own version of Book 6 to continue the story and sell it without a license. Guess what's going to happen? You will be sued into bankruptcy. What OpenAI is doing is lot more similar to that than literally copying things. And it's still wrong and illegal. It has been distorted beyond recognition by a few companies, who have state everything on a broad and permanent copyright I agree with you, what Disney and others have done with copyright extensions is immoral and should be illegal. But it's not illegal. pretending that training a model on some thing equals copying it isn’t based in reality No, it isn't based in reality. Which is why nobody made the claim. They're packaging up a derivative work and selling it. Don't have to look hard to see examples that this is just as infringing as outright copying.

by 300bps 3y ago

ausbah 3y ago

I mean they already do that right now.

300bps 3y ago
Sure. "Easier to say sorry than please". Until you get sued into bankruptcy by millions of litigants.
- stale2002 3y ago
  Or, instead of that, everyone gets away with it, under broad "transformative/fair use" protections and nobody gets sued. That's what's happening now. But sure, people will always make these unfalsifiable arguments about some hypothetical doomsday that is far enough in the future that they can never get called out when it doesn't arrive.

api 3y ago

> Imagine companies thinking they can scrape all the world's information for free and then package it up and sell it. That's one of the main business models of the Internet, only usually it's not even scraped. People upload it willingly in exchange for free hosting and free social media. The other big business model of the Internet is mass surveillance driven targeted advertising.

300bps 3y ago
It's hard to guess who you're talking about but that is not what Google does. They scrape public web servers, respecting robots.txt directives and then they provide links back to the original site. And if someone hosts user-submitted copyrighted information without a license, the copyright holder can submit a complaint via DCMA and the hoster has safe harbor from liability. OpenAI is literally scraping copyrighted works, packing it up and selling it without a license. Art, books, magazines, everything. No safe harbor from DMCA for doing that.
- xp84 3y ago
  I’m willing to admit that training a model on a book is not exactly the same as a person reading a book and remembering what they learned. Are you willing to admit that training a model on the book is not the same thing as copying that book? A big problem here is that copyright law was a massively problematic thing even before transformer model tech was developed. It has been distorted beyond recognition by a few companies, who have state everything on a broad and permanent copyright rather than the limited one we started out with. We probably need new definitions in the law, because pretending that training a model on some thing equals copying it isn’t based in reality. It’s an emotional appeal meant to gin up outrage.
  - 300bps 3y ago
    Are you willing to admit that training a model on the book is not the same thing as copying that book? This type of interaction is not helpful. It's an argumentative strawman argument. Who said training a model is the same thing as copying a book? Of course it isn't. But who said it had to be? Here's an idea... read the first five books in the series "A Song of Ice and Fire". Then sit down and write your own version of Book 6 to continue the story and sell it without a license. Guess what's going to happen? You will be sued into bankruptcy. What OpenAI is doing is lot more similar to that than literally copying things. And it's still wrong and illegal. It has been distorted beyond recognition by a few companies, who have state everything on a broad and permanent copyright I agree with you, what Disney and others have done with copyright extensions is immoral and should be illegal. But it's not illegal. pretending that training a model on some thing equals copying it isn’t based in reality No, it isn't based in reality. Which is why nobody made the claim. They're packaging up a derivative work and selling it. Don't have to look hard to see examples that this is just as infringing as outright copying.