counterpoint: what if we just make an exception for tech companies and double fuck consumers?

This is the correct answer. Never forget that US copyright law originally allowed for a 14 year (renewable for 14 more years) term. Now copyright holders are able to:

• reach consumers more quickly and easily using the internet
• market on more fronts (merch didn't exist in 1710)
• form other business types to better hold/manage IP

So much in the modern world exists to enable copyright holders, but terms are longer than ever. It's insane.

That would work if you were rich and friends with government officials. I don’t like your chances otherwise.

You are a glass half full sort of person!

Another win for piracy community

When a corporation does it to get a competitive edge, it is.

What OpenAI is doing is not piracy.

Ho no, what will we do without degenerate generative AIs?!

Fartsniffer 🤣

gosh Ed Zitron is such an anodyne voice to hear, I felt like I was losing my mind until I listened to some of his stuff

I don't think it's actually such a bad argument because to reject it you basically have to say that style should fall under copyright protections, at least conditionally, which is absurd and has obvious dystopian implications. This isn't what copyright was meant for. People want AI banned or inhibited for separate reasons and hope the copyright argument is a path to that, but even if successful wouldn't actually change much except to make the other large corporations that own most copyright stakeholders of AI systems. That's not actually a better circumstance.

This has been the legal basis of all AI training sets since they began collecting datasets. The US copyright office heard these arguments in 2023: https://www.copyright.gov/ai/listening-sessions.html

MR. LEVEY: Hi there. I'm Curt Levey, President of the Committee for Justice. We're a nonprofit that focuses on a variety of legal and policy issues, including intellectual property, AI, tech policy. There certainly are a number of very interesting questions about AI and copyright. I'd like to focus on one of them, which is the intersection of AI and copyright infringement, which some of the other panelists have already alluded to.

That issue is at the forefront given recent high-profile lawsuits claiming that generative AI, such as DALL-E 2 or Stable Diffusion, are infringing by training their AI models on a set of copyrighted images, such as those owned by Getty Images, one of the plaintiffs in these suits. And I must admit there's some tension in what I think about the issue at the heart of these lawsuits. I and the Committee for Justice favor strong protection for creatives because that's the best way to encourage creativity and innovation.

But, at the same time, I was an AI scientist long ago in the 1990s before I was an attorney, and I have a lot of experience in how AI, that is, the neural networks at the heart of AI, learn from very large numbers of examples, and at a deep level, it's analogous to how human creators learn from a lifetime of examples. And we don't call that infringement when a human does it, so it's hard for me to conclude that it's infringement when done by AI.

Now some might say, why should we analogize to humans? And I would say, for one, we should be intellectually consistent about how we analyze copyright. And number two, I think it's better to borrow from precedents we know that assumed human authorship than to invent the wheel over again for AI. And, look, neither human nor machine learning depends on retaining specific examples that they learn from.

So the lawsuits that I'm alluding to argue that infringement springs from temporary copies made during learning. And I think my number one takeaway would be, like it or not, a distinction between man and machine based on temporary storage will ultimately fail maybe not now but in the near future. Not only are there relatively weak legal arguments in terms of temporary copies, the precedent on that, more importantly, temporary storage of training examples is the easiest way to train an AI model, but it's not fundamentally required and it's not fundamentally different from what humans do, and I'll get into that more later if time permits.

The "temporary storage" idea is pretty central for visual models like Midjourney or DALL-E, whose training sets are full of copyrighted works lol. There is a legal basis for temporary storage too:

The "Ephemeral Copy" Exception (17 U.S.C. § 112 & § 117)

U.S. copyright law recognizes temporary, incidental, and transitory copies as necessary for technological processes.
Section 117 allows temporary copies for software operation.
Section 112 permits temporary copies for broadcasting and streaming.

Right. The problem is not the fact it consumes the information, the problem is if the user uses it to violate copyright. It’s just a tool after all.

Like, I’m capable of violating copyright in infinitely many ways, but I usually don’t.

and learns how to copy its various characteristics

Because you are a human. Not an immortal corporation.

I am tired of people trying to have iNtElLeCtUaL dIsCuSsIoN about/with entities that would feed you feet first into a wood chipper if it thought it could profit from it.

Yup. Violating IP licenses is a great reason to prevent it. According to current law, if they get Alice license for the book they should be able to use it how they want.
I'm not permitted to pirate a book just because I only intend to read it and then give it back. AI shouldn't be able to either if people can't.

Beyond that, we need to accept that might need to come up with new rules for new technology. There's a lot of people, notably artists, who object to art they put on their website being used for training. Under current law if you make it publicly available, people can download it and use it on their computer as long as they don't distribute it. That current law allows something we don't want doesn't mean we need to find a way to interpret current law as not allowing it, it just means we need new laws that say "fair use for people is not the same as fair use for AI training".

If an LLM consumes the same copyrighted content and learns how to copy its various characteristics, how is it meaningfully different from me doing it and becoming a successful writer?

That is the trillion-dollar question, isn’t it?

I’ve got two thoughts to frame the question, but I won’t give an answer.

1. Laws are just social constructs, to help people get along with each other. They’re not supposed to be grand universal moral frameworks, or coherent/consistent philosophies. They’re always full of contradictions. So… does it even matter if it’s “meaningfully” different or not, if it’s socially useful to treat it as different (or not)?
2. We’ve seen with digital locks, gig work, algorithmic market manipulation, and playing either side of Section 230 when convenient… that the ethos of big tech is pretty much “define what’s illegal, so I can colonize the precise border of illegality, to a fractal level of granularity”. I’m not super stoked to come with an objective quantitative framework for them to follow, cuz I know they’ll just flow around it like water and continue to find ways to do antisocial shit in ways that technically follow the rules.

Say his name y’all

Suchir Balaji