The Irony of 'You Wouldn't Download a Car' Making a Comeback in AI Debates

Technology@lemmy.world – 70 points – 2 months ago

Those claiming AI training on copyrighted works is "theft" misunderstand key aspects of copyright law and AI technology. Copyright protects specific expressions of ideas, not the ideas themselves. When AI systems ingest copyrighted works, they're extracting general patterns and concepts - the "Bob Dylan-ness" or "Hemingway-ness" - not copying specific text or images.

This process is akin to how humans learn by reading widely and absorbing styles and techniques, rather than memorizing and reproducing exact passages. The AI discards the original text, keeping only abstract representations in "vector space". When generating new content, the AI isn't recreating copyrighted works, but producing new expressions inspired by the concepts it's learned.

This is fundamentally different from copying a book or song. It's more like the long-standing artistic tradition of being influenced by others' work. The law has always recognized that ideas themselves can't be owned - only particular expressions of them.

Moreover, there's precedent for this kind of use being considered "transformative" and thus fair use. The Google Books project, which scanned millions of books to create a searchable index, was ruled legal despite protests from authors and publishers. AI training is arguably even more transformative.

While it's understandable that creators feel uneasy about this new technology, labeling it "theft" is both legally and technically inaccurate. We may need new ways to support and compensate creators in the AI age, but that doesn't make the current use of copyrighted works for AI training illegal or unethical.

For those interested, this argument is nicely laid out by Damien Riehl in FLOSS Weekly episode 744. https://twit.tv/shows/floss-weekly/episodes/744

You are viewing a single comment

View all comments Show the parent comment

But then it does go on to quote materials verbatim, which shows it’s not “just” ‘extracting patterns’.

Is is just extracting patterns. Is making statistical samples of which token ("word", informally speaking) is likely followed given the previous stream.

It can only reproduce passages of things it has seen many, many times. I cannot reproduce the whole work. Those two quotes can be seen elsewhere on the internet plenty of times. And it's fair use there, so it would be fair use with a chat bot as well.

There have been papers published where researchers were able to regenerate an image that was present in the training set of Stable Diffusion. But they were only able to find that image (and others) in particular, because they were present in the training set multiple times, and the caption was the same (it was the portrait picture of some executive at a company).

when given the book and pages — quote copyrighted works

Yeah, you are not gonna be able to do that with an LLM. They will be able to quote only some passages, and only of popular books that have been quoted often enough.

Even if they started to use my service to literally copy entire books?

You cannot do that with an LLM.

Why are you defending massive corporations who could just pay up? Isn’t the whole “corporations putting profits over anything” thing a bit… seen already?

I hate that some corporations are burning money, resources and energy on this, and the solution is not to restrict fair use even further. Machine Learning is complex, but if I had to summarize in some way is "just" gathering statistics of which word comes next (in the case of a text model). This is no different than getting a large corpus of text, and sample it for word frequency, letter frequency, N-gram frequency, etc. It is well known that this is fair use. You only store the copyrighted works to run the software and produce a very transformative work that is a summary many orders of magnitude smaller than the copyrighted work. This is fair use, and it should still be. Changing that is gonna harm the public, small companies and independent researchers way more than big tech companies.

As I said in another comment, I would very much welcome a way to force big corpos to release their models. Make a model bigger than N parameters? You needed too much fair use in one gulp: your model has to be public, and in the public domain. I would fucking welcome that! But going in the opposite direction is just risky.

I don't understand why small individuals think that copyright is their friend, and will protect them from big tech companies. Copyright will always harm the weak and protect the powerful as a net result. It's already a miracle that we can enjoy free software and culture by licenses that leverage copyright in our favor.

You cannot do that with an LLM.

If I want to go and read a Harry Potter book, I presumably have to pay someone something (excluding library services because those are services provided for actual people, not AI's)?

This LLM clearly has read Harry Potter and Chamber of Secrets, and is merely refusing to display the data it already has on it. "Data" in this case meaning the work, the book.

I'm not for current copyright laws, but I find defending these hypocritical companies despicable. I'm sure you're able to imagine that if it suited OpenAI, they might argue the exact opposite of what they're arguing. Companies don't really argue things in good faith, rather always arguing for the thing that will be the most profitable for them, no matter the veracity.

Yeah, you are not gonna be able to do that with an LLM. They will be able to quote only some passages, and only of popular books that have been quoted often enough.

You entirely ignored this part.

You basically proved my point in doing so, BTW. You cannot do what you claimed with an LLM. And I'm not saying, and I never said before "ChatGPT" or "OpenAI". I don't understand why you think that I might be "defending these hypocritical companies", when I literally said the opposite at the end.

You are entirely fooled by the output of ChatGPT and you are not arguing in good faith (or you are entirely unable to understand what I said).

Edit/addendum: And to stress out my point, given that the person to whom I've replied to showed the output of ChatGPT as if it were any kind of proof, this is what other LLMs say. This is 4o mini:

Large Language Models (LLMs) like me do not have the ability to quote whole sections of copyrighted texts verbatim. While I can generate text based on patterns and information learned during training, I do not store or recall specific texts or books. Instead, I can provide summaries, analyses, or discuss themes and concepts related to a book without directly quoting it. If you have a specific topic or question in mind, feel free to ask!

And this is Llama 3.1 70B:

Large Language Models (LLMs) can generate text based on the patterns and structures they've learned from their training data, which may include books. However, whether they can quote whole sections of a book depends on several factors.

LLMs are typically trained on vast amounts of text data, including books, articles, and other sources. During training, they learn to recognize patterns, relationships, and context within the text. This allows them to generate text that is similar in style and structure to the training data.

However, LLMs do not have the ability to memorize or store entire books or sections of text. Instead, they use the patterns and relationships they've learned to generate text on the fly.

That being said, it's possible for an LLM to generate text that is very similar to a section of a book, especially if the book is well-known or widely available. This can happen in a few ways:

Overlapping patterns: If the book's writing style, structure, or content is similar to other texts in the training data, the LLM may be able to generate text that resembles a section of the book.

Memorization of key phrases: LLMs may memorize key phrases, quotes, or passages from the training data, which can be recalled and used in generated text.

Contextual generation: If the LLM is given a prompt or context that is similar to a section of the book, it may be able to generate text that is similar in content and style.

However, it's unlikely that an LLM can quote a whole section of a book verbatim, especially if the section is long or contains complex or unique content. The generated text may be similar, but it will likely contain errors, omissions, or variations that distinguish it from the original text.

Feel free to give them a shot in: https://duck.ai

First off, this been bothering you for a month? O.o

Secondly, I personally can't steal a single book, popular or not.

I am arguing in good faith. The training data obviously has copyrighted works in it.

These companies are being treated differently than if you personally used copyrighted works without paying.

https://www.copyright.com/blog/heart-of-the-matter-copyright-ai-training-llms-executive-summary/

Using Copyrighted Works in LLMs LLMs use massive amounts of textual works—many of which are protected by copyright. To do this, LLMs make copies of the works they rely on, which involves copyright in several ways, such as:

Using copyright-protected material in the training datasets of LLMs without permission can result in the creation of unauthorized copies: copies generated during the training process and copies in the form of representations of the training data embedded within the LLM after training. This creates potential copyright liability.

Outputs—the material generated by AI systems like LLMs—may create copyright liability if they are the same or too similar to one of the copyrighted works used as an input unless there is an appropriate copyright exception or limitation.