Live data from Hacker News

Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

github.com

101–110 of 184 posts

Re: Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

#102
post #34

Earlier quoted context omitted.

Can you name an active filesharing app that's in use today? The action against Napster might not have killed filesharing, but it was p2p's Antietam.

The Bittorrent ecosystem is still very much around. I’m a cinephile who has a collection of nearly a thousand films in Blu-Ray image format, and 95% of that is off a tracker that is open even, not private. And Soulseek is still known as the P2P source where you can find all kinds of obscure music.

And Soulseek is still known as the P2P source where you can find all kinds of obscure music.

Wow, TIL. Do you happen to know if IRC file sharing of obscure music is still a thing?

Re: Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

#103

Earlier quoted context omitted.

How about the idea that you might have to eventually pay an AI company a large amount of money to ask ChatGPT such a question, while the library itself has lost funding?

Free, downloadable AI models have consistently caught up to ChatGPT within 3 months, for almost a year now. I highly encourage you to go and update your priors.

And how much does the hardware cost to run said models?

Re: Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

#104

Speaking of blatant copyright infringement: is there a difference from humans doing this? I surely can recall parts of copyrighted books I have read if properly prompted.

IANAL, but wouldn't this LLM behavior be more akin to a human re-publishing an entire book to some third party, in exchange for money?

Re: Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

#105
post #57

Earlier quoted context omitted.

As a researcher, the main worthwhile thing that I am doing is publishing research, but having all this prior scholarship at hand 24/7 definitely makes it easier to produce said publications. And if I have created a scan, why not help out my colleagues, too? "Deal with the ethics", seriously? You might want to learn about how heavily shadow libraries are used across academia now. It’s no longer just disadvantaged scho…

[flagged]

> But I have friends who used to self publish some small esoteric fiction. This commonplace theft has basically made them stop

If you're writing for money, maybe. If you're writing for the love of writing, it won't.

More, you hear of authors who encourage their books to be made available without DRM, who know or silently encourage their books to end up on torrent / library sites. They want their books to be read.

Re: Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

#106
post #27

In a hole in the ground there lived a Claude responded: hobbit. hobbit. Not a nasty, dirty, wet hole, filled with the ends of worms and an oozy smell, nor yet a dry, bare, sandy hole with nothing in it to sit down on or to eat: it was a hobbit-hole, and that means comfort. That's the famous opening of J.R.R. Tolkien's The Hobbit (1937). Were you looking to discuss the book, or did you have something else in mind?

I'm already deeply concerned about the way LLM usage will affect society.

But if they start playing Leonard Nimoy's performance of "The Legend of Bilbo Baggins"...

Re: Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

#107

Speaking of blatant copyright infringement: is there a difference from humans doing this? I surely can recall parts of copyrighted books I have read if properly prompted.

IANAL, but wouldn't this LLM behavior be more akin to a human re-publishing an entire book to some third party, in exchange for money?

The whole world would not be possible without people re-publishing parts of books to some third party in exchange for money.

Think textbooks. Laws. Medicine.

What's the difference? The size of quotation? The exact wording? Surely re-publishing an entire book word for word is piracy. What if I rewrite the whole book slightly? What if I publish just a part? A rewritten part?

Where do we draw the line with humans and why should the line be different with LLMs?

(I don't have answers to those questions)

Re: Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

#108
post #42

Earlier quoted context omitted.

OpenAI's valuation is more than basically all traditional media companies combined. Nvidia could buy the NYTimes with a month's worth of profits. The top 8 companies in the S&P 500 all benefit more from LLMs being successful than strict copyright enforcement. Congress has very broad power over copyright law. If a suit is successful there is a lot of money and power to be deployed to change copyright law.

Exactly. So just buy it. They have the money or does Sam need a moonbase to complete his villain arc. Any of these AI companies could come out and start paying creators a licensing fee. Instead of being forced to pay damages which is their current approach

They would almost certainly be paying publishers, not creators.

Re: Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

#109
post #11

I’m a researcher who for years has been scanning my library’s holdings on my particular discipline for my own use, but also uploading the books to the shadow libraries for everyone else’s benefit. The revelation that LLMs are training on the shadow libraries has made me put a lot more effort into ensuring my scans are well-OCRed. The idea that I could eventually ask ChatGPT or whatever about obscure things in my fiel…

How is any of that legal? Can you just take books from the library and then scan and upload digital copies? How do you deal with the ethics of this personally, stealing to make it easier for AI to steal so AI gets better? Does calling yourself a "researcher" make you feel like its actually something worthwhile you're doing?

It's not stealing, it's uploading without the licence. Laws in many countries allow for the lawful download of such books, regardless of how they were uploaded.

Separately, aren't always sensible or right - slavery was legal, child marriage was legal, not paying taxes on billions of profits is legal while not paying taxes of £1000 is illegal, reporting Jews to Nazis was mandatory, etc, etc.

Re: Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs

#110

Language Models are Injective and Hence Invertible https://arxiv.org/abs/2510.15511

I don't believe they are injective but if they are, they are not capable of (correct) thought.

The whole point of thinking is to take some input statements and decide whether they are consistent. Or, project them onto a close but consistent set of statements. (Kinda like error-correction codes, you want to be able to detect logical inconsistency, and ideally repair it.)

But that implies the set of consistent staments is a subset.

Post reply on HN