The Vanishing Page: AI Firms Scan Then Destroy Rare Book Editions | Hacker News Reader

The Vanishing Page: AI Firms Scan Then Destroy Rare Book Editions

dallasexpress.com

50 pointsthrow0101a75 comments

Humanity hasn't caught up to the reality of the times; we should adjust copyright laws so anything over say 10 years old can be hosted for free on the Internet. We should have colossal databases of music, books, movies, etc. that can be freely traded and downloaded without breaking any laws. Maybe in the past the commerical protection of art and IP was necessary but with the Internet and AI I think we need to, as a species, get over it and focus on information archival and dissemination over protecting income streams.

10 years may be too little but generally I agree.

What's blowing my mind lately though is the people complaining about copyright today are some of the same people I saw 30 years ago screaming "Information wants to be free" and running around with the DeCSS code on their T-Shirts. But suddenly, now that they hate AI, copyright is important and the information should be locked away. I can't wait for this all to settle down.

It's possible to be opposed to excessive copyright and also be opposed to the mass theft of media for profit and subsequent abuse of copyright that the scrapers are guilty of. The AI pillagers are taking everything in, then gatekeeping access. And as they acknowledge with their "nothing after 2022" preferences, they (or their users at least) are salting the earth for future seekers of knowledge.

> It's possible to be opposed to excessive copyright

Define excessive. Who makes that determination and how?

the people who are opposed to it

There's a substantial economic literature on this topic.

Among other factors: time value of money and future-value discounting mean that economically ever-expanding copyright terms offer virtually no present-value benefit.

Most works see virtually all of their economic value in the first few years of publication. There are some (rare) long-tailed exceptions. The original US copyright term of 14 years, with extensions, actually fits the economics pretty well.

What extreme copyright duration does do is:

1. Create vast copyright-holding monopolies, often of works for which authors were poorly compensated if at all. (For the latter: academic publishing.)

2. Create a vast category of "orphan works" which aren't or cannot be published, in the latter case because rights simply cannot be clearly established, or competing claims (survivors, estates, publishers) exist.

"Seventeen Famous Economists Weigh In On Copyright : the Role of Theory , Empirics , and Network Effects" <https://jolt.law.harvard.edu/articles/pdf/v18/18HarvJLTech43...> (2005)

"The true impact of shorter and longer copyright durations: from authors’ earnings to cultural creativity and diversity" Jimmyn Parc & Patrick Messerlin (2020) <https://www.tandfonline.com/doi/full/10.1080/10286632.2020.1...>

Orphan Works: <https://en.wikipedia.org/wiki/Orphan_work>

People must realize that it is easy to built some sentiment against on part of legislation that seems to restrict them, however, it is much more difficult to establish legislation that serves everyone. In the end we need smart people in politics and lawmaking instead of populists and libertarian cry-babies.

far too often people care about things for purpose, not because they care. There are things of equal sorrow that don't get cared for because it serves no purpose. If your cause isn't ammunition for something that people already want to do then it will struggle to gain traction.

For example, nobody cared about data centre water or electricity use when it was serving us up cat pictures. Collectively the past 20 years of doing that is still significantly more use than AI data centres so far. We suddenly care about it because its anti-AI ammunition, not because we care.

People suddenly care about data center water and electricity use because a lot more data centers are being built now (there are more under construction now in the US than the total we had in 2020).

Things being done as part of a massive boom are usually harder to accommodate then things done as part of longer term growth and data centers are no exception.

I don't think the average person is poring over data centre plans and requesting blueprints from local governments using FOI requests. The average person is understandably apprehensive about AI and water and electricity use serve as a convenient attack vector.

I don't think I've ever seen people query with such fervour; the energy usage of our existing internet structure. or supply chains to provide us all with electronic devices, despite them being inherently problematic in similar ways.

I was in a British supermarket today and there were apples on the shelves that were transported there from New Zealand. Again, no protests. Do we really care about energy usage or do we simply find convenient ammunition to attack ideas that we are already apprehensive about? We live in a world that is overheating as a consequence of the energy bounty of fossil fuels and yet people are still driving cars, SUVs are more popular than ever. There is no global carbon tax in sight while every nation on the planet slinks from their responsibility for all the co2 in the atmosphere, blaming each other and charging towards our scientist's worst case outcomes. So who really cares about energy consumption on its own?

There are no such people. The people who object to ai's violation of copyright today, always cared about copyright.

They validated the concept of copyright by employing it themselves to create the gpl, to make information free and arm it's ongoing freedom.

I imagine you're in favor of the artists being compensated in some other way then? Or should they just suck it up?

I think it’s pretty clear in GP’s first sentence that they believe artists should get compensation during the first 10 years. I personally would change it to something more like 20 years but I share the idea.

Yeah I could be convinced to go to 20. And I think the rights shouldn't be able to be held by companies, only creators specifically.

I think part of the problem is that current copyright law doesn't compensate artists, only mega corporations who figured out the right bureaucratic tricks (copyright assignment) to get paid in lieu of the artists.

After 10-20 years, yeah, I think they should suck it up.

On the surface it seems bad, although many of these books were probably rotting in place rather than being read. If the companies are willing to make the digital version available this may actually prreserve the books.

More troubling to me is the focus on pre-2022 data. This indicates that the companies are seriously worried about model collapse, where the future models become worse becuase they are being fed by generated data instead of real human data.

This is actually my worse case scenario for AI: the models become good enough to disrupt entire industries in the next 10 years, but then stay frozen at that level. Model drift then kicks in because the world will keep changing but the models do not, and in a few decades we actually regress instead of progress becuase they won't be enough skilled humans to drive advances.

Yeah, I really don't know where this goes from here. Like certain sources, particular Reddit and Twitter, have to largely be considered spoiled at this point. It's hard to quanitfy the amount of bots but my intuition tells me it's really high.

But changing how non-AI people write, that's an interesting angle. Because where do we go from here? In 100 years will we still be overvaluing pre-AI sources? That doesn't make sense. Of course a lot can (and will) change in 100 years.

But i'm reminded of pre-atomic steel, which is steel made before the first atomic bombs were detonated and thus have really low background radiation. This is necessary for making MRIs and such. People will go and find it from shipwrecks and such (ironically, many of which are from WW2). It's also a finite resource. What happens when we run out?

Now pre-2022 texts aren't consumed (other than destructively scanning books of course) but it is also finite. We can't make more of it.

As long as economics incentivizes next quarter thinking nothing will be done. We can't even handle human portion of global warming which is arguably more important than just some AI model collapse scenario.

Entropy is a thing.

Ancient Rome stopped creating aqueducts because they had all the ones they needed. They failed to pass that knowledge on to the next generation and so they just forgot how to create aqueducts.

Once AI starts making the majority of content, people will simply forget how to make content. Before we know it we're all fat slobs in floating chairs like in Wall-E.

Best case scenario I think is similar to what we see in Ian M. Banks Culture books where the machines basically take care of us out of the goodness of their hearts and we just kinda fuck off into obscurity.

That's the western world right now with manufacturing. Meanwhile China did the manufacturing and India is coming up, so it's not a lost skill globally.

The next generations are also losing computer skills, being brought up with devices that handhold you and have no opportunity for customisation and are essentially thin clients for mega corporations.

What would happen if we created large volumes of books with AI but faked their dates before 2022 to induce model collapse?

There is good discussion on the previous thread from 6 days ago: https://news.ycombinator.com/item?id=49068738

It’s been stunning how often this story is reposted from various news sources across Reddit and Hacker News.

Each time it’s some different outlet - but when you dig in, the piece is just verbal framing around the original story written by 404 media:

https://archive.is/9MQrK

What’s even more stunning is that the original article doesn’t provide evidence that “rare” books are being destroyed. That doesn’t even appear in the original article title.

Yes, since we complained about them violating copyright so much, it does not violate copyright if they destroy the original, so they are now destroying the original. We got our demands met.

That's not how copyright works, at all.

Yes, actually it is

I'm going to buy a Taylor Swift CD, rip it as MP3s, destroy the CD, and put the MP3s online...

Will I get sued?

The "putting it online" part is where you are f-ing up.

But it's transformed now!

Edit(seriously): >Transformativeness is a characteristic of such derivative works that makes them transcend, or place in a new light, the underlying works on which they are based. In computer- and Internet-related works, the transformative characteristic of the later work is often that it provides the public with a benefit not previously available to it, which would otherwise remain unavailable.(Wikipedia)

I am sure Napster or the like argued that what they were doing was transformative.

Not sure how AI companies are arguing about PUBLIC benefit if you have to pay...

It has to be sufficiently transformed. Simply scanning it doesn't qualify. The legal system by contrast is leaning toward training of AI as being sufficiently transformative.

Destroying your copy of a copyrighted work does not violate copyright.

Format shifting is allowed. Making copies isn't. By destroying the original, it's format shifting.

Right. But destroying a book even without making a copy is also allowed. It's very very common, in fact; something approaching a billion books are destroyed each year in the US.

It absolutely is. By destroying the original they retain the one copy they were licensed. It explicitly discusses that this was judged fair use in a court of law, because of the destruction of the original, as this becomes a transformation of the original work rather than a duplication.

Don’t hate the player, hate the game.

So what are they doing about the online data they ingested?

Copying it. Because they clearly believe that they’re allowed to do so.

For physical books, they believe they’re not allowed to, hence the destruction.

It’s abhorrent, I agree… maybe if your world domination plan involves destroying rare books, you should change the plan rather than saying “well our hands our tied, the law says we have to”

I mean they used to have a different plan, which involved not destroying the books, and then they got sued for not destroying the books, by us, so now they are destroying the books. What did you think would happen?

They got sued for violating copyright in general, not for “not destroying books” in particular. Destroying the books was a legal trick they used to continue scanning them without breaking the law.

If I were running the AI company and my lawyers came back with “teeeechnically we can still scan them if we burn them after”, I would reply with “I guess we’re not scanning them”, but I lack the sociopathic instincts required to be a tech CEO I guess.

The sociopathy here is describing a perfectly acceptable activity as somehow evil.

You should be ashamed of yourself.

Indeed, won’t someone please think of the poor AI company CEO’s, we should be ashamed of ourselves for criticizing their perfectly acceptable destruction of our creative works. Thank god you’re here to swoop in and defend them at every turn.

When Google scanned its books, it did not destroy any of the books. This was because of the idea of the "cultural object" where the book has value as a physical artifact. The experience of reading a physical book is different than reading an electronic text. There is an argument that reading a book in its original form is better because it is closer to the original experience. This can also be said of records and tapes. The experience is designed to work in the original format. A book is a designed object with cultural significance. The container for the words can be as important as the actual words. There should be clearer distinctions and policies between mass market production where items can be destroyed and there will still be many of them left and unique and original content with limited copies available. This is not hard to do. Not to do it shows carelessness.

It’s true that they’re destroying old books. The claim that they’re destroying rare books seems to be repeated without much in the way of evidence.

I'm not really the most AI-friendly person around but this news cycle is just as equally annoying.

"Rare and out-of-print" is fabricating a lot of aura here. It's technically correct (the best kind of correct) but I've yet seen evidence that these are culturally significant copies being destroyed for scanning.

From https://nltimes.nl/2026/06/25/rare-book-dealers-fear-tech-fi...:

> The attachment contained 3,000 English-language titles organized by ISBN number, including books such as Distinct Element Modelling in Geomechanics by K.R. Saxena (1999); Barrett's Traditional Fairy Tales (2021), an academic study of Irish folklore; and Laser Shock Peening of Advanced Ceramics by Pratik Shukla (2018).

The oldest is from 1999! If that's the best they can actually enumerate to bolster this outrage farming cycle you could just wonder how irrelevant the rest are.

Really, please, kill this news cycle. There's a lot of issues deserving proper attention right now and this one is a straight-up nothingburger.

There are plenty of scanners that do not destroy the book being scanned. Here is one, I am sure there are many others: https://www.youtube.com/watch?v=b9LcTZU-HHI

It should be more than obvious, that the problem are not the scanners.

https://news.ycombinator.com/item?id=49127284

Forever preserving the content of a rare book by consuming one (1) copy is the kind of thing we should always be doing more of.

Roughly nobody would have had access to that copy. Roughly everyone can now benefit from its content.

Where can I read any of these destroyed copies from anthropic?

What do you imagine a rare book is? It's not the only extant copy of the writing in that book, in fact it's not even like it represents a significantly limited accessibility of that book.

Roughly Anthropic owners benefit from the content, unless the scans are made available somehow.

> Roughly everyone can now benefit from its content.

You mean, roughly everyone who uses that particular model provider’s products. For truly rare books, this has an anticompetitive flavor, since it ensures others can’t train models from the same knowledge.

I have a deep rooted doubt that "roughly everyone can now benefit from its content" If there's any value in the content it will immediately be monetized with an aggressive pay structure and had it sat in a library, that same value would be free.

One (1) company destroys and (illegally?) copies a rare book. This knowledge is now of that company, not of humanity. Other companies will feel the need to do the same and before you know it knowledge is walled off in the gardens of the AI companies, and the rare books are gone and destroyed.

The scan doesn't go away, and can be released when the copyright expires.

Oh no, a lump of cellulose is gone. Will no one think of the fibers?

> can be released when the copyright expires

But it ... won't be? The companies have no motivation to do so. The copyright on lots of these books are surely already expired, so if they wanted to they could be putting these up now. I'm sure internally this is viewed as a corpus of knowledge they have that their competitors do not, so they will not release them unless something forces them to.

Yes, without any justification let's assume the worst. This is true good faith arguing.

We have a partial list from https://nltimes.nl/2026/06/25/rare-book-dealers-fear-tech-fi...:

Even if the authors died immediately after publication there is no way the copyright has expired.

Only if they even sell copies. Which they won't. They'll sell the 'intelligence' trained on that knowledge.

When are you or anyone more likely to get access to a rare book:

a) When a company has made a digitized copy of it

b) When a few previous hard copies are kept somewhere in the world

Idk what Anthropic plans to do with this, but note that my original comment was not about Anthropic at all. I simply think it's a good tradeoff to destroy a few copies of rare books if that helps digitizing them. I am not against simply legally forcing companies that do this to release those copies in due time. In the meantime I am also happy to have them as part of the LLM corpus.

> illegally?

No, copying rare books is very likely not illegal, because, presumably, they are quite old.

When I ask for it in the royal library.

It's going to be hard for you to do more of that, considering that AI forms are destroying the books. You know, the whole point of the article...

> Roughly everyone can now benefit from its content.

You mean every Anthropic?

They should at least give the scanned copies to a non-profit or government entity to preserve and release them when the copyright expires.

Why don't they just put the pdf versions on Internet Archives? Then they would at least be preserved for the future

That would be illegal.

Why don't they rebind them and put it back in some library? They don't have to throw it out.

Because legally then it's a copy and not a format conversion.

This story has a lot of details left out and is a magnet for illinformed commentators: the reason the originals are destroyed is because it is legal to do so for the purposes of format conversion, but not legal to retain the originals under copyright law.

Who knew Ray Bradbury could be so wrong.

Not all books are sacrosanct. There doesn't seem to be any evidence that they are destroying rare and valuable books.

Nowhere are the supposed rare books listed.

Confuses "(not all that) old books" with "rare books".

Ragebait, nothing more.

They are academic titles from the last few decades for the most part, and are rare because the print runs were typically short, and the prices typically eye-watering. “Semantic analysis of cotton processing terminology in Des Moines, 1993-1996, 8th edition”, that sort of fun.

I'm reminded of a quote that's been (mis)attributed to Dorothy Parker.

"This is not a novel to be tossed aside lightly. It should be thrown with great force."

Yay capitalism! Imagine how much value is being created for shareholders by this destruction! Another example of "great work" being done!

Jokes aside, I don't know what I would do in this situation either. Maybe find a way to mark "rare" or "out of print" books and sell/auction the "pages"? Or make the scanned book available via some mechanism?

You could “ingest” them all into a “model” and sell access to the information within them via “token usage”

Reply on news.ycombinator.com