As AI tech companies increasingly buy and destroy books to feed to their AI models, Anna's Archive is calling for volunteers to help preserve them for the public record.

top 79 comments

sorted by: hot top controversial new old
[–] 62 points 2 months ago* (2 children)

Anna's Archive says ‘time is running out’ as ‘knowledge is permanently monopolized on private servers’

Not only that, of course, but easily edited. It's a "mandala effect" generator. AKA modern gaslighting. They name the thing, point to legitimate examples, then abuse our new "understanding" to cover their crimes. 1984 indeed.

Oceania was at war with Eurasia; therefore Oceania had always been at war with Eurasia

  • source
  • hideshow 2 child comments
  • [–] 58 points 2 months ago (2 children)

    Getting rid of physical copies of books so we can be monthly subscription slaves, disgusting

  • source
  • hideshow 2 child comments
  • [–] 16 points 2 months ago (1 child)

    It's worse than that. I'm sure they're getting wet dreams about eradicating as many books as possible so that they can finally sell AI slop books at a premium price, claiming that will be "the last way to get something in [author] style!".

  • source
  • parent
  • hideshow 1 child comment
  • [–] 42 points 2 months ago (6 children)

    Fucking scanning books without destroying them, the philistines! /s

  • source
  • hideshow 6 child comments
  • [–] 5 points 2 months ago (5 children)

    Is this gif from that tom hanks movie?

  • source
  • parent
  • hideshow 5 child comments
  • [–] 9 points 2 months ago (4 children)
  • [–] 8 points 2 months ago (2 children)

    No I know AI companies are doing that, I just don't think they would be making gifs about it.

    But there's a Tom Hanks movie that came out a few years ago where in the beginning of the movie he lives in a bunker because the UV index outside is so high the sunlight is basically radioactive, and he's building a robot and he uses a machine to chop the spines off of books to feed them into a scanner to train the robot brain.

    That all takes place in the very beginning and then he leaves the bunker with the robot and some shit happens. But that's what I thought this gif was from.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 31 points 2 months ago (1 child)

    I guess the next step is picking specific people to memorize specific books and pass that knowledge down to an assistant/acolyte who will keep the memories alive. Rinse and repeat. Fahrenheit 451 anybody?

  • source
  • hideshow 1 child comment
  • [–] 17 points 2 months ago (2 children)

    If you read the blog post, they don't provide help or guidance for regular bookworms on how to scan. No information about what they expect.

    Also, flatbed scanning a whole book takes many hours. It's one page at a time, line up each page, press the book flat and scan, for hundreds of pages.

    You'd likely need to buy, or have access to a book scanner. Which is fine, there are cheap models, but who knows if that would work for them? They don't say. It doesn't seem like they're expecting help from you or me.

  • source
  • hideshow 2 child comments
  • [–] 14 points 2 months ago (7 children)

    I own some rare books but no book scanner. What should I do?

  • source
  • hideshow 7 child comments
  • [–] 19 points 2 months ago (2 children)

    Those who want to conduct large-scale scans and upload many titles can reach out to the archive for support, as the archive says that it “can help pay for the scanning fees and other rewards.”

  • source
  • parent
  • hideshow 2 child comments
  • [–] 6 points 2 months ago (1 child)

    Hmm, it's just a few books but I do think they're pretty rare. Not something I'm willing to send away. I wonder if a library would be willing to help?

  • source
  • parent
  • hideshow 1 child comment
  • [–] 12 points 2 months ago

    My local library has some large format scanners. I've considered using them to do my photo albums. I have the individual photos scanned and backed up, but the layout and margin notes are worth preserving also.

  • source
  • parent
  • [–] 2 points 2 months ago (1 child)

    If your local library doesn't own a book scanner, any colleges near you may have one. You usually don't have to be a student to walk into the library, and the book scanners I've used don't require a login. YMMV on whether this is how it's set up where you are.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 3 points 2 months ago

    I might try this. Im concerned that some of the books may still be under copyright. Not all of my rare books are that old. I collect indie poetry books and comics as well as old religious texts/pamphlets.

    Also concerned that some might be literally some of the only remaining copies and I really don't want to risk damaging them.

  • source
  • parent
  • [–] 27 points 2 months ago (9 children)

    Remember, USA and it's AI companies did that.

  • source
  • hideshow 9 child comments
  • [–] 18 points 2 months ago* (5 children)

    Ah the American dream. Be greedy, selfish and only care about yourself.

  • source
  • parent
  • hideshow 5 child comments
  • [–] 3 points 2 months ago (2 children)

    Last time I checked, no other country(apart from Russia) wanted to burn rare books and destroy the world.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 21 points 2 months ago (2 children)

    I agree about not destroying books, but Anna's archive is ok with providing all of their info to llms

    https://web.archive.org/web/20260305153357/https://annas-archive.gd/blog/llms-txt.html

  • source
  • hideshow 2 child comments
  • [–] 9 points 2 months ago

    I am all up for starving machine learning of new knowledge but I view this as the least friction path. Otherwise, if the machines try to brute force their entry, what harm can kt cause?

  • source
  • parent
  • [–] 10 points 2 months ago (2 children)

    Any reputable orgs I can donate to for assisting the cause? I just don't feel I have the means otherwise to make a dent in the buying and preservation.

  • source
  • hideshow 2 child comments
  • [–] 1 point 2 months ago* (1 child)

    Perhaps there is a public library in your area that accepts donations? The one I live by is STARVING for support, they accept books, old computers, anything.

    They do good work for the community too, like free tech literacy classes, chess for kids, etc.

    But if you are dead set on an online NGO, then maybe archive.org? You have probably heard about their Wayback Machine, but they preserve books and other media too.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 9 points 2 months ago

    They're destroying everything like locusts

  • source
  • [–] 6 points 2 months ago (5 children)

    Anyone know any good book scanners was looking around and not sure what i should get.

  • source
  • hideshow 5 child comments
  • [–] 6 points 2 months ago

    Ages ago I had mimicked the "overhead scanner" with a mounted digital camera, but now I'd go with a mounted smartphone + light bars + bluetooth photo trigger. Scrolling through Bezos' was pretty depressing in terms of cost/quality and even a $150 device seemed like a repurposed webcam, with a sole non-free confirmed buyer saying it died after 3 months while freebies praise it lol. As long as you don't scan drawings/designs/comic books - just set lights evenly and that'd be enough, and even moreso if you OCR it and turn a scan into an e-book where visual fidelity stops to matter.

  • source
  • parent
  • [–] 1 point 2 months ago (1 child)

    I don't know about buying, but the ones I have seen basically hold the book open at a 45 degreeish angle, to not damage the spine. Then you press the pages with a 45 degree set of glass plates. The actual "scan" is a camera of some kind that slodes back.and forth between the two halves on a mount.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 4 points 2 months ago (8 children)

    the AI ran out of Reddit slop material to scan, because reddit has been purging alot of content/accounts even suspected of spamming, so they go for physical books now.

  • source
  • hideshow 8 child comments
  • [–] 2 points 2 months ago (7 children)

    gigo.

    books are better sources imho.

  • source
  • parent
  • hideshow 7 child comments
  • [–] 2 points 2 months ago (6 children)

    pretty much why they go for books.

  • source
  • parent
  • hideshow 6 child comments
  • [–] 1 point 2 months ago* (5 children)

    that's what I'm saying. I've never understood the hoopla of reddit selling it's corpus, aside from being fucking gross (spot on for spez) it's a shitton of garbage. there's gold in there, but it's absolutely buried in shite

  • source
  • parent
  • hideshow 5 child comments
  • [–] 3 points 2 months ago (2 children)

    For an AI to troubleshoot problems with your computer (something it's really good at today), it needs content like reddit's. Books don't and won't detail weird software issues and their workarounds. One the current training set becomes irrelevant to whatever software people are running, this is a specific area where Reddit's data is way more valuable than books.

    Of course Reddit locking down access to that information leads to fewer people contributing. Reddit is dead, it's just struggling to go down.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 2 months ago

    there’s gold in there, but it’s absolutely buried in shite

    yep, there is a lot of nuanced depth, but measured against the overall sea of chuds lol... and yep, the moment they started putting up walls and controlling the community it was doomed. it's gonna thrash for a while then, when the only ones left are advertisers and bots, it'll shit itself and stop twitching.

  • source
  • parent
  • [–] 2 points 2 months ago (1 child)

    im guessing reddit has been banning so aggressively lately, that AI scraping isnt getting as much "useful data" as much anymore (chatgtp/google dropped "references for reddit in thier slop summaries). so they are trying to force logins now to keep up with the user generated content so they can squeeze every last penny they can from reddit, i dont think ads were ever a big presence in reddit? its the data. people try to post content, like posts, subreddits, those can instantly get removed by redit filters. i think lost alot of traffic recently to, which was likely AI scraping.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 13 points 2 months ago (5 children)

    These AI companies are just being Nazis, books on pyres…again.

  • source
  • hideshow 5 child comments
  • [+] -8 points 2 months ago (4 children)

    just because it looks similar doesnt mean it is similar you thetralic midwit

  • source
  • parent
  • hideshow 4 child comments
  • [–] 3 points 2 months ago* (2 children)

    Similar: having characteristics in common.

    Looking similar does mean that two things are similar. You could have tried to make your point in a variety of ways, but you chose the way that can be countered by simply understanding English.

    0/10

  • source
  • parent
  • hideshow 2 child comments
  • [–] 0 points 2 months ago (1 child)

    Okay I'll give you a hint. The bad thing about the nazis burning books wasn't that they were destroying physical copies of books. I'll give you a few keywords intent, selection, political control

  • source
  • parent
  • hideshow 1 child comment
  • [–] 1 point 2 months ago

    Why be condescending now?

    You already fucked up the point you were making. By all means, insult people's intelligence at the start of the conversation, before you fuck up. Doing it at the same time as fucking up looks bad... but insults after you fuck up looks even worse.

  • source
  • parent
  • [–] 3 points 2 months ago

    I hate this timeline... It's things like this that make me hope we actually do live in a simulation, and the devs are about to release a major bug fix...

  • source
  • [–] 3 points 2 months ago (1 child)

    With modern phones having a document scan option with our phone cameras, is it possible we can scan and combine into a PDF file?

    I am going to test it myself, see how easy it is, and find a way to ensure anyone can follow with low cognitive load, but hopefully it’s as easy as I think to do.

    The sad part, is to do so would require a slow tedious task for books with hundreds of pages. It would be nice to see a large group tackle chunks at a time to the point we eventually have groups dedicated to genres and scouring to find them to preserve them.

  • source
  • hideshow 1 child comment
  • [–] 4 points 2 months ago (3 children)

    So, my question is, of all these books being scanned and then destroyed, aren't there many more copies of these books? And BTW, knowledge isn't permanent. It can be lost.

  • source
  • hideshow 3 child comments
  • [+] 10 points 2 months ago* (last edited 2 months ago) (3 children)
  • [–] 1 point 2 months ago (2 children)

    When someone can't answer a question a common response to change the subject. Do you have an answer to my question?

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 2 months ago (1 child)

    They just cut and bin the books after they've been scanned, the bins are taken to the local disposal, most municipalities have a garbage dump, some burn the trash.

    An ai trained on unique data is more valuable if you don't share the data it was trained on to your competition, your LLM gains an edge.

    If I found a unique handwritten book from the 18th century, than I proceed to read it and memorise then burn it, I'll be the only one to posses the knowledge of it.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 0 points 2 months ago

    There are very few, if any, rare, one of a kind books that would go through this process. If one of these books were scanned for an LLM it would be done delicately in a more time consuming manner and the book would be returned.

  • source
  • parent
  • load more comments (-1 replies)
    [–] 1 point 2 months ago (1 child)

    Okay, but aren't there more than 1 copy of the books they get their hands on? Or do they steal from museums?

  • source
  • hideshow 1 child comment
  • [–] 0 points 2 months ago (4 children)

    I have a feeling this is click bait...

  • source
  • hideshow 4 child comments
  • [–] 3 points 2 months ago (3 children)

    The headline does not contain any clickbait-typical elements, for instance

    • Most of The Words Capitalized
    • A question mark at the end
    • "You won't believe'
    • etc.
  • source
  • parent
  • hideshow 3 child comments
  • load more comments
    view more: next ›
    49 comments in c/piracy@lemmy.dbzer0.com view that post
    10 comments in c/technology@lemmy.zip view that post
    2 comments in c/books@lemmy.world view that post
    2 comments in c/technology@beehaw.org view that post
    1 comment in c/technology@lemmy.ml view that post
    1 comment in c/technology@midwest.social view that post