3 ms·
I'm surprised you're confident they don't. You think someone's picking up a first edition Steinbeck and destroying it so it gets into the next training corpus?
by kfrzcode 1mo ago
I'm surprised you're confident they don't.
You think someone's picking up a first edition Steinbeck and destroying it so it gets into the next training corpus?
Or is it like, technical manuals and incredibly dry almanac content?
I think the details actually matter here. I'd love to see a realtime list of these titles being scanned and processed. Then we'd know if anyone actually gives a care.
That said I'm willfully ignorant of most "headline news" so I'd be curious to see how heinous the issue really is.
- ForHackernews 1mo ago> You think someone's picking up a first edition Steinbeck and destroying it so it gets into the next training corpus? I doubt the low-paid workers stripping, scanning and shredding the books are checking what anything is. I'd imagine they are dumped with a big crate of old books and an hourly quota they have to rapidly process. Arguably old technical manuals with limited print runs are even rarer or more valuable to historians. We have lots of copies of Steinbeck, albeit not first editions.
- qarl2 1mo ago> I doubt the low-paid workers stripping, scanning and shredding the books are checking what anything is. You don't think they are, at the very least, checking to see if the book has been scanned before? A search like that could very easily be extended to check for rare books too. Why do you assume it's not being done?
- voidhorse 1mo agoWhy assume it is? We don't know, and the fact that the process is so opaque and would have been totally undisclosed if not for journalists doesn't exactly make me confident they are "doing things by the book" on this one.
- qarl2 1mo agoBecause you usually find out if someone is doing something before you get mad about it. No?
- voidhorse 1mo agoYou should already be upset. A company with far greater resources than you has potentially prevented you from accessing a portion of the world's information. Because they haven't been transparent, you don't know. You should react to the facts, which are that a company has taken moves, largely surreptitiously to prevent you from accessing intimation. There is zero reason to assume good will in this scenario. You should demand more transparency and more insight into operations, period. Unless you have a vested interest in the company.
- qarl2 1mo agoYou understand that what you just said is true of almost every company in the United States... right? Sorry, but I'm going to save my efforts for the people who are ACTUALLY doing bad things - not just the people who I don't like who MIGHT be doing bad things. Otherwise it kinda seems hysterical. IMHO.
- voidhorse 1mo ago> You understand that what you just said is true of almost every company in the United States... right? No. It definitely isn't. There are plenty of US companies, one of them large and recognizable, that actually make it their mission to give you access to information in some fashion. We have different definitions of "bad thing". In my opinion, unilaterally potentially erasing human history is already bad. Is it the worst thing someone could do? No, probably not, but that doesn't mean I shouldn't be bothered by it. I'm not sure what kind of argument you're really trying to make here. I don't find LLMs that valuable, a few companies being the only providers of "frontier" models bothers me, and their possibly destroying works of historical interest also bothers me. Until they come out and find a way to make their scans available to the public, I'm not going to be happy about this. In my opinion, you shouldn't be happy about it either. These companies have everything to gain by making it harder for you to access knowledge (so that you are forced to pay them continuously for their "knowledge" in the form of information sparse LLM tokens). They have almost nothing to gain by enhancing your access to knowledge.