whats the point of scanning fiction and old shit lol so it can roleplay better ? roflmoa
No so it has more data, context, and information of it.
Be no good if you asked it about a summary of say Harry Potter, characters in it, or anything and it goes: Sorry, I have no idea what that is. All I know is historical facts.
It also learns more about contextual situations it might not know from purely factual data, new ways of speaking, more exposure to sentence structures and prose. It can infer from it new connections between words and in an abstract way, understand human psychology on what and why they write fictional things, and why others enjoy it and what they get from it.
Tons of reasons.
Do people write fiction to solely teach people to roleplay better, or read it themselves to learn how to roleplay better?
-We don't control what happens to us in life, but we control how we respond to what happens in life.
-Hard times create strong men, strong men create good times, good times create weak men, and weak men create hard times. -G. Michael Hopf
Disclaimer: Post made by me are of my own creation. A delusional mind relayed in text form.
Scanning of books is not "nazi behavior", the nazis were trying to erase history, AI companies are scanning books. you can not scan a book without destroying it, you have to removed the bindings, cut the pages into loose sheets.
This is yet another lie to make these companies look bad. Not saying they haven't been doing anything bad, they have, but a lot of what is spread is half-lies like this, did they destroy books, yes, was it like a book burning, not even close.
Anthropic got hit for torrenting books, they decided to buy millions of books (second hand mostly) scan them. They had to be discarded to, this is purely from a legal standpoint, its basically to get around having it as a digital copy instead of the physical, which is now discarded, they can't keep both legally, Anthropic argued (and won i believe) that they had just changed the medium to digital.
Keeping books which are now loose sheets would kind of suck anyways. one million books which are loose sheets, how fun to save, but i guess they must to not get the nazi stamp lol
his only argument in the video is maybe one of the books they bought was "the last copy" and then somehow people draw the conclusion that this is very close to how the nazis were trying to erase every single copy of books. give me a f'ing break.
Last edited by vurt on Wed, 5th Aug 2026 08:58; edited 1 time in total
funny how doing the opposite of what the nazis did is now nazi.. like spreading of information, making sure its digitalized to never go away = nazi. Hahaha. these people sure knows how to draw conclusions out of their ass.
The narrative in sweden has been the same in MSM. "Tech companies are buying books to destroy them".
lol... they are so full shit, like always. When the only argument they have is to lie about something to try to make it look bad, then you know they have absolutely nothing. If you want to have an argument it should be about torrenting of books, that actually IS an argument that could be used against them.
HOLY SHIT ... there isnt enough power for all these data centers who bought ram yesterday .... they wont get permission, waiting list is now EIGHT YEARS!!! when ddr6 is shitting on ddr5 they will have power for their ddr5 LOOOOOOOOLLLLLL
and there is not enough power to insure for extra frosty winters ...
and a data center wont get more efficient as they already have peak efficiency chips n stuff
@PickupArtist
To save you the typing. why not from now on just post this on any tech, ai, crypto, game news, or such topic when you share a video
It sums up your replies pretty well. And saves you all the typing
Spoiler:
-We don't control what happens to us in life, but we control how we respond to what happens in life.
-Hard times create strong men, strong men create good times, good times create weak men, and weak men create hard times. -G. Michael Hopf
Disclaimer: Post made by me are of my own creation. A delusional mind relayed in text form.
It seems amazons AI shopping assistant helps with anything....decided to ask it a random stupid basic coding test question to see if it would, and it actually answered.
Spoiler:
You can even make it do images...I think they probably want to fix that.
Spoiler:
-We don't control what happens to us in life, but we control how we respond to what happens in life.
-Hard times create strong men, strong men create good times, good times create weak men, and weak men create hard times. -G. Michael Hopf
Disclaimer: Post made by me are of my own creation. A delusional mind relayed in text form.
It's probably just an out of the box model with little to no extra fine-tune or reinforcement training, so it's going to do that, yeah. Without that, just them having instructions/system prompt isn't going to do enough to override it's general behavior.
been quite amazed with how good chatgpt SOL is though + 2 resets in like 2 days which is great. so many funny and woah moments while working on a new game.
I had an image file which i was going to implement in a certain way, just called "lower.png" when it was doing something else it found that image, looked at it (its just the lower part of a bulkhead, kind of long steel part) and it just implemented it straight away without asking how i wanted it, and it did it exactly how i had intended it which is kind of complex.
I was just about to create long prompt for how to do it, not needed, that kind of stunned me, like "are you a fucking mind reader?"
The numbers don't decide, the system is a lie. The river running dry, the wings of butterflies.
And you may pour us away like soup. Like we're pretty broken flowers.
We'll take back what is ours. One day at a time.
Probably pattern matched with other projects. I'm guessing it was something that had a logical use and fortunately it worked out. Could have just as well done something you didn't want with it to be fair.
The big models especially have gotten good at figuring out what people want, just because of how much data they have on unclear interactions with AI and poorly worded requests.
yes something like that, unsure if the method i was going to ask for is common, but it could be.
its overall very good at understanding even if you mess up your prompts horribly. There are so many times where i write "can" instead of "cant" so what i am asking for becomes the opposite, but it still understands and just does it right.
Jason Kelce Stars in Anti-Data Center Advert: ‘We Want Your Pee’
Former Philadelphia Eagles center Jason Kelce has appeared in a new advertisement criticizing the amount of water used by artificial intelligence data centers and offering an unusual solution.
The campaign, released Tuesday by Garage Beer and Liquid Death, shows Kelce apparently urinating into a jar as he tells viewers: "AI data centers waste millions of gallons of water. That's why Liquid Death and Garage Beer have decided to team up. We want your pee … to cool these data centers."
Kelce then performs a song asking for "endless gallons of pee" and encouraging the public to send urine to an AI data center.
@madmax17
Ai does not "use up" water to cool it.
Saying AI uses (consumes) water to cool it by the gallon. Is (sort of) like saying your water cooled PC "used" 60 gallons of water to play GTA6 for an hour. Because it pumped about 60 gallons of water through the waterblock while you played.
It's not exactly same-same. But the claims is overblown and misleading on purpose. And the public perception of it is framed that way to carry a stigma that is erroneous and stated poorly with intent.
Water usage is a concern. BUT how most people opposing it phrase and frame it are 'technically right' but worded specifically in a way to make it seem far more 'threatening' than it is.
And again, I am not defending data centers. As I say: 100 things to dislike them for. This one is misleading, misrepresented, and flagged higher that it should be to make a 100 + 1. that isn't (shouldn't be) even in the top 10 reasons out of those 100.
Hate them all you want, but hate them for valid reasons. Just because it's feed to you in a simple way to digest compared to more complex but more important but intangible valid complaints against them, doesn't mean you should swallow it.
-We don't control what happens to us in life, but we control how we respond to what happens in life.
-Hard times create strong men, strong men create good times, good times create weak men, and weak men create hard times. -G. Michael Hopf
Disclaimer: Post made by me are of my own creation. A delusional mind relayed in text form.
it's a closed loop since quite a while back, MSM tries to act like nothing has evolved. they want to rile people up, it creates discussions and most of all clicks, which means money.
people being angry is very profitable, happy people or people who's like "meh" doesnt do anything for them. they have figured this out ages ago.
data centers started with the internet, so don't be on the internet if you want to be some sort of "anti-data centers activist". but its like always - fake activism, just trying to virtue signal, trying to come of as being above others.
Yea. while I am in no way FOR mass and corporate/govt AI. Nor massive data centers just for such purposes. WE know they will exploit them far more than the purposes stated they made them for.
Just bugs me it seems the main complaint spread, is a mostly incorrect one simply because it's the easiest to weaponize. When the more important reasons to be against them is MUCH more worrisome and dire.
Its making people focus on arguable points, instead of focusing on more concrete indefensible imminent opposition points.
-We don't control what happens to us in life, but we control how we respond to what happens in life.
-Hard times create strong men, strong men create good times, good times create weak men, and weak men create hard times. -G. Michael Hopf
Disclaimer: Post made by me are of my own creation. A delusional mind relayed in text form.
Last edited by DXWarlock on Wed, 19th Aug 2026 17:30; edited 1 time in total
@madmax17
Ai does not "use up" water to cool it.
Saying AI uses (consumes) water to cool it by the gallon. Is (sort of) like saying your water cooled PC "used" 60 gallons of water to play GTA6 for an hour. Because it pumped about 60 gallons of water through the waterblock while you played.
It's not exactly same-same. But the claims is overblown and misleading on purpose. And the public perception of it is framed that way to carry a stigma that is erroneous and stated poorly with intent.
Guess that depends on the data center itself, different types are out there.
Yeah there is, but none consume water.
The worst case is open air evap type: water isn't "destroyed." It evaporates, condenses, rains back down.
No worse than an open body of water 'consuming' water by evaporation, or how surface water also works on the earth to start with.
And the standard usage data used all the time comes from: "AI query = X bottles of water" Liquid Death drink brand beer commercial. It's a drink company marketing. Not actual credible data. The Liquid Death ad is closer to dishonest corporate activism, than analysis.
-We don't control what happens to us in life, but we control how we respond to what happens in life.
-Hard times create strong men, strong men create good times, good times create weak men, and weak men create hard times. -G. Michael Hopf
Disclaimer: Post made by me are of my own creation. A delusional mind relayed in text form.
So a few weeks ago I got myself a subscription to Claude AI, and I got to say... I'm completely addicted
Like vurt says, it's been very impressive how it interprets and 'understands' what you are trying to do.
I've started a project with Claude AI, to see if I could revive an old browsergame I created back in the day, and looking at what I've got after just a week... It's still gonna be a big undertaking, but the 'collaboration' with Claude has been very satisfying and pretty damn motivating.
It picks up complex bugs/design problems way before they become a thing, it suggests improvements that actually improve things. And most importantly, it's consistent, even through very lengthy conversations that get sidetracked for a while. You see it improving as well. If it makes a mistake and I correct it, it doesn't make it anymore, even in other conversations.
I still wouldn't trust it to generate all the code itself, though. It models it's syntax to the examples you feed it (like old code) and if you don't ask it whether it's actually good code, it just goes along.
But building the app part by part seems the perfect balance. I frequently ask it to audit the entire solution, which seems to help a lot too. The code it suggests at this point I can pretty much copy-paste into my solution.
And I'm still just using Sonnet 5, haven't even touched Opus or Fable.
The numbers don't decide, the system is a lie. The river running dry, the wings of butterflies.
And you may pour us away like soup. Like we're pretty broken flowers.
We'll take back what is ours. One day at a time.
Signature/Avatar nuking: none (can be changed in your profile)
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
so it can roleplay better ? roflmoa
