I wonder at what point people start taking them to court. It seems like the usual idiot tech bro excuse of thinking terms of service/use somehow override the law which is hilariously naive.
You cannot override the law in a TOS.
Like if they wrote down that they were allowed to murder you written into their TOS and proceeded to murder you they'd still go to jail for murder.
If Ukraine gets Gripen then they get access to an air platform that can fire the best air to air missile in the world.
Meteor is deadly.
They could use F-16's in the SEAD/DEAD/bomb truck role whilst using the Gripens to shoot down russian jets and helicopters from extreme range.
How so?
The trained model includes vast swathes of copyrighted material. It's the rights holders who get to decide whether someone can use it.
Just because it makes it inconvenient or harder for someone to train an AI model does not justify wholesale stealing.
A lot of models are even trained on large numbers of pirated material like books downloaded from pirate sites etc. I guarantee you OpenAI and others didn't even buy a lot of the material they use to train the AI models on.
Reddit Migration@kbin.social – 4 points – It really does make me wonder at what point the apollo dev does take Huffman to court for defamation.
Their nation is spending something like half its budget if not more on defence, they can manage a few dozen jets.
Besides, they'll have to phase out their soviet era jets at some point because of the sheer age of them soon. Those Su-24 etc airframes are probably in rough shape by this point.
No it doesn’t, the training data isn’t inside the LLM.
This is factually incorrect. You can extract the data. How do you think the legal cases are being brought?
The model has to contain the data in order to produce works.
Wholesale commercial copyright infringement where you're profiting off of others work on a large scale is a whole different ball game.
They're training their models on large amounts of pirated content and profiting off it.
Of course the rights holders are going to say "wait a minute, why are you making money off my content without my permission? And how much of my work did you pirate to use?"
You cannot hand wave away mass piracy to train their models, and then distribute said models based on an act of mass copyright infringement.
Do you not understand the basics of the law?
its idiotic to think that its reasonable to demand such a thing.
Again, the law is the law. If they mass pirate a bunch of media which then the model contains chunks of they are breaking the law.
I can't believe this is a hard concept for someone to understand.
Honestly, I didn't mind them. If they had them on kbin I wouldn't mind as long as that money went to server upkeep etc. Somebody has to pay for hosting.
His name is Steve Huffman, using Spez means he gets to avoid a lot of the public criticism through google searches etc.
Control is excellent and also has a mod from one of the developers that adds stuff like HDR, support for newer versions of DLSS, better raytracing etc.
EDIT: Also, if you're playing RDR2 ensure to enable the nvidia reflex low latency stuff and run at a high framerate to mitigate some of the input lag. It's a marked improvement over PS4.
That baby shits lead.
Man.. I went from slashdot, to digg, to reddit..
I still have my slashdot account though, slashdot is a shadow of what it used to be. Pretty sure my digg account is long dead. We'll see with reddit I guess.
I've been a lurker on reddit for over 10 years. Here I feel like I can actually post and comment without worrying about negativity :)
Here's a bit of negativity; give it time. The larger a community and platform the wider the views. What starts as a tight-knit community eventually expands into something gigantic where things aren't so tight-knit any more.
It's like gravity, the larger something is the more it pulls in.
Place your bets whether these aren't ports but just emulators with the PS1/PS2 ISO's.
Cyberpunk 2077 was a hot mess at launch. The one good thing I think is the Judy storyline was pretty well done.
This entire comment screams of 0 technical knowledge
Yes, your comment does.
There is literally software to extract this stuff from models now.
This "it's just math" is techbro idiocy. It's like the idiots regurgitating crypto coin bullshit.
(Also thanks from a K400 owner, haha.)
The ability to finally disable those super annoying gestures permanently was a life saver. It was driving me up the wall.
Allow NSFW content as well, that way reddit gets no money from you regardless as you have to flag it as a NSFW subreddit which means no ads.
Love it.
I used to work on secure sites for defence work. Honestly, not going to go back for that stuff any more. No more travelling and all the headaches that comes with.
Fully work from home now. Any job that expects me to work in an office is not a job I will even bother to interview for.
And as said they didn’t “train chat GPT on a piracy site” the scraping algorithm put some stuff form there in the training data. There is no person doing that.
"Your honour my program that I created to slurp up data from the internet using my paid for internet connection, into my AI trained model that I own and control happened to slurp up copyrighted data.. I um, it's not my fault it slurped up copyrighted data even though I put no checks in place for it to check what it was slurping up or from where."
That is the argument you are putting forth.
Do you think any judge/court of law would view that favourably?
User on kbin here, just tried to sign up to lemmy.world.. looks like everything crashed and burned when tried to sign up there.
Interesting thing; Bloodborne is now booting in the PS4 emulator fpps4 to the menu music.
Like I said, don't worry about it. If you don't get it you don't get it.
From what I heard during development a load of the developers left because they joined Arkane to make games like Prey, Dishonored etc. Not multiplayer games.
Reddit demands not to face any consequences for its actions. This is the news at ten.
At a guess if something like here starts becoming popular with people I suspect they'll comment here for example. Kbin is growing, see what happens I guess.
That and an API that offloads data loading direct to the GPU. This is using DirectStorage 1.2, so it is bypassing the cpu in loading the data and decompressing it.
The PS5 has a total of 16GB of RAM which is split depending on the developers.
My PC has 32GB RAM and 16GB VRAM. Meaning I have three times the amount of ram available. The PC port of this likely has the option with a HDD of preloading data into ram before it is used (like loading the data into ram for the next level) where the PS5 did not. There are ways to do these things to speed up loading of data, but when you have limited amounts of RAM available means it limits your options.
For a console they want to maximise the ram usage for what is in current use. So yes, nvme SSD is likely required for that.
This is one of the features of DirectStorage Memory-to-memory decompression.
A corporation only ever adopts open standards when it is advantageous to them, the moment it is the counter they close it down or stop cooperating.
This has been repeated with Microsoft, Google, Oracle (Sun too), SCO for those of us old enough etc.
Embrace, extend and extinguish has been done over and over again from them all. Meta/Facebook is no different.
Good thing I just recently upgraded my internet connection to a full gigabit.
🍿
Eh, I mean they probably did it because of russian government pressure but ultimately I've seen a lot of russian opposition media hold what are questionable views regarding Ukraine, often a colonial or imperialists view.
They don't want war, they just want to do these things peacefully to impose russia on Ukraine etc.
At this point I couldn't give two shits what happens in russia, what's important is helping Ukraine.
Something to consider before anyone goes preordering, From's last pc game Elden Ring has a nasty shader compilation stutter that they never fixed to my knowledge.
Don't forget to hit that like button. And hit that bell and subscribe for all the latest live court reactions.
Yep, I think they need to get SteamOS on these if they want to compete. Windows is simply not good for devices like this, at all.
Works fine here, this might be a kbin/lemmy type formatting thing going on.
It contains large parts of the data in order to create. In my link I provided it shows that the models do contain chunks of the original works.
Otherwise, how would it create the words etc.
I am amazed that we now have people on the level of crypto coin idiocy going on about ai models who don't understand this.
You're contradicting yourself.
In one sentence you say it doesn't memorize (with "reasonable effort") then in the next you admit it does.
"Reasonable effort" is weasel wording.
Make up your mind.
So, me pointing out the flaw in your argument is trolling?
What?
If you choose to use weasel wording to try and get out of something that is your call.
So, now it's gone from "reasonable effort" to most definitely you can say without any doubt that all the trained models contain no copyrighted data at all?
Come on. Make up your mind.
I think it's those stupid hard coded buttons on my remote that I accidentally press every so often then have to repeatedly try and back/exit out of the stupid thing it launched that I cannot remove/uninstall from my tv.