It’s pretty simple.
The reason is:Your country is not allowed to access this resource
But if the resource is GPUs, I guess there are ways around that
Lol, “recent”?
Bronze, copper, zinc, silk…
At the earliest signs of human civilization we have; trade has always been tightly controlled. Nothing about this is new, you just don’t know your history.
They mean on YouTube, specifically.
That’s wild someone even tried to watch an embedded YouTube video, but even wilder people apparently could tell it was a YouTube message calling a video a “resource” and not about the actual resource the article is about.
NGL, I too thought they were referring to the concept of trade embargoes…
I am referring to the fact that I can’t view the article because the website has geoblocked me. I was unaware that IP geoblocking has been a thing since the bronze age.
I can’t try if you’re attempting that whole “snark” thing or genuinely need to ask…
But I was saying that “gatekeeping resources” has been around as long as we’ve had evidence of human society.
If you only meant websites not being available in every country…
That’s been around as long as websites have.
Under any interpretation, what you’re talking about is not new.
Merchantilist is back, baby!
lol yeah so GN did this exposé a year ago. Nice that mainstream media is finally picking it up 😑
THE NVIDIA AI GPU BLACK MARKET | Investigating Smuggling, Corruption, & Governments
YouTube
GN doing more investigative journalism than a lot of the actual press.
“Respawnfirst.com” is “mainstream media”?
I love that Steve’s crusade for gamers to have good framerates at reasonable prices led to geopolitical intriuge and potentially domestic corruption charges 🤣
I don't understand these kind of posts, are they real news? or just people posting stuff just because or what? Sorry for my ignorance but I see no sense in this.
Not sure why you are being downvoted.
It is an opinion piece, e.g. a blog.
But it does cite “real news” like press releases and more journalistic articles, so it’s a well informed opinion piece.
It’s not even an opinion piece. There’s no opinion. It’s blogspam. They source another source that sourced a Reuters article.
The purposely picked out a thin thread to use as click bait. This is the line from the original Reuters article “The newspaper’s source alleged”.
I’ll bet real money the downvotes are linked to the OPs spam operation.
Ah, perhaps. I just skimmed it.
Nah it’s not a bot net. It’s just people on Lemmy. Folks on the internet (in aggregate) don’t mind clickbait as long as it’s a sentiment they already agree with, and will reflexively defend it. Sometimes, they even consciously prioritize that over information hygiene.
And I’ve found the Fediverse is no exception, sadly.
Many in the MSM reported this over a year ago.
This is going viral cause it now serves a political purpose in different corners of the oligarchy.
Rubio cause he is a true believer Neocon Cold War shill(Rubio is who leaked the now disputed notion Jensen was being honeypotted by a Chinese spy) and most of the AI companies want to put a protective moat around their AI so they can cartelize the tech and it’s various inputs going into it. using the most powerful government in the world to enforce it. Nvidia was the central power agent in the middle that has an incentive in the other direction to want a more open AI market where they can sell their chips anywhere.
So deploy the China boogieman and then pressure US and US allies to refuse any adoption of China’s open weight models and restrict supply chains(something we are actively doing as we speak) and watch as the broad masses go along with new regulations and restrictions serving the AI companies interests in the name of security/safety and conservatives cheer on taking it to the Chinese and liberals cheer it on in the name of “accountability” for Trump’s corruption and punishing Nvidia for breaking (questionable) US law on sales to China….Or just cause they are distracted with the shiny drama of Jensen being the one oligarch seemingly hooking up with age appropriate women
Don’t believe me? WSJ reports Commerce is about to name a list of favored ai companies (openai, anthropic, microsoft, amazon amongst others) as “national champions.” Qualifying them for special treatment and support by the US government: faster export licenses, government backing abroad(fancy way of saying using the power and threat of US might to advance their interests), federal financing, regulatory protection, and stricter import/export controls.
I think it’s ultimately a sign of weakness(and a bit of face saving for the grifters) and I personally think China has already won this space, but there is very much still a world where the US, in a seemingly bipartisan fashion, is going to build protectionism around this tech to try and prop it up at the expense of everything else.
Did you know that DeepSeek can be massaged into admitting where the true training data came from?
Try it out!
Assuming sarcasm there (my sarcasm meter isn’t great), but this drives me crazy every time it’s brought up seriously in threads.
Nope.
When ran on a local card, on my hardware at home, it will readily admit it was created by American companies.
You can rent a card in the cloud to try it for yourself. It’s fascinating.
The Chinese only know how to copy and scale, not innovate. Innovation comes from individualistic cultures, while scaling solutions come from communal cultures. It actually makes a lot of sense philosophically.
The Americans create it and the Chinese scale it, it’s been the same story for almost a century now.
You do realize that, regardless of what output an LLM gives you, it doesn’t actually know anything right?
The reason you can massage it is because it’s trying to write a plausible story for you, lmao
I think you should experiment with local models!
Train them, embed some custom things, and see what happens!
Compare the results to public services with the same prompts, etc.
As for the semantics of “knowing”, I won’t engage there 😉
Hot beef is exactly right. And before you bring it up, I’ve been experimenting with local models since GPT-J and old GAN finetunes… TBH, I’m not sure how that’s relevant.
You’re not sure how at home training is relevant when we’re discussing the output of LLMs? Or home sepecifc stuff?
Home inference.
And that’s a very broad statement. But nothing we can do locally will get an LLM to intuit or reveal its own dataset.
We can get it to output Claude traces it was trained on. I am not trying to be pedantic, but the semantics/phrasing there is very important.
Not byte for byte, but one can prove an llm was trained on data by asking specific questions. It’s why we’re having all of the copyright lawsuits.
Are all of you seriously trying to argue that local experiments with model training data ARE NOT RELEVANT to our conversation about LLMs and their training data?
Am I losing my fucking mind here or is the hive mind that strong? I’m all ears buddy.
Hot beef is literally trying to argue this point. Are you? It is actually ridiculous.
I am not with the Lemmy hive mind consensus on LLMs, not even close. But we are discussing something more specific.
Yeah. I think people wildly misunderstand how these things work, then want to argue about them. It’s crazy.
All I’m trying to say is that if I train AI on my mythical creature, the “Fijokopil” , a flying tiger ant, then the LLM will 100% be able to define a “Fijokopil”.
And, ChatGPT won’t know what a “Fijokopil” is.
I think I gotta take a break, it’s clear that my implications are going right over peoples heads.
the true training data
Guy who robbed a guy who robbed a bank very upset when he finds out he’s been robbed.
The entire AI industry is scams in scams
NVIDIA is paying customers to buy their shit (and never deploy them)
China’s humanoid robots are bought by local governments to sell training data back to the manufacturers.
While the US leaning on protectionism is a sign off weakness, so is China’s use of “Frontier models” & US chips.
I don’t think any credible analysis has China leading given their dependence on US tech, I also don’t think DeepSeek’s numbers are any more real than openAI so the idea that they are more efficient because they are cheaper is funny.
what is probably true is China is close enough that with/without protectionism that US AI companies will not be making profits big enough to justify their IPOs about time soon.
“Listen carefully, Feyd,” the Baron said. “Observe the plans within plans within plans.”
Listen, I’m not saying I’d eat worm excretions to gain psychic powers in a post “AI” world, I’m just saying I wouldn’t rule it out…
I mean, how’d those plans within plans within plans work out for him in the end?
The real victims here are OpenAI and Anthropic. They would have a monopoly right now if it wasn’t for this evil company. Think of big AI for once.
Living up to their name lol
What is with the recent idiotic trend of geoblocking random countries?