Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Startup founders urge U.S. government not to shut off Chinese open weight AI

politico.com

Post reply on HN
Print view View original

601–610 of 983 posts

Re: Startup founders urge U.S. government not to shut off Chinese open weight AI

#601

Post by cmiles74 »

Oh man, if you weren't sure if the administration is staffed with people who do not understand the technology they are trying to manage, doubt no more!

From the article...

Treasury Secretary Scott Bessent said Tuesday on Fox Business’ “Mornings with Maria” that the administration would investigate whether Chinese artificial intelligence companies had improperly distilled American models to power their own, saying that the U.S. can “sanction” companies that engage in intellectual property theft.

Re: Startup founders urge U.S. government not to shut off Chinese open weight AI

#602

Post by chucksmash »

scoofy wrote: post #555
mythrwy wrote: post #547

Earlier quoted context omitted.

[flagged]

[flagged]

The world is full of nuance. Just like whataboutism and "both sides" can be convenient rhetorical devices for people lying to themselves or others, accusing someone of either can be a convenient figleaf for avoiding thinking about the flaws of one's own side.

The fact that Trump won in 2016 obviously overshadowed many things about that election. I didn't vote for him but I can still remember acutely the disappointment I felt during the primaries when the presumptive frontrunners were another Clinton and another Bush. Like out of 300 million people, these will be the choices presented to me? Whataboutism is bad mkay, but two things can also be true at the same time. Nuance.

Re: Startup founders urge U.S. government not to shut off Chinese open weight AI

#603

Post by scoofy »

pessimizer wrote: post #592
scoofy wrote: post #555

Earlier quoted context omitted.

[flagged]

> "Both sides are the same" is the last bastion of denial. This is hyperpartisan bullshit that you obviously wouldn't say if someone were choosing between a red shirt and a blue shirt. If you support some political candidate, defend them on the merits of their positions and capabilities, not by repeating content-free jingles. Or even worse, based on unexplored counterfactuals: "X" did this, don't you wish you picked…

>This is hyperpartisan bullshit that you obviously wouldn't say if someone were choosing between a red shirt and a blue shirt.

I voted for Dean Phillips in the primary. I'm extremely angry at the DNC. I thought they treated Mitt Romney and John McCain like garbage for absolutely no reason. I would have happily voted for Romney over both Clinton and Trump in 2016. I'm not some partisan hack. That said, it wasn't a difficult choice last election.

Donald Trump is an exceptionally corrupt, delusional, moronic President. He pretends that doing it out in the open makes it okay.

He's literally flying around in a Qatari bribe. This level of corruption and self-dealing is exceptional.

Re: Startup founders urge U.S. government not to shut off Chinese open weight AI

#604

Post by nolok »

mythrwy wrote: post #559
scoofy wrote: post #555

Earlier quoted context omitted.

[flagged]

I didn't say "Both sides are the same". I said both candidates were morons. As far as terms, I don't like really like either side we have on the menu right now.

[flagged]

Re: Startup founders urge U.S. government not to shut off Chinese open weight AI

#605

Post by throw1234567891 »

vkaku wrote: post #398
throw1234567891 wrote: post #389

Earlier quoted context omitted.

Maybe they would be more successful if they were posting on an open platform instead of a walled garden. This “free speech” platform is pretty useless at being “open”.

Hey, I'm on this one too. Nothing stays forever one way. You all find me on the other ones, add me there and give me a reason to post to a wider audience. I'm on LinkedIn too ... most people won't even engage the same way there and I just want a simple easy way to engage.

I read the thread. It sounds dangerous sending any donations to a Russian national based in Moscow in the current political climate. Maybe that’s why nobody on LinkedIn wants to chime in.

Re: Startup founders urge U.S. government not to shut off Chinese open weight AI

#607

Post by mythrwy »

nolok wrote: post #604
mythrwy wrote: post #559

Earlier quoted context omitted.

I didn't say "Both sides are the same". I said both candidates were morons. As far as terms, I don't like really like either side we have on the menu right now.

[flagged]

That doesn't describe the candidates on offer though.

Look, I just want good rational, wise, humane low touch governance without excessive political ideology nonsense. I'm sick of the wars, sick of the lies, sick of the economic yo yo of boom/bust, sick of progressivism and "woke", sick of religion being injected into the conversation, sick of idiot candidates put there by the wealthy. And no, I'm not going to pretend that Joe Biden administration or Kamala was acceptable. Nor am I going to pretend that socialism is really gonna work this time, it's not and will leave most people worse off over the long term.

I realize this might be too much to ask, for but a boy can dream.

Re: Startup founders urge U.S. government not to shut off Chinese open weight AI

#608

Post by sillysaurusx »

kube-system wrote: post #579
sillysaurusx wrote: post #343

Earlier quoted context omitted.

> Proprietary model weights are IP Not copyrightable IP, or at least it hasn't been challenged yet. I have experience with this: I made llama-dl, a way to download the original llama model. Meta issued a DMCA, I appealed to the HN community for funds, someone funded, and our lawyer successfully counterclaimed. Never heard from Meta again. A lack of response to the counterclaim doesn't mean the issue is settled. But n…

Congrats on your efforts, but by definition that does not count as legal precedent. At a bare minimum , a legal precedent is some sort of written holding by a court with authority. Even an out-of-court settlement doesn't count.

It was referenced by other legal cases. Why would they mention it if it didn’t help their case?

Re: Startup founders urge U.S. government not to shut off Chinese open weight AI

#609

Post by adrian_b »

paulddraper wrote: post #284
petilon wrote: post #272

Earlier quoted context omitted.

> The U.S. does NOT have copyright protections on the data that AI models were and are trained on, it was mass theft. Absolutely false. Courts have decided that it is fair use [1]. That makes sense. It should not be used to justify Chinese theft. https://www.reuters.com/world/us-judge-approves-anthropics-1...

Correct. This seems to be poorly understood. You can use an original work to write an encyclopedia (e.g. Wikipedia). That is fair use. You cannot copy an original work and distribute it. LLMs are an encyclopedia.

But no Chinese company has copied and distributed an American LLM.

Even if they had wanted that, it would have been impossible, because Anthropic et al. do not let anyone access them directly.

Querying the available API can be used to extract only an extremely small part of the information stored in an LLM.

That part can be used for the post-training of another LLM, to obtain some desirable properties, but the extracted information is far too little to be called "copying" or "distributing". Moreover, after the post-training it does not appear anywhere in the weights verbatim, so its use is at least as transformative as the training of the original American model.

Besides these facts, there is no evidence that the Chinese companies have actually done this, even if it sounds plausible.

Re: Startup founders urge U.S. government not to shut off Chinese open weight AI

#610

Post by IncreasePosts »

adrian_b wrote: post #576
IncreasePosts wrote: post #525

Earlier quoted context omitted.

What researchers want to join a company that is in a race to the bottom to serve inference tokens? If any open weight companies start actually producing state of the art models then maybe, but if it's just copying what others have done for cheaper, you aren't going to find many researchers interested in that

The Chinese AI models are certainly not copied from any US sources. All of them have quite different structures, and the reasons for choosing those structures have been clearly explained in published research papers. The structures of the US "SOTA" models are unknown and nothing useful has been published about them, so they certainly were not a source of inspiration for China. Big LLMs like those published by the Chi…

What would the benchmarks be if GLM5.2 and Kimi K3 be if they didn't use distillation of frontier models?

I don't think any of the Chinese AI companies have stolen IP from the US AI companies (at least, I've seen no evidence of it), but the evidence of distillation is pretty apparent.

My issue with this is that if distillation occurs frequently enough, it is going to zero out most SOTA research into AI. Having cheap AI is great, but that alone won't advance the state of the art. There needs to be groups that are pushing the boundaries, and unless some kind of protection is put into place, there will be zero financial incentive to do so if anyone can come along and effectively steal your model and get financially rewarded for serving it far cheaper than the original group can, because much less R&D cost is needed. I say this as someone who sees distillation as "legal", since if you can train anything you can look at, and you can look at the output of those frontier models, then you can train on them.

Post reply on HN