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

OpenAI says it has evidence DeepSeek used its model to train competitor

ft.com

Post reply on HN
Print view View original

511–520 of 1001 posts

Re: OpenAI says it has evidence DeepSeek used its model to train competitor

#511

Post by echelon »

hnlmorg wrote: post #474
827a wrote: post #465

Earlier quoted context omitted.

There is a third possibility I haven't seen discussed yet: That DeepSeek, illegally, got their hands on an OpenAI model via a breach of OpenAI's systems. Its easy to laugh at OpenAI and say "you reap what you sow", I'm 100% in that camp, but given the lengths other Chinese entities have gone to when it comes to replicating Western technology; we should not discount this. That being said, breaching OAI's systems, re-t…

[flagged]

> based purely on racial prejudices.

At some point these straw men start to look like ignorance or even reverse racism. As if (presumably non-Han Chinese) Americans are incapable of tolerance.

There are plenty of Han Chinese who are citizens of democratic nations. China is not the only nation with Han Chinese.

America, for instance, has a large number of Asian citizens, including a large number of Han Chinese. The number of white, non-Hispanic Americans is decreasing, while the number of Asian Americans is increasing at a rate 3x the decrease in whites. America is a melting pot and deals with race relations issues far more than ethnically uniform populations. The conversations we have about race are because we're so exposed to it -- so racially and culturally diverse. If anything, we're equipped to have these conversations gracefully because they're a part of our everyday lived experience.

At the end of the day, this is 100% a geopolitical argument. Pulling out the race card any time China is criticized is arguing in bad faith. You don't see the same criticisms lobbied against South Korea, Vietnam, Taiwan, or Singapore precisely because this is a geopolitical issue.

As further evidence you can recall the conversations we had in the 90's when we were afraid Japan would take over. All the newspapers wrote about was "Japan, Japan, Japan" and the American businesses they were buying up and taking over. It was 100% geopolitical fear. You'll note that we no longer fill the zeitgeist with these discussions today save for a recent and rather limited conversation about US Steel. And that was just a whimper.

These conversations about China are going to increase as the US continues to decouple from Chinese trade. It's not racism, it's just competition.

Re: OpenAI says it has evidence DeepSeek used its model to train competitor

#513

Post by nataliste »

A Wolf had stolen a Lamb and was carrying it off to his lair to eat it. But his plans were very much changed when he met a Lion, who, without making any excuses, took the Lamb away from him.

The Wolf made off to a safe distance, and then said in a much injured tone:

"You have no right to take my property like that!"

The Lion looked back, but as the Wolf was too far away to be taught a lesson without too much inconvenience, he said:

"Your property? Did you buy it, or did the Shepherd make you a gift of it? Pray tell me, how did you get it?"

What is evil won is evil lost.

Re: OpenAI says it has evidence DeepSeek used its model to train competitor

#514

Post by insane_dreamer »

insane_dreamer wrote: post #174

Usually I'm very much on the side of protecting America's interests from China, but in this case I'm so disgusted with OpenAI and the rest of BigTech driving this "arms race" that I'd be happy with them burning to the ground. So we're going to reverse our goals to reduce emissions and fossil fuels in order to hopefully save future generations from the worst effects of climate change, in the name of being able to do w…

downvoted -- I guess I upset some people defending OpenAI? Good.

Re: OpenAI says it has evidence DeepSeek used its model to train competitor

#515

Post by aiono »

aprilthird2021 wrote: post #478
rockemsockem wrote: post #467

Earlier quoted context omitted.

I think the prevailing narrative ATM is that DeepSeek's own innovation was done in isolation and they surpassed OpenAI. Even though in the paper they give a lot of credit to Llama for their techniques. The idea that they used o1's outputs for their distillation further shows that models like o1 are necessary. All of this should have been clear anyway from the start, but that's the Internet for you.

> the prevailing narrative ATM is that DeepSeek's own innovation was done in isolation and they surpassed OpenAI I did not think this, nor did I think this was what others assumed. The narrative, I thought, was that there is little point in paying OpenAI for LLM usage when a much cheaper, similar / better version can be made and used for a fraction of the cost (whether it's on the back of existing LLM research doesn'…

That's only the case if you don't need to use the output of a much more expensive model.

Re: OpenAI says it has evidence DeepSeek used its model to train competitor

#516

Post by JTyQZSnP3cQGa8B »

glenstein wrote: post #220

All the top level comments are basking in the irony of it, which is fair enough. But I think this changes the Deepseek narrative a bit. If they just benefited from repurposing OpenAI data, that's different than having achieved an engineering breakthrough, which may suggest OpenAI's results were hard earned after all.

> OpenAI's results were hard earned after all

DDOSing web sites and grabbing content without anyone's consent is not hard earned at all. They did spent billions on their thing, but nothing was earned as they could never do that legally.

Re: OpenAI says it has evidence DeepSeek used its model to train competitor

#517

Post by tehjoker »

827a wrote: post #465
Imnimo wrote: post #320

I think there's two different things going on here: "DeepSeek trained on our outputs and that's not fair because those outputs are ours, and you shouldn't take other peoples' data!" This is obviously extremely silly, because that's exactly how OpenAI got all of its training data in the first place - by scraping other peoples' data off the internet. "DeepSeek trained on our outputs, and so their claims of replicating…

There is a third possibility I haven't seen discussed yet: That DeepSeek, illegally, got their hands on an OpenAI model via a breach of OpenAI's systems. Its easy to laugh at OpenAI and say "you reap what you sow", I'm 100% in that camp, but given the lengths other Chinese entities have gone to when it comes to replicating Western technology; we should not discount this. That being said, breaching OAI's systems, re-t…

Basically, without some kind of shred of evidence, this is completely chauvinist to make this accusation.

Re: OpenAI says it has evidence DeepSeek used its model to train competitor

#518

Post by YetAnotherNick »

827a wrote: post #465
Imnimo wrote: post #320

I think there's two different things going on here: "DeepSeek trained on our outputs and that's not fair because those outputs are ours, and you shouldn't take other peoples' data!" This is obviously extremely silly, because that's exactly how OpenAI got all of its training data in the first place - by scraping other peoples' data off the internet. "DeepSeek trained on our outputs, and so their claims of replicating…

There is a third possibility I haven't seen discussed yet: That DeepSeek, illegally, got their hands on an OpenAI model via a breach of OpenAI's systems. Its easy to laugh at OpenAI and say "you reap what you sow", I'm 100% in that camp, but given the lengths other Chinese entities have gone to when it comes to replicating Western technology; we should not discount this. That being said, breaching OAI's systems, re-t…

Deepseek v2 and v2.5 was still very good but not par with frontier models. How would you explain that?

Re: OpenAI says it has evidence DeepSeek used its model to train competitor

#520

Post by joe_the_user »

rockemsockem wrote: post #467
riantogo wrote: post #461

Earlier quoted context omitted.

Why would it cast any doubt? If you can use o1 output to build a better R1. Then use R1 output to build a better X1... then a better X2.. XN, that just shows a method to create better systems for a fraction of the cost from where we stand. If it was that obvious OpenAI should have themselves done. But the disruptors did it. It hindsight it might sound obvious, but that is true for all innovations. It is all good stuf…

I think the prevailing narrative ATM is that DeepSeek's own innovation was done in isolation and they surpassed OpenAI. Even though in the paper they give a lot of credit to Llama for their techniques. The idea that they used o1's outputs for their distillation further shows that models like o1 are necessary. All of this should have been clear anyway from the start, but that's the Internet for you.

The idea that they used o1's outputs for their distillation further shows that models like o1 are necessary.

Hmm, I think the narrative of the rise of LLMs is that once the output of humans has been distilled by the model, the human isn't necessary.

As far as I know, DeepSeek adds only a little to the transformers model while o1/o3 added a special "reasoning component" - if DeepSeek is as good as o1/o3, even taking data from it, then it seems the reasoning component isn't needed.

Post reply on HN