China’s open-weights AI strategy is winning • Hacker News Board

Re: China’s open-weights AI strategy is winning

#951

Post by maxerickson » Tue, Jul 21, 2026, 5:58 PM UTC

elictronic wrote: post #940

maxerickson wrote: post #488

Earlier quoted context omitted.

Talking about their product segments the same way they do is not being pedantic. Insisting on your personal definition is!

You sound like someone who believes online betting markets aren’t gambling because the company says so. Cryptocurrency is commonly used for personal transactions. You might not ever admit your argument is dumb but tomatoes are fruit is stupid.

I'm perfectly happy to acknowledge that tomatoes are a vegetable in the kitchen.

If you want to understand their reproductive biology, you are stuck understanding them as fruit.

It's maybe telling that you bring up a confusion that is harmful instead of explaining why grouping massively parallel processors under the term "gpu" is harmful.

Re: China’s open-weights AI strategy is winning

#952

Post by SJC_Hacker » Tue, Jul 21, 2026, 6:14 PM UTC

engineer_22 wrote: post #884

consumer451 wrote: post #787

Earlier quoted context omitted.

> I use AI chat every day, I find it endlessly useful. It’s replaced google search. Extremely subsidized agentic search is very superior to Google at the moment, and of course it is. Google is a public company. The AI summary model has to work instantly, is likely as dumb as a 8T param model, and gives you incorrect details constantly. This sucks so much for Google. If you click on "AI Mode," suddenly the facts becom…

Your comment about watts stuck out to me. I became curious how much power we might be talking about. I’m into Claude for $20/mo (petty bourgeois tier), let’s assume for sake of examination this is only paying for power, infrastructure already amortized Price of grid power approx $0.15/kWh $20 / ($0.15/kWh * 30 days) = 4.4 kWh per day. This same amount of energy can lift a 2-ton SUV 1/2 mile into the air. To me this s…

With a crane and a good pulley system, a human should be able to pull that off inside of hour. Not all that impressive to me.

Re: China’s open-weights AI strategy is winning

#953

Post by runeks » Tue, Jul 21, 2026, 7:09 PM UTC

giancarlostoro wrote: post #927

I still wonder why some of our key universities in the US aren't building Open Weight Models. To this day we have "BSD" and "MIT" as key licenses for a lot of software (I know weird tangent but still... this emphasizes how significant these institutions were once!), but these top schools don't really build anything the way they used to anymore. I'm sure someone will say students build LLMs all the time, and this and…

> I still wonder why some of our key universities in the US aren't building Open Weight Models.

I'm guessing the $100m price tag plays a role. Plus the fact that it will be outdated next quarter.

Re: China’s open-weights AI strategy is winning

#954

Post by KerrAvon » Tue, Jul 21, 2026, 7:53 PM UTC

stanzeo wrote: post #919

DrScientist wrote: post #878

Earlier quoted context omitted.

Who funds the majority of cutting edge scientific research? Is it companies or is it governments? If governments around the world see LLMs built from public knowledge as pre-competitive as the public knowledge itself, then why wouldn't they sustainbly fund it?

Because the business model for governments is difficult. Income are mostly taxes, so you have to be able to explain constantly why it is a good idea to keep on spending budget on that funding.

Governments don't have a business model. They don't need to make a profit. (Also, debt doesn't work the same way for the federal government that it does for a business or even a US state.)

see, for just one take on this, https://www.ucdavis.edu/magazine/why-federally-funded-resear...

Re: China’s open-weights AI strategy is winning

#955

Post by charintstr » Tue, Jul 21, 2026, 8:29 PM UTC

Everyone was saying the exact same thing, that open would win and compute was overvalued, about deepseek back in 2025 ignoring that up until now open models had essentially returned to about 6-9mo behind OAI/Ant post-Deepseek moment. I think this is a dynamic frontier and open modeling a strategy that has to be pursued by China to pool their knowledge in order to keep up with the US. However, if China actually breaks ahead, then open becomes a burden rather than a strength. So I don't see this as a China better thing, just asking more "who is ahead" or not.

Re: China’s open-weights AI strategy is winning

#956

Post by zrn900 » Tue, Jul 21, 2026, 8:47 PM UTC

For China, AI is infrastructure. For the US, it's another cash cow to be made scarce so that investors can milk it.

That's why China chases open-source AI and commoditizes it. It will improve the AI until it is used in all kinds of things, ranging from vehicles to factories, from medicine to housemaid robots. China will make the AI the commoditized infrastructure to build bigger things on it.

You can't do that while keeping AI as a scarce resource and charging a premium for it. Like how the US does to satisfy endless investor greed.

Re: China’s open-weights AI strategy is winning

#957

Post by Danox » Tue, Jul 21, 2026, 11:22 PM UTC

danny_codes wrote: post #852

blackqueeriroh wrote: post #790

Earlier quoted context omitted.

Because the scale of the subsidization is unimaginably different.

The scale of public spend is probably much higher in the US..

From a distance, the Chinese spending seems to be more focused and spread out across their infrastructure at all levels. Is that a difference between engineers running the system and lawyers running the USA?

I certainly would like to see Thorium get some love after basically 50 years of being mothballed moth, and the same goes for the rail system within the United States.

Re: China’s open-weights AI strategy is winning

#958

Post by Danox » Tue, Jul 21, 2026, 11:25 PM UTC

twobitshifter wrote: post #782

idontwantthis wrote: post #502

Earlier quoted context omitted.

I am also confused by this point. The American government could force OpenAI and Anthropic to open their models, but then they would instantly evaporate, right? It doesn't seem like a choice that they can make, so framing it as a "winning" strategy doesn't make any sense to me. In what world could those companies have existed and opened their models?

Why did Google give Android away for free?

To service their ad Empire and the same thing applies to creating the Chrome browser and Google maps…

Re: China’s open-weights AI strategy is winning

#959

Post by dinfinity » Wed, Jul 22, 2026, 12:01 AM UTC

zeroonetwothree wrote: post #641

dinfinity wrote: post #459

Earlier quoted context omitted.

12 months ago "way too many stupid errors" was constant news. Today, you rarely hear about those anymore. Sure, the novelty of the errors has worn off a bit and thus the reporting. Nevertheless the quality has improved immensely in this regard. Also, AI video generation is now so good and accessible that it is very, very regularly used for memes, disinformation and proper (short) movie projects. AI image generation e…

Ah yes, memes are clearly the sign of massive progress.

Yes, if they are made with new technology and the quality is good enough they certainly are.

AI generated video memes were NOT a thing a year ago. Yes, the AI video generation in itself was a meme (Will Smith eating spaghetti), but now there are tons of convincingly good AI video memes about things like sports events generated by average Joes. It is proof of the capability and accessibility of the technology even if the use of it is mundane.

Everybody and their dog playing snake on their Nokia 3310 was similarly mundane, but also a sign of the end of the era of Gameboys and the beginning of (normie) mobile gaming.

Re: China’s open-weights AI strategy is winning

#960

Post by bilbo0s » Wed, Jul 22, 2026, 12:15 AM UTC

nonethewiser wrote: post #94

_aavaa_ wrote: post #3

> I have serious concerns about how these models might reflect Chinese government perspectives (try asking them about Tiananmen Square). And I have serious concerns about the American ones. Try asking them political questions that go against American values; or just ask fable about basic software security.

>Try asking them political questions that go against American values 1. Can you give me some examples? 2. Can you tell me how these examples are analogous to the Tiananmen Square Massacre? Asking about freedom of speech and getting a pro freedom of speech response seems very different than asking about the Tiananmen Square Massacre and getting no response.

Just Devil’s Advocate, but couldn’t you ask these models about controversial topics in US or Chinese history. Like, what happened at Tiananmen? Or what happened during the Tulsa riots?