Most Chinese models are already doing the migration to Huawei chips, will be the biggest lost for Nvidia.
Artificial Intelligence thread
- Thread starter 9dashline
- Start date
first, maybe you should show so loyalty to China and stop using Claude. What a joke. Everyday you come on here and say you support Chinese AI effort and then you continue to give money to Western AI companies.I have submitted my resume
I actually put my money where my mouth is and have paid for coding plans with both Kimi and GLM. I'm frankly tired of people like yourself.
It will be a huge loss for Nvidia for sure, but it will be a victory, at least in the short term, to the American AI companies as Huawei chips are inferior to Nvidia chips that American AI companies have enjoyed.
Do you think AI datacenters is just one massive chip? lol
The only thing that matters is $ per token and US is basically done.
See the thing about Nvidia is their chips aren't actually that good once you factor in software-hardware optimization, which they can't do because it's Chinese companies publishing all frontier techniques and, contrary to popular belief, Chinese labs don't actually publish their latest advances on day 1.
Engineer
Major
The problem with this thinking is that it assumes chips work independently of one another. That's not how they work in reality.
In a homelab, Nvidia will offer the best performance for sure, but a chip's raw performance become quickly meaningless in a cluster where the biggest bottleneck is inter-chip communication. Remember, Huawei's bread-and-butter business is telecommunication. Huawei doesn't need beat Nvidia in chip's raw performance. Huawei just needs Ascend chips to be good enough, then the company's strength in telecommunication is enough to steamroll everybody.
data flywheel that Anthropic built and how that flows to other labs. Note, the key is advanced SWE tasks they get and user responses.
Again, note that our usage of GLM and Kimi coding plans help with their future models.
Well to be fair, lots of mainland engineers also use Claude. It's quite popular. I wouldn't regard it as some kind of indictment of disloyalty.
If they spent their own money to use it, I will call it disloyalty, cause they have better options out there. They can use Chinese models or even OpenAI. Using Claude when Anthropic is openly the most anti-China company is surely being a tone deaf traitor when China is being suppressed using US sanctions. Its aiding the enemy by giving them money in the new cold war.
If their company uses it, then no other choice.
Its about cost benefit ratio. If im paying $200 a month for subscription but costing them way more to serve because im maxing out the qouta, then its a win win for me. In the long arc of things China is going to win the AI race anyway, its structural, and there is nothing Dario or the USA can do about it... what I do or dont know doesnt make a dent of difference in the grand scheme of history. In the meantimes its the VC suckers and 401k bagholders thats subsidizing my AI use
Okay, I make a counter point to that, although you could be right.
1. Software sometimes can be localized (for whatever reasons). People use email in the West. They use WeChat in China. So I think American AI will never disappear. What form it has as a commercial entity has not been finalized.
2. What clearly has changed, is that the Americans have to compete in the open source AI space to maintain relevance. The US government will always buy American, but that does not go for corporate America.
3. American AI will find a way to stick around. By that I mean the current AI leaders in America, will change tack, and all become hyperscalers. That is only logical, and is what Jensen Huang said a few months ago. AI is booming, and we need more compute. But, data centers are a big capital expense. So who, which companies, has that kind of money for capital investment?
4. Once DeepSeek came out, the entire game changed. All this talk about the frontier and AGI, that is the smokescreen. They are lying to the public, and more important they are lying to themselves to protect their fragile egos, like a gimp.