China claims their AI is better than the US again… does it even matter?

Chinese tech company Alibaba just dropped a new version of its Qwen 2.5 AI model, claiming it’s better than DeepSeek-V3.

Is this actually a big deal, or just another PR move? Are we looking at another market reaction?

https://finance.yahoo.com/news/alibaba-releases-ai-model-claims-051704166.html

Can’t wait for the Temu AI

Tyler said:
Can’t wait for the Temu AI

They’ll train their model for $50K while others spend $5M.

Thames said:

Tyler said:
Can’t wait for the Temu AI

They’ll train their model for $50K while others spend $5M.

And somehow, they’ll find a way to make users pay for the training and call it a deal.

Thames said:

Tyler said:
Can’t wait for the Temu AI

They’ll train their model for $50K while others spend $5M.

  • shipping and you wait three weeks for an answer.

Tyler said:
Can’t wait for the Temu AI

Jokes aside, their engineers are actually solid. People dismiss them because they sell cheap junk, but at the scale they operate, they’ve got serious AI talent working behind the scenes.

Tyler said:
Can’t wait for the Temu AI

Imagine a Temu LLM…

The US doesn’t ‘own’ any AI models, private companies do. Unless you’re saying the government owns everything these companies produce… which sounds kind of like China’s system.

Paz said:
The US doesn’t ‘own’ any AI models, private companies do. Unless you’re saying the government owns everything these companies produce… which sounds kind of like China’s system.

From a European perspective, it honestly looks more like big tech owns the US government than the other way around.

@Micah
That’s not just a look, that’s reality.

Paz said:
The US doesn’t ‘own’ any AI models, private companies do. Unless you’re saying the government owns everything these companies produce… which sounds kind of like China’s system.

In a way, the US does control these models, because companies can’t sell them to whoever they want. AI, like GPUs, is treated as highly sensitive tech. So calling it ‘US AI’ isn’t completely wrong.

Competition is good for the global economy, which means in the long run, it’s good for the stock market.

I have no opinion on the models themselves, and I have zero clue how the market will react in the short term. But honestly? I don’t really care.

They say it beats ChatGPT-4o and DeepSeek-V3, but no mention of OpenAI’s other models or DeepSeek-R1, which are top-tier right now.

They’re catching up fast, but they’re still trailing a few months behind the best models out there.

And no, the market won’t freak out this time. The shock factor is gone.

@Kingsley
China AI better than US!

Sai said:
@Kingsley
China AI better than US!

Winnie the Pooh

Ask DeepSeek why Xi is called Winnie the Pooh—it won’t answer. Ask it about China’s biggest controversies—it’ll give some vague response.

So yeah, their models are powerful, but they come with CCP-approved ‘truth.’ Western AI models at least try to be objective about all countries, not just the ones they don’t like.

@Yun
‘Western models are objective’—except they won’t tell you how to make explosives or show anything remotely NSFW.

Censorship exists everywhere, it’s just applied differently.

@Yun
Or, hear me out… use the AI for actual useful stuff instead of asking it political questions?

@Yun
Petty takes are the best takes.

I don’t see the hype. Anything built in China is ultimately controlled by the CCP, meaning the US and allies won’t touch it for anything serious.

It’s also not investable—no one is dumping NVDA stock to buy DeepSeek. The market overreacts to every little thing, but this doesn’t actually change anything.