Tired of ads? Enjoy an ad-free experience by signing up.
👩‍🍳 How we use AI at Tech in Asia, thoughtfully and responsibly.
🧔‍♂️ A friendly human may check it before it goes live. More news here

US giants dominate AI spending over Chinese rivals

Analysts said US tech giants are set to spend more than US$700 billion on AI infrastructure this year, far exceeding Chinese peers.

However, rising demand for AI applications in China could push local companies to increase their investment.

Google, Microsoft, Meta, and Amazon account for most of that total, while Morgan Stanley estimated Chinese cloud providers would spend about US$105 billion this year.

UBS put last year’s AI capex by China’s internet giants at 400 billion yuan, or US$59 billion.

China should not be judged against US cloud giants alone, as local firms face curbs on advanced US chips.

They have instead relied more on software and algorithm improvements, said Tilly Zhang of Gavekal.

🔗 Source: South China Morning Post

🧠 Food for thought

Implications, context, and why it matters.

US chip controls slow access without cutting it off

  • US export controls aim to limit China’s access to AI technology without fully shutting it down.
  • Since October 2022, the rules have blocked certain advanced chips plus chipmaking tools from reaching China. That has pushed Chinese firms toward legal imports of weaker alternatives and other workarounds 1, 2.
  • In 2024, Chinese companies legally imported about 1 million Nvidia chips designed for that market. Huawei, the Chinese telecoms and technology company, is estimated to make 200,000 AI chips in 2025 2.
  • A January 2026 rule from the US Department of Commerce left room for sales of advanced AI chips to China. A Council on Foreign Relations (CFR) analysis said a cap of about 1 million Nvidia H200 units could still support the world’s largest AI data center, comparable to xAI’s Colossus in Memphis, Tennessee 3.

China is turning limited compute into an edge

  • Hardware limits have pushed Chinese AI labs to squeeze more from less computing power by improving algorithms.
  • DeepSeek, a Chinese AI startup, trained its highly ranked R1 model on Nvidia H800 chips before those chips faced restrictions in October 2023 4.
  • The pressure has also sped up work on open-source, or open-weight, models whose weights are shared. Chinese companies such as Alibaba are gaining ground worldwide 5.
  • Alibaba’s Qwen models have passed US rivals in total downloads on Hugging Face, an online platform where developers share AI models. Brookings testimony said Qwen also moved ahead of Meta’s Llama in popularity, and that Airbnb’s customer service agent relies heavily on Qwen 5.

Recent Google developments

Stay ahead in Asia’s tech landscape

You've reached your 2 free content limit for the month. Sign up for free to read the full story.

🏄 For casual readers / 👶 Free

Basic

US$0

Free forever

Get instant access to this article and more every month

0 premium content

Unlimited news briefs

5

5 articles

Ad-free reading experience

Just US$0 per day

⌛Sign up in 20s. No payment details needed.

📖 For learners / 👍 Starter

Lite

US$4.92/month

Billed annually at US$59/year

Get instant access to this article and more every month

4

4 premium content

Unlimited news briefs & articles

Ad-free reading experience

Just US$0.17 per day

Cancel anytime

Our subscriber community includes professionals from these companies:

Stay updated on the go with our mobile app.

Get latest insights with smoother, more personalized experience through TIA mobile app.