Western AI labs are pulling back from releasing large models openly. Meanwhile, Chinese developers are stepping in to fill the gap. They are building powerful AI models designed to run on everyday hardware, making them more accessible and easier to use. A recent security study highlights just how much Chinese AI has taken over this space.

Chinese Models Dominate Open-Source AI Deployment

A study conducted by SentinelOne and Censys tracked over 175,000 exposed AI hosts across 130 countries over nearly a year. It found that Alibaba’s Qwen2 consistently ranks second in global deployment, right behind Meta’s Llama. This shows how popular and widespread Chinese models have become.

Interestingly, Qwen2 is often deployed alongside other models. When systems run multiple AI models at once, Qwen2 appears in more than half of those setups. This suggests it has become the go-to alternative to Llama for many users around the world.

Why Are Chinese AI Models So Widely Adopted?

One reason is that Chinese labs openly publish large, high-quality weights that are optimized for local use. They focus on making models that can run efficiently on common hardware, including setups with limited resources. This makes them easier to adopt, run, and integrate into various environments.

Compared to Western companies like OpenAI and Google, Chinese developers are less constrained by safety reviews and regulatory hurdles. They tend to release models openly, which encourages more experimentation and use in different settings. This pragmatic approach helps more researchers and developers run powerful AI on their own hardware without needing huge budgets.

As a result, Chinese models like Qwen2 are often the best or only options for those wanting accessible, high-performance AI. They’re built with practicality in mind, making AI more available to a wider audience.

The Impact of Regulatory and Commercial Pressures

Western AI companies are facing increasing regulation, safety checks, and commercial pressures to keep their most advanced models behind APIs or private releases. These restrictions limit the ability of researchers and developers to access and modify the latest models freely.

In contrast, Chinese labs are demonstrating a different philosophy. They publish models openly, emphasizing usability and local deployment. This openness has helped Chinese models gain a strong foothold in the open-source community, especially as Western labs slow down or restrict their releases.

Overall, this shift could reshape the open-source AI landscape. Chinese models are filling the void left by Western labs’ cautious approach, making advanced AI more accessible worldwide.

In summary, Chinese AI models are gaining ground because they are openly available, easy to deploy, and built for practical use. As Western companies tighten restrictions, Chinese developers are becoming the main source for powerful, open-source AI tools.

Inspired by

• https://www.artificialintelligence-news.com/news/chinese-ai-models-175k-unprotected-systems-western-retreat/

Sources

• sentinelone.com