US-China AI War Heats Up: Who Will Dominate the World?

Competition between the United States and China in the development of artificial intelligence (AI) has entered a new chapter. While Washington has long been known as the pioneer of the most technically sophisticated AI models, technology companies from Beijing have taken the opposite path: creating AI that is sufficiently capable, but far cheaper, more efficient, and easier for global markets to adopt.

That strategy is beginning to bear fruit. Over the past year, Chinese AI models have increasingly been used by companies, governments and businesses across various countries, from Southeast Asia to the Gulf region.

A report by The Washington Post notes that the AI race is no longer determined solely by who possesses the best technology, but by who can deliver technology that is actually used by society.

Cherie Shi, Global Business Manager at MiniMax, said the measure of AI success is no longer merely achieving the highest technical feats.

“The frontier of AI is no longer about reaching technical benchmarks, but about how many people in the real world actually use our model every day,” she said, as quoted by The Washington Post.

The statement reflects a shift in strategy among Chinese AI companies. Their focus has moved from chasing the title of most advanced AI to becoming the most widely used AI.

Although American AI companies such as OpenAI and Anthropic are still regarded as leaders in technological capability, a number of analysts believe that advantage may not last long.

A report published by the JPMorganChase Center for Geopolitics in May stated that the capabilities of leading AI models are expected to converge, with the differentiating factors becoming quality of implementation, integration with industrial needs, and the ability to attract users.

The institution even assessed that AI laboratories in China are developing models designed to “win the adoption race.”

Meanwhile, the cost of using AI has become a concern for many companies. After previously racing to integrate AI into their business operations, they now face a surge in token costs — the unit of AI data consumption.

This situation has opened up significant opportunities for Chinese AI companies offering lower usage costs alongside models that are more economical in computing resources.

A sales representative from a major Chinese AI company, whose identity was withheld, said many companies do not actually need the world’s best AI capabilities.

“Many companies only need a fraction of the capabilities offered by Anthropic or OpenAI. If we can deliver 80 per cent of the benefit at a far lower cost, that is enough. We will capture that market share,” the representative said, as quoted by the same outlet.

Gunja Gargeshwari, Chief Revenue Officer of Bright Data, has observed a similar trend.

“There is no denying it — their cost efficiency and token usage are truly remarkable.”

According to Gargeshwari, a growing number of Bright Data clients have begun experimenting with Chinese-made AI models.

The success of Chinese AI companies is visible not only in technology, but also in business expansion.

Singapore, for instance, selected Alibaba’s Qwen AI model to support a national AI initiative. In Malaysia, Chinese technology companies are involved in building a smart city project worth 20 billion US dollars. Meanwhile, Saudi Arabia has partnered with ByteDance and Huawei to apply AI to urban infrastructure.

One of the most aggressively expanding companies is Zhipu AI, also known as Z.ai.

Z.ai Product Director Li Zixuan likened the AI competition between America and China to a luxury car comparison.

The analogy suggests that American AI indeed remains superior in quality. However, Chinese AI is considered highly competitive, more affordable, and has a greater chance of being widely adopted.

Despite being placed on a United States trade restriction list over alleged ties to the Chinese military, Zhipu continues to expand its international business. The company has opened offices in Singapore and Dubai and has begun securing various government contracts across Asia.

OpenAI has even highlighted Zhipu’s aggressive expansion. In a blog post last year, OpenAI wrote that the company was attempting to:

“Embed Chinese AI systems and standards in various developing countries before competitors from the United States or Europe can do so,” the OpenAI statement read.

Zhipu’s flagship product, GLM, is now claimed to have more than 4 million users across 218 countries.

Beyond low prices, many Chinese AI models are available as open source. This means companies and governments can download the models and run them in their own data centres without depending on cloud services.

According to Cherie Shi, this approach appeals to many governments seeking to ensure all data remains under their own control.

Such models are also in demand from companies handling sensitive data, such as the banking and healthcare sectors.

Aleksander Mordvinov, founder of a financial technology start-up in Thailand, said he built his company’s entire AI system using Qwen.

“Every company using cloud services could eventually suffer a data breach. I cannot take that risk,” he said.

For that reason, he chose to use AI that can be run locally.

Martijn Rasser, Vice President of the Special Competitive Studies Project (SCSP), said uncertainty in United States policy has also begun to affect user confidence.

“Many users are worried because access to American AI products cannot always be relied upon. That is clearly not a value-add,” he said.

The comment came after the administration of President Donald Trump restricted foreign nationals’ access to Anthropic’s most advanced AI models.

The policy even briefly boosted the share prices of several Chinese AI companies, including Zhipu.

At a technology forum in Paris, Alibaba Chairman Joseph Tsai urged European countries not to depend on just one AI provider.

“The problem now is that all your eggs are in one basket. Why not choose a second basket to spread the risk?”

According to Tsai, having alternative AI from China would give countries and companies greater flexibility.