AI Race Update: China Faces Uphill Battle Against the U.S.

Buzz

Ngày cập nhật gần nhất: 15/4/2026

Content

Where Does China's AI Industry Stand?

If Washington Tightens AI Regulation, Will China Have an Advantage?

How Does Semiconductor Export Ban Impact?

What is China's Goal with AI?

How Does China Plan to Govern AI?

What Has the U.S. Learned?

The intense race for technological dominance between the United States and China has never cooled down. This year, the competition has entered a new realm, with artificial intelligence algorithms like ChatGPT becoming the hottest buzzword in the first 6 months.

AI holds tremendous potential, from chatbots creating content to algorithms powering self-driving cars. The world's two largest economies are both sprinting to gain an edge over each other, developing the most powerful AI technologies. However, according to many analysts and experts in the field, China is significantly lagging behind Western countries in the current AI race. The only advantage Beijing has over Washington is that the Chinese government has taken steps to regulate AI technology.

On the flip side, such regulatory actions, especially in the realm of AI content creation, pose certain obstacles for Chinese tech conglomerates in algorithm development. Take Douyin, the mainland Chinese version of TikTok, which has specific regulations requiring clear markings on AI-generated content to avoid misleading viewers.

However, in the U.S., there are perspectives suggesting that if the U.S. government sets standards and regulations for AI governance, they might lose the advantage in research to China.

Where Does China's AI Industry Stand?

Chinese tech giants have been investing in AI for many years, ranging from deploying thousands of surveillance cameras on the streets to implementing facial recognition systems for... monitoring paper conservation in public restrooms. When it comes to tracking and identification technology, China is currently at the second position, with no one daring to claim the top spot.

However, in other AI-related research areas, China is lagging considerably behind. Partly due to the government's tight control over information and communication. As a result, Chinese tech conglomerates focus more on researching AI with clear commercialization potential, rather than algorithms like ChatGPT or Midjourney for chatbots and image creation tools.

If they want to use language models, for the most part, China has to buy pre-developed LLMs from foreign countries.

Jeff Ding, associate professor of political science at George Washington University, who focuses on researching the AI race between the U.S. and China, believes: “Many Chinese companies have lower investment capital and less freedom in basic AI research. So most of the time, they have to follow trends established by U.S. companies earlier.”

If Washington Tightens AI Regulation, Will China Have an Advantage?

In a recent study, Professor Ding found that the majority of large-scale language models developed in China are outdated compared to LLMs developed in the U.S. by about 2 years. This gap is challenging to narrow, even if the U.S. takes strict regulatory measures in AI research.

This gap also makes it challenging for China to attract talent worldwide. Everyone wants to work in units with abundant capital and the freedom to experiment in new machine learning research areas.

Experts argue that as U.S. lawmakers debate how to regulate AI to mitigate risks while not stifling future development, they will have a clear advantage. In discussions, they can find a way that China cannot.

Not to mention, the semiconductor export restrictions imposed by the U.S. on Chinese companies could become a significant barrier in China's efforts to race ahead in AI development.

How Does Semiconductor Export Ban Impact?

Since late last year, the Biden administration has started restricting semiconductor exports to China, encompassing products, solutions, technologies, and devices created by U.S. companies, as well as those produced by foreign corporations using U.S. technology. Washington states that this decision aims to prevent U.S. technology from being used for China's military purposes or finding its way to Russia.

Alongside other industries, this undoubtedly creates pressure and challenges for Chinese tech companies. They almost cannot access the world's most powerful processing chips currently designed for one sole purpose – handling massive-scale AI algorithms. Beijing labels the U.S. semiconductor export ban as 'authoritarian behavior' aimed at ensuring technological monopoly.

Recent research reveals that among the top 20 most popular LLMs in China, 17 algorithms are entirely dependent on Nvidia's GPUs, with only 3 language models trained and operated using domestically-produced chips.

So, Chinese companies also have to find ways to secure the supply of AI processing chips in the black market. In this realm, an A100 card, the most powerful in the Ampere generation, costs twice as much as the official price. As for the H100, not exported to China, Nvidia had to create a lower-grade version, known as the H800, to serve this market.

What is China's Goal with AI?

Despite the aforementioned challenges, Chinese AI research companies have achieved remarkable breakthroughs in various forms of artificial intelligence technology, including facial recognition, motion detection, and AI combined with virtual reality.

These technologies have actively contributed to the operation of massive citizen surveillance systems in China, and they have no hesitation in promoting this technology to other countries. Evidence of this is seen in the contracts to build smart cities in Belgrade, Serbia, and Nairobi, Kenya.

In Western countries, pressure from lawmakers fundamentally restrains the pace of developing AI technologies for human tracking purposes. Instead, companies in the U.S. or the U.K. focus on creating chatbots for entertainment or innovation and then bring them to individual user communities.

How Does China Plan to Govern AI?

As mentioned, some observers and researchers believe that Beijing's approach to AI governance might limit the development and innovation capabilities of corporations and research units.

However, in the aspect that the U.S. is still struggling to manage, all AI development companies in China must adhere to regulations on intellectual property rights, personal data protection, and algorithms governing content display, especially content generated by machines. In April, Chinese regulators introduced a draft regulation for AI content creation, enabling solutions like Stable Diffusion and ChatGPT.

Certainly, every AI created in China must comply with Beijing's requirements. As a result, filters are in place to remove content deemed as violations. This clearly imposes a limitation on the creative capacity of the AI development industry in China.

These limitations are the reason China currently has not publicly tested any large-scale language models, as nobody wants users to create content contrary to the Chinese government's stance.

What Has the U.S. Learned?

The drawback of strict AI regulation is stifling innovation, but there are principled factors that Washington can learn from and adjust to avoid excessive control while addressing concerns about AI's negative impact on society: Protecting human privacy, labeling AI-generated content, and alerting governments and managers if AI poses a potential threat.

Johanna Costigan, a researcher at the Asia Social Policy Institute, believes that if the U.S. enacts AI regulations, it may not be as forceful and rigid as what China has established. However, it can still adhere to existing laws, prevent algorithmic biases, and safeguard the rights of individuals: “Regulations can always be created to coexist with innovation.”

Source: The Washington Post

Frequently Asked Questions

What are the main factors contributing to the AI race between the US and China?

The AI race between the US and China is driven by technological advancements, investment in AI research, regulatory approaches, and the competition for global dominance in various AI applications. Both nations aim to develop superior AI technologies that can enhance their economic and military capabilities.

How does the semiconductor export ban affect China's AI development?

The semiconductor export ban imposed by the US significantly hampers China's AI development by restricting access to advanced processing chips crucial for handling large-scale AI algorithms. This creates challenges for Chinese tech companies as they struggle to secure the necessary technology for their AI projects.

Is China currently lagging behind the US in AI technology research?

Yes, China is significantly lagging behind the US in several key areas of AI technology research. Analysts have noted that many Chinese language models are outdated compared to their US counterparts, impacting the country's ability to attract top talent and innovate in this field.

What impact does AI regulation have on innovation in China?

AI regulation in China tends to limit innovation by imposing strict guidelines that corporations must follow. While these regulations aim to protect data and ensure compliance, they can also stifle creative development in AI technologies and restrict the use of advanced models.

How is the US government's approach to AI regulation different from China's?

The US government's approach to AI regulation is generally more flexible and aims to balance innovation with safety. In contrast, China's regulations are more stringent, focusing heavily on controlling content and intellectual property, which can restrict the creative potential of AI technologies.

What advancements has China made in AI despite regulatory challenges?

Despite regulatory challenges, China has achieved significant advancements in AI technologies such as facial recognition and motion detection. These technologies are integral to national surveillance systems and have been marketed internationally through contracts for smart city developments.

Mytour's content is for customer care and travel encouragement only, and we are not responsible.

For errors or inappropriate content, please contact us at: [email protected]