The scene was repeated at several events over several days in the Chinese capital, Beijing, and in March in the southern tech hub of Shenzhen, engineers were seen helping crowds set up the popular AI “agent” OpenClaw on their laptops.
“I’m worried that we will fall behind in technological development,” said Sun Lei, 41, a human resources manager at Cheetah Events. She said she hopes the tool will help obtain and screen resumes on various recruiting platforms.
More than a year after OpenAI’s Chinese rival DeepSeek wowed the world with its advanced AI models, China has become a testing ground for the mass use of AI tools. While AI models built in the U.S. still dominate in raw computing power, people and businesses in China are rapidly embracing the technology, facilitating rapid and widespread adoption in nearly every possible field.
With the adoption of AI in the workplace and daily life at a rapid pace around the world, ordinary Chinese people are using it for everything from booking and planning trips to ordering food and hailing rides. According to a report by the government-run China Internet Network Information Center, more than 600 million people out of a population of 1.4 billion were using generative AI as of December, an increase of 142% from the previous year.
There has also been a recent surge in the use of “agent” AI like OpenClaw, including by many Chinese companies, and data consumption by AI models has also increased. According to OpenRouter, an AI “gateway platform” that tracks data and enforces security between different AI models, the weekly share used by Chinese AI models recently surpassed that of U.S. models, as measured by what computer scientists call tokens, or units of data such as parts of words.
With the introduction of AI, China will become a “world leader”
Jason Tong, a 64-year-old retired IT engineer in Shanghai, has been using AI chatbots like Doubao and Kim for everyday queries since they were first introduced several years ago.
He started paying closer attention to his health and in early March joined a blood sugar monitoring service run by a Shanghai-based company that uses AI models to generate customized health advice. He felt that a personalized and quick response would be helpful.
Tong believes that widespread adoption of AI applications in daily life is inevitable. “This is bound to happen, just as horse-drawn carriages were eventually replaced by trains.”
Chinese products incorporating AI, such as cars and robots, have made great strides, from humanoid robots with advanced cognitive capabilities to AI systems that drivers can use for more complex tasks, such as making restaurant reservations.
“The competition[in AI]is clearly moving from models to ecosystems,” said Lizzie Li, a researcher specializing in economics and technology at the Center for China Analysis at the Asian Social Policy Institute. “Users in China are essentially acting as large-scale real-time testers.”
Chinese technology companies such as Tencent, Alibaba, and Baidu are also competing to commercialize AI. Tencent has integrated OpenClaw into China’s own “super app” WeChat. Although WeChat is primarily a messaging tool, it can also be used to order food, pay for things, and more. Alibaba is incorporating “agent” AI into its workflows.
OpenClaw facilitates widespread use of China AI applications
OpenClaw was originally created last year by Austrian software developer Peter Steinberger and quickly gained fervor thanks to its ability to use a variety of tools to complete complex tasks.
Zhao Yikang, a Chinese university student in Macau, uses OpenClaw in both his studies and daily life.
During his internship at a real estate company in the southern Chinese city of Zhuhai, he was struck by how low-cost and efficient it was, and used it to automatically generate promotional videos and manage his social media accounts.
“AI can understand things instantly,” Zhao said. “You just have to act as a commander and tell them what to do.”
Zhao, who was preparing to start a photography service business after graduation, asked AI to build his company’s website. Within 10 minutes, a fully functional site was generated for less than 5 yuan (70 cents).
At one point, Chinese authorities issued multiple warnings about potential security risks related to the OpenClaw AI “agent,” including data leaks following a surge in installations, but widespread interest persisted.
Janet Tan, a partner and managing director at consulting firm AlixPartners who focuses on technology, said Chinese companies are increasingly setting internal goals to encourage the use of AI to improve efficiency.
“There are many application scenarios,” said Wang Xiaogang, co-founder of Chinese AI software company SenseTime and chairman of ACE Robotics. “The industry is evolving very rapidly, and people are very open and eager to try AI in many scenarios.”
US export controls both help and hinder AI use in China
China has aimed to position itself well by investing heavily in training human resources and securing access to abundant, affordable electricity for the development and advancement of power-hungry AI.
To achieve breakthroughs in technology, including AI, China’s leaders have pledged to increase national research and development spending by more than 7% annually in a five-year plan through 2030. The ‘AI Plus’ national blueprint outlines steps to integrate AI into many areas of life, from healthcare to education. The court said judges in Shenzhen handled a 50% increase in cases last year, in part due to the use of AI tools to assist in judicial proceedings.
But U.S. regulations that limit access to some of the world’s most advanced computer chips remain a bottleneck to China’s AI progress.
“Tool export controls are slowing down China’s chip manufacturing capabilities and are an Achilles heel for many AI labs that need advanced AI chips,” said Sam Sachs, a senior fellow at New America who specializes in Chinese technology policy.
However, the regulation also led to improved coordination of design, manufacturing, and adoption across China’s technology supply chain. “Over time, this dynamic could facilitate rather than hinder China’s ambitions,” Sachs said.
China is becoming an “AI innovator”
When China’s DeepSeek released a preview of its long-awaited V4 AI model last month, one of the big changes was that it was now partially supported by a computer chip made by Chinese tech giant Huawei. This means less dependence on top U.S. chipmakers such as Nvidia.
A recent report by Stanford University’s Institute for Human-Centered AI found that the gap between the United States and China in the performance of top AI models has “virtually closed.”
US policymakers and top AI companies such as Anthropic and OpenAI have accused Chinese AI startups of stealing US AI technology. China says such claims are baseless.
Lian Jie Hsu, principal analyst at research and advisory group Omdia, believes the AI gap between the US and China will continue to narrow, despite US export restrictions and China’s Great Firewall, the ruling Communist Party’s extensive internet filtering and censorship system.
Analysts including Su believe that hurdles such as the Great Firewall are likely to have a limited impact on China’s use of AI, as the technology is already being tested, integrated and expanded in China’s controlled internet environment.
“It won’t be long before China moves from being a fast follower to being a parallel innovator,” he said.
