Tech magic

From viral stage performances to street-side applications, AI in China is moving out of the R&D labs and into the concrete reality of daily life

Recently, Tesla CEO Elon Musk shared a video clip showing humanoid robots, developed by China’s Unitree Robotics, presenting coherent, rhythmic movements alongside human dancers at a concert. The post, which Musk described as “impressive”, drew over 20 million views globally. Yet, what the video, which has gone viral, shows is more than a mere mechanical stunt. It marks a pivotal step forward in the country’s technological development. Only months ago, during the Spring Festival Gala, similar robots were seen giving a performance, but that was a scripted, rigid one. The rapid leap from controlled display to dynamic public performance mirrors a broader, structural transformation occurring across China: artificial intelligence is migrating from the abstract realm of algorithms and laboratories into the tangible, complex and vibrant reality of everyday life.

For international observers, China’s advances in AI are often framed through the lens of computing power chips or manufacturing metrics. However, this perspective overlooks a defining feature of the Chinese model, which is its application-oriented development path rooted in a “people-centered” approach. Under the strategic umbrella of “Digital China”, intelligent technologies are not only being stockpiled as strategic assets; they are being aggressively deployed into real-world consumer and service settings. This strategy is effectively turning the entire country into a vast, open-air laboratory for innovation, creating a feedback loop that accelerates both technological maturity and social acceptance.

This trend is most visible in the physical reshaping of urban spaces, where AI is becoming a tangible presence. A compelling example can be found on Beijing’s Qianmen pedestrian street. The city launched Galaxy Pod, a groundbreaking autonomous retail store. Resembling a compact space capsule dropped in the midst of centuries-old architecture, this unit is operated entirely by a self-developed humanoid robot. Without any human intervention, the robot handles complex orders given orally, processes payments and delivers products to customers. More than just a vending machine, the pod serves as a cultural touchpoint by offering AI-powered photography and greeting visitors in the local Beijing dialect. It blends heritage with futuristic innovation, tangibly demonstrating Beijing’s vision as a global consumption hub. When technology is presented in such a familiar, low-risk format, it demystifies the concept of “robots” for the average citizen, transforming fear into curiosity and, eventually, into utility.

Beyond physical robots, the integration of large language models into public services exemplifies how China is leveraging new quality productive forces to enhance governance efficiency. DeepSeek, a cost-efficient Chinese large language model, has been rolled out nationwide since early 2025 with remarkable speed. In the bitter cold of Mudanjiang, a city in Heilongjiang province, AI is powering service windows in government halls. These systems understand the nuances of everyday speech and deliver instant, accurate responses to complex social security queries. Accessible around the clock, they have redefined public service standards by extending support beyond traditional office hours, offering unprecedented flexibility to residents. This deployment aligns with the government’s broader push to modernize public services delivery and respond more effectively to residents’ needs, ensuring that the benefits of digital advancement are felt directly by the people.

Crucially, the visibility of this technological wave is not confined to tech-savvy urban centers; it has become a ubiquitous presence across diverse demographics and regions. This universal reach ensures that AI serves as an infrastructure for many rather than a luxury for a few. In rural areas, the expansion of broadband and AI-assisted agriculture has narrowed the digital divide for over 200 million users. For the aging population, an AI experience pavilion in Shanghai demonstrates how technology adapts to human needs rather than the other way around. There, companion robots interpret emotional cues and smart exoskeletons assist with mobility, offering scalable solutions for dignified independent living. Similarly, at the 2025 “Technology Without Barriers” event, participants highlighted how AI is opening professional worlds for the blind by enabling careers in audio editing and coding. These examples underscore a key aspect of China’s strategy: making AI tangible and accessible to everyone, regardless of age, location or physical ability.

From a governance perspective, this strategy of bringing AI into the public eye serves a critical regulatory function. When emerging technologies operate in plain sight, they stimulate informed public discourse and generate real-time feedback for policymaking. A case in point is Jiangmen, Guangdong province, which launched Baidu’s Apollo Go autonomous ride-hailing service in September 2025. Just seven months after the local operating entity was established, the city had already put in place a comprehensive regulatory framework. It designated testing zones in Jianghai district and set operational hours from 9 am to 5 pm to avoid peak traffic. This philosophy of creating “safe sandboxes” for innovation was equally evident at the 2025 Humanoid Robot Half Marathon in Beijing.

Co-organized by sports authorities, tech regulators and urban planners, this world-first event functioned as a dynamic testing ground where market experimentation and public oversight converged. Such initiatives allow developers to push their hardware to the limit while regulators observe safety parameters in a controlled environment. Far from a tech-driven free-for-all, these examples exemplify China’s evolving model of “responsive governance”, where regulation advances through iterative learning and proactive institutional design.

The deep embedding of intelligent systems into consumer sectors has substantial economic implications. The widespread adoption of AI is acting as a catalyst for China’s shift toward a service-led economy. In the first three quarters of 2025, the service sector accounted for 58.4 percent of GDP and contributed nearly 61 percent to overall economic growth. Within this context, service retail rose by 5.2 percent, outpacing goods retail. This growth is being fueled by a demand for “experience” and “quality”, usage patterns that AI is uniquely positioned to satisfy. Whether it is digital cultural services, or personalized travel recommendations generated by AI agents, technology is creating new value by transforming lived experience into a modern economic currency.

At a time when many nations remain entangled in abstract debates over the risks versus benefits of artificial intelligence, China’s experience suggests a compelling alternative: “oversight through deployment”. Rather than relying on static regulations drafted in a vacuum, policymakers are creating realistic environments where technologies can be refined alongside their users. This approach offers a pragmatic lesson for both developing and advanced economies. It demonstrates that the resilience of an AI ecosystem is not measured solely by the sophistication of its code, but by how deeply it is embedded in the rhythms of society.

When people encounter AI through helpful, familiar experiences — such as a robot navigating a subway station or a kiosk answering pension-related questions — they are more likely to accept it. As a senior resident in Shenzhen remarked regarding a patrol robot: “If it helps me get where I need to go, why should I fear it?” By fostering this form of participatory governance, where residents become co-shapers of the technological future, China ensures that the era of intelligence is guided by a vision of shared prosperity and tangible benefit for all.

The author is an associate professor at the Business School at the University of International Business and Economics.

The author contributed this article to China Watch, a think tank powered by China Daily. The views do not necessarily reflect those of China Daily.

Contact the editor at editor@chinawatch.cn.