The world of humanoid robotics is undergoing a remarkable transformation. Once confined to sci-fi stories and elite research labs, robots are now becoming affordable, mass-produced tools thanks to China’s rapid innovation. Powered by a “good enough, fast, and accessible” mindset, Chinese companies are challenging costly Western prototypes and opening new possibilities for industry and everyday life, reshaping who will benefit as robots move from marvels for the few to commonplace assets for the many.

We, as humans, have always been captivated by robots, an obsession vividly captured in science fiction, from R2D2 in Star Wars to Doraemon, the friendly robotic cat. Perhaps one of the most prescient depictions of human-like machines can be found in Ray Bradbury’s 1949 short story, Marionettes, Inc., where technology blurs the line between mechanical mimicry and genuine emotion.

In Bradbury’s tale, two men seek escape from their marriages by turning to robot duplicates. Braling’s automaton, “Braling Two,” lets him vacation in Rio unnoticed, while Smith, hoping to replicate the scheme, discovers his own wife has already swapped herself for a robot. The story’s chilling twist was that Braling Two develops real feelings and locks away its creator, eager to permanently assume his life.

Though more than seventy years old, Marionettes, Inc. captures anxieties and dreams that feel newly relevant as robots become not only more intelligent, but more human in appearance and behavior. Until recently, such advances seemed possible only in the realm of elite research labs in Japan, Germany, or the US, countries where robotics was shaped by years of government funding and academic rigor.

Yet, just as science fiction often foreshadows reality, today’s robotics landscape is being dramatically reshaped. The emergence of affordable, commercially viable humanoid robots, driven increasingly by Chinese manufacturers, signals a pivotal shift in both technology and global influence. This transition from specialized Western innovation to mass-market Chinese production raises fundamental questions: What happens when humanoid robots become tools for the many, rather than marvels for the few, and how might this reshape who leads the next era of robotics?

In the US, robotics development has traditionally been a slow and costly journey, one defined by a relentless pursuit of technological prowess. Few companies embody this “capability-first” ethos better than Boston Dynamics, a firm launched in 1992 as an offshoot of MIT’s Leg Laboratory and with early backing from the Defense Advanced Research Projects Agency (DARPA). Its founding mission was to build robots able to tackle the world’s toughest engineering challenges.

This ambition crystallized with landmark projects like BigDog, a quadrupedal robot designed in 2005 to traverse dangerous, uneven terrain alongside soldiers. Though BigDog never saw active military deployment, its design and the research it inspired came to define Boston Dynamics’ culture of prizing innovation above marketability.

From BigDog’s legacy emerged celebrated breakthroughs such as Spot, the agile four-legged robot, and Atlas, a humanoid machine famed for feats like carrying heavy loads and executing backflips. Yet, the pursuit of unrivaled capability has come at a considerable cost. Despite technological achievements, Boston Dynamics has struggled to convert its world-class research into profitable products. Its shifting ownership, first Google X in 2013, then SoftBank in 2017, and Hyundai Motor in 2020, reflects repeated industry attempts to monetize its technological edge, mostly to little financial avail.

Spot’s $74,500 price tag illustrates this divide. Designed for niche, high-stakes tasks like documenting construction, monitoring hazardous sites, and providing situational awareness where human labor is risky and technical failure expensive. As Michael Perry, the company’s former VP, remarked, Spot fills roles “where the cost of human labor is high and the financial penalty for equipment failure is enormous.” For most would-be buyers, however, the technology remains out of reach, a testament to the market limits of the capability-first philosophy.

More recently, the global robotics landscape has been transformed by the rapid ascent of Chinese companies such as Unitree, DeepRobotics, and Astribot. Unlike their Western predecessors, these firms are propelled not by government research grants alone, but by China’s broader push toward high-end manufacturing and its strategic goal of tech-sector leadership. Their approach diverges sharply from legacy Western firms in both product philosophy and target markets.

At the heart of this new wave is a “good enough” mindset that values bringing functional, affordable robots to the market as quickly as possible, even if that means sacrificing some top-end performance. The strategy is embodied by Unitree’s G1 humanoid: at 127 cm tall and 35 kg, it is both much smaller and lighter than Boston Dynamics’ Atlas, which stood at 150 cm and weighed 89 kg in its 2024 iteration.

These disparities in size and weight reflect more than just design choices, but rather represent a fundamental difference in function. The G1’s 23 degrees of freedom (DOF) enable it to perform basic movements and carry payloads up to 2 kg, whereas Atlas’s 50 DOF and 10–11 kg payload capacity enable more advanced industrial tasks. While Western companies strive to push the limits of robotics for specialized, niche markets, Chinese firms aim for a versatile, “good enough” robot suitable for a wide range of entry-level uses.

Usefulness, too, is defined differently. Chinese firms prioritize mass market adoption, believing that widespread use generates invaluable feedback and iterative improvement. This stands in contrast to the Western belief that technological superiority itself will eventually generate a broader user base across industries. Crucially, China’s expansive domestic supply chain, bolstered by crossover components from the drone and electric vehicle industries, allows its firms to manufacture robots at prices Western companies cannot match. Unitree’s G1, for instance, starts around $16,000. Boston Dynamics’ 2024 Atlas, by contrast, is expected to cost over $140,000; not only do the two robots serve different needs, but they are priced for entirely different markets.

Another distinctive hallmark is the emphasis on openness. Where Western robotics firms often maintain proprietary, closed systems, Chinese companies like Unitree and UBTech advance open-source development and modularity. Unitree supplies comprehensive SDKs and robust ROS support, encouraging academic labs and industry alike to build upon their platforms. This culture of accessible collaboration accelerates knowledge sharing and product iteration, much as NVIDIA once did by distributing GPUs and fostering a broad developer ecosystem for its CUDA platform.

Taken together, the interplay of low cost, mass-market appeal, and open-source collaboration in the Chinese model creates a powerful flywheel: affordable robots stimulate broad adoption, widespread user engagement drives rapid improvement, and shared development resources accelerate learning across the ecosystem. Each element amplifies the next, fueling continuous growth and innovation.

Though Chinese robotics firms may not yet match the technical achievements of industry pioneers like Boston Dynamics or Figure AI, their strategy is already reshaping what progress looks like. Companies such as Unitree have reached profitability, an elusive goal for their Western counterparts. Unitree achieved profitability by unlocking accessibility and scale, not by superior technology or unmatched capabilities. Backed by both state support and investments from tech giants like Tencent and Alibaba, China’s robotics sector is charting a path that mirrors its earlier surges in solar panels and electric vehicles, combining speed, openness, and market-driven feedback to build influential technology from the ground up.

In an industry once dominated by expensive prototypes and niche applications, the rise of the “good enough, fast, and accessible” approach signals a fundamental shift. As China’s flywheel gathers momentum, the question is no longer just who builds the most advanced robots, but who brings them out of the lab and into the hands of millions. 

Image credits: Northern Bites Bowdoin Robocup via Flickr

Andy Liao
Andy is a graduate of the University of Toronto and a researcher of Chinese electric vehicles. He previously consulted for the World Bank on fintech in emerging markets and worked with the IFC analyzing over-indebtedness among women in Sri Lanka. In his free time, Andy enjoys reading and is currently training for a half marathon.