Eighteen robots displayed at the 2024 WAIC in Shanghai, China.
Ni Tao is IE’s bi-weekly columnist, giving exclusive insight into China’s technology and engineering ecosystem. His Inside China column explores the issues that shape discussions and understanding about Chinese innovation, providing fresh perspectives not found elsewhere.
If there was anything truly worth seeing at the recent World AI Conference and High-Level Meeting on Global AI Governance 2024, it was undoubtedly humanoid robots.
On a scorching day last week, I pushed past hordes of people gathered inside the Shanghai World Expo Exhibition & Convention Center to get a clear view of a spectacle.
Standing in the milling crowd, I craned my neck. Eighteen Chinese-made humanoid robots were arranged in two rows for public display. They were somewhat awkwardly dubbed “Eighteen Arhats,” an apparent allusion to a group of legendary Buddhist figures.
Around me, the crowd eagerly snapped photos while tech influencers on livestream bellowed into their smartphones.
The WAIC, an annual extravaganza held in Shanghai to showcase the latest tech achievements from home and abroad, was a big disappointment this year.
This year’s event felt subdued without the buzz generated by last year’s frenzy over generative AI and large language model (LLM). Despite the conference’s name, China’s domestic AI players had little to show this time to be deserving of their previous bombast. Consequently, one might say that fancy-looking humanoid robots became the sole attraction—or saved the day.
Exhibitors at WAIC 2024 occasionally demonstrated their robots’ walking abilities, but only within the confines of their booths and securely tethered to prevent mishaps. Tesla’s humanoid robot Optimus Gen 2 was why many visitors were drawn to the tech show, but it only sat immobile behind a glass wall.
“What good does that do?!” a journalist friend grumbled. But it’s not the first time that Tesla chose to play it safe by showcasing its robot inside a glass box.
My biggest impression of the WAIC 2024 was that Chinese humanoid robot companies were too numerous to list. What surprised me the most was the sight of a legion of startups previously unheard of and with dubious track records now also jumping on the humanoid bandwagon.
Some put up a humanoid figure or even a half-finished shell without any discernible technological edge or unique features. Questions about their products’ commercial prospects often elicited vague responses.
One exhibitor equivocated, “We just entered this segment, and so we will keep rolling out more advanced features over time,” deflecting my question on application scenarios.
These observations at the WAIC 2024 epitomize the current state of affairs of China’s humanoid robot sector. With new startups flooding in, some without prior robotic experience and aided by venture capital injections, overheating, and speculation are sure to follow.
In China, any segment chosen by decision-makers as a new pocket of growth will quickly become the crown jewel of the tech community and the darling of venture fund managers.
In recent years, the once-obscure domain of humanoid robotics has steadily emerged from the shadows, receiving a significant boost from last year’s “Robotics Plus” action plan.
This is a far-reaching strategy devised by China’s Ministry of Industry and Information Technology, in collaboration with 16 other ministry-level government agencies, to spur the growth of the country’s robotic field.
Bubbles are beginning to surface at an alarming rate. Critics may accuse me of misplaced anxiety. However, in an immense market like China’s, speculative behavior through brutal price wars often undercuts and sometimes even drives out genuinely down-to-earth and competent players. This time, the peril of speculation has again reared its ugly head in the case of humanoid robotics.
Perhaps sensing this danger, a subforum on legal and ethical governance of humanoid robots during the WAIC this year unveiled a set of “Guidelines for Governance of Humanoid Robots.”
Crafted by legal and AI experts from around the country, the guidelines delve into topics such as open-source innovation, legal and ethical considerations, and intellectual property rights associated with AI-driven humanoid robotics.
This is the world’s first known document outlining governance principles for humanoid robots.
Admittedly, some of the guidelines’ provisions are remarkably forward-thinking and are apparently inspired by current societal debates. For instance, in the first part on goal and vision, the guidelines’ authors emphasize the need to ensure that intelligent design and application of humanoid robots do not compromise human safety, dignity, or security.
They advocate for a symbiosis between humans and machines to enhance societal development and prevent technological misuse.
These all look very nice on paper. They also address long-standing concerns about the widespread adoption of humanoid robots: What will happen to blue-collar jobs? Will robots take over and wreck human livelihoods? What’s worse, if humanoid robots are armed and used in modern warfare, how should humanity respond?
In light of these concerns, I spoke with Wang Xingxing, founder and CEO of Unitree, one of the world’s top humanoid robotics firms. Wang took ethical considerations a step further.
“In addition to compliance with local laws and regulations, users of intelligent robots must also employ them responsibly and avoid modifications or applications that pose direct or potential dangers to mankind,” he told me.
Wang’s remarks resonate because questions like who is using humanoid robots and in what way do pose a real-world dilemma beyond the control of manufacturers like Unitree.
Unitree had a brush with controversy when its quadruped robot was shown in online videos to be firing live rounds from a gun mounted on its back. This expose plunged Unitree into the maelstrom of fear and loathing. The underlying message is clear: After a weaponized robot dog, is a gun-toting humanoid next?
In anticipation of these scenarios, industry self-regulation has become paramount.
Against this backdrop, some industry leaders have welcomed the guidelines, viewing them as a timely reiteration of ethical principles already built into their operations. While currently more aspirational than binding, they serve as a necessary reminder to the industry about its responsibilities.
“We are in AI and embodied intelligence mainly to improve human life. Therefore, the premise is definitely not to harm humans,” said Yue Xi, co-founder of Robot Era, a leading Chinese humanoid robot startup.
Xi stated that the industry already takes ethical concerns about humanoids not harming humans seriously. However, he questioned some of the language in the guidelines, particularly in the section on risk management.
It mandates adherence to preset rules and procedures without allowing for flexibility in task goals or execution methods for humanoid robots.
In Xi’s view, this is out of touch with reality. Humanoid robots generally possess autonomous learning capabilities, unlike traditional industrial or other robots that strictly follow programmed paths.
He believes future robot guidelines may need to be more explicit on how tasks to be carried out by humanoids can adapt to predefined scripts.
Overall, he lauded the guidelines, noting that humanoid robots have seen explosive growth in the past two years compared to the previous four decades.
He noted that with more companies entering this race, the release of guidelines is crucial to steer everyone toward compliance with ethical, legal, and technical norms.
Interestingly, when I asked executives at several leading humanoid robot companies about their perception of the guidelines, they either claimed ignorance or sounded indifferent.
This raises questions about their involvement in drafting and whether their voices were adequately represented.
In contrast, friends who work for international media eagerly requested a copy of the guidelines for review. The difference between domestic and international perspectives on the same document is quite revealing: In today’s China, industry self-regulation and regulatory scrutiny can sometimes outpace industry expansion.
Only time will tell whether this benefits or harms the industry and society.
Stay up-to-date on engineering, tech, space, and science news with The Blueprint.
By clicking sign up, you confirm that you accept this site's Terms of Use and Privacy Policy
Ni Tao Ni Tao worked with state-owned Chinese media for over a decade before he decided to quit and venture down the rabbit hole of mass communication and part-time teaching. Toward the end of his stint as a journalist, he developed a keen interest in China's booming tech ecosystem. Since then, he has been an avid follower of news from sectors like robotics, AI, autonomous driving, intelligent hardware, and eVTOL. When he's not writing, you can expect him to be on his beloved Yanagisawa saxophones, trying to play some jazz riffs, often in vain and occasionally against the protests of an angry neighbor.
0