Hey there, tech enthusiasts and China watchers! Your friendly neighborhood American blogger in the Middle Kingdom is back, and this time, things are getting heated in the world of robotics, specifically, humanoid robots. If you thought the electric vehicle (EV) scene in China was intense, buckle up, because the robot revolution is revving its engines, and it’s causing some serious fireworks in the normally cool-headed world of Chinese venture capital.
For those of you just tuning in, China is going absolutely bonkers for humanoid robots. Think less clunky factory arms and more sci-fi movie droids, capable of walking, talking, and maybe even doing your dishes someday. This isn’t just some niche tech fad; we’re talking about a potential industry that could reshape manufacturing, service sectors, and even our homes. The Chinese government is pushing hard for technological self-reliance, and robotics, especially embodied intelligence (that’s the fancy term for robots that can interact with the physical world in a smart way), is right at the forefront of that drive.
But as with any tech gold rush, especially here in China where things move at warp speed, questions are starting to swirl around whether this humanoid robot hype is the real deal or just another tech bubble waiting to pop. And let me tell you, the debate is getting spicy, folks.
Recently, a major dust-up erupted in the Chinese investment community, a full-blown digital shouting match, if you will, sparked by none other than Zhu Xiaohu, the managing partner at GSR Ventures. Now, Zhu isn’t just some random VC guy; he’s a bonafide tech investment heavyweight. Think of him as the guy who got in early on Ofo (the bike-sharing company that soared and then spectacularly crashed) and Ele.me (the food delivery giant now part of Alibaba). He’s known for his sharp, contrarian takes, and this time, he’s throwing some serious shade on the humanoid robot party.
In a recent interview that sent shockwaves through the Chinese tech and finance circles, according to an article on 金石杂谈 (Jinshi Zatan), Zhu declared, “We are batch exiting humanoid robot projects.” Boom! Talk about dropping a robot-shaped bomb. His reasoning? He’s “not seeing a clear path to commercialization” and feels there’s “too much consensus” around the sector. In Zhu-speak, that’s basically a sell signal.
Now, to understand Zhu’s perspective, you need to grasp his investment philosophy. He’s a big believer in what he calls “anti-consensus investing.” His mantra is “buy in disagreement, sell in agreement.” Think of it like the stock market: you make money when you buy low when everyone else is hesitant, and sell high when everyone is clamoring to get in. He argues that profits are often made when opinions are divided, and high, unified expectations can often lead to inflated valuations and subsequent crashes – like that dreaded “open is close” trading day that haunts many investors’ nightmares. He points to past tech booms, like the pharmaceutical craze of 2020 and the lithium battery and baijiu (Chinese white liquor) frenzies of 2021, as examples of markets where uniform expectations led to a quick rise and then a sharp fall.
Zhu’s past successes in “anti-consensus” plays are legendary in Chinese VC circles. Take Ofo, for instance. GSR Ventures poured money into Ofo’s Series A, B, and C rounds back in 2016. But by December 2017, seeing that a merger with rival Mobike was unlikely, Zhu reportedly sold his stake to Alibaba at a 20% premium, making a clean exit while others, including Xiaomi, Tencent, Matrix Partners China, ZhenFund, and CPE Yuanfeng, were left holding the bag. Alibaba, it seems, became the “big sucker” in that deal, a common term in Chinese internet slang for someone who ends up overpaying in a hyped-up market.
Another example is Ele.me. GSR invested a few million dollars in 2011. When Alibaba acquired Ele.me for a whopping $9.5 billion, GSR walked away with a return of tens of times their initial investment. Zhu even celebrated in his WeChat朋友圈 (WeChat Moments, the ubiquitous social media platform in China), saying, “Before the all-out war of the Three Kingdoms breaks out, let financial investors retreat unscathed, for which we must thank Alibaba BABA.” Again, Alibaba was jokingly cast as the “big sucker.” This narrative even extends to more recent events, with some quipping that Alibaba’s $6 billion investment in Kimi, a Chinese AI chatbot, might be another instance. Ouch. No wonder some are suggesting Alibaba’s investment department needs a shakeup.
However, Zhu’s track record isn’t without its critics. As some netizens point out, he’s also had his share of misses and controversial stances. One online comment summarized it bluntly: “He didn’t like Zhang Yiming (founder of ByteDance, TikTok’s parent company), he didn’t invest in Xiaohongshu (a hugely popular Chinese social media and e-commerce platform), which was invested by his colleague (Zhang Yutong, now embroiled in a lawsuit with GSR over Kimi investments). He doesn’t mention these.” The comment further notes Zhu’s initial skepticism about large language models (LLMs), only to express interest in investing in DeepSeek (a rising Chinese LLM star) after its success. Last year, Zhu famously said he wouldn’t invest in any LLM companies, only to do a 180-degree turn in an early 2025 interview, stating, “If DeepSeek opens up for financing, I will definitely invest, price is not important, the key is to participate.” Talk about a change of heart!
But it’s Zhu’s “bubble” pronouncements on humanoid robots that have really stirred the pot. And that’s where Zhang Ying, the founder of Matrix Partners China, one of China’s top-tier venture capital firms, enters the fray. Zhang fired back at Zhu’s skepticism, stating, “Tell Boss Zhu to stop messing around.” He went on to argue that the “robotics field is booming, a major track, with a hundred flowers blooming. It’s normal to have some bubbles in the process. Over time, the humanoid robot track will definitely produce big companies.” Basically, Zhang’s take is: relax, Zhu, it’s a new industry, bubbles are part of the game, and the long-term potential is massive.
Zhang didn’t stop there. He added, “After Boss Zhu says something like this, I’m curious which high-quality companies in the robot sector will still talk to him, let alone take his money.” In less diplomatic terms, as one Chinese internet commentator translated it, Zhang was essentially saying, “Boss Zhu, nobody’s gonna play with you anymore.” Ouch, again!
This isn’t the first time Zhu and Zhang have publicly disagreed. Last year, they clashed over the best exit strategies for VCs. Zhu argued that demanding dividends had become the consensus for most early-stage VCs, aiming to recoup their principal within five years given the uncertain exit environment. Zhang countered that sophisticated US dollar Limited Partners (LPs, the investors in VC funds) wouldn’t accept or believe in a five-year dividend-based return. He maintained that top-tier LPs still believe in the long-term potential of the Chinese market and capital market exits for high-quality companies. Dividends, in his view, are merely a supplementary tool and represent a “cheap sale” for US dollar LPs, a meaningless effort.
Adding to the complexity, Wang Xingxing, founder of Unitree Robotics (the company behind those viral dancing robots), has also weighed in on the current state of humanoid robot development. He basically said that humanoid robots are still at the “L2” level of autonomous driving – early stages, folks. Everyone’s hoping to leap to “L3” capabilities – mass production and commercial viability – but that jump is a massive, several-fold increase in difficulty. It’s like going from driving assist to fully self-driving, a huge leap in technological capability and real-world application.
Despite these challenges, the hype around humanoid robots in China has been undeniably fueled by companies like DeepSeek, leading to a re-evaluation of Chinese tech assets. Stock prices for robot-related companies have skyrocketed. Even fund managers like Yan Siqian from Penghua Fund and Zhang Lu from Yongying Fund have become champions in the fund performance rankings, boasting annual gains of up to 65% and 57% respectively, largely thanks to their bets on humanoid robotics.
This year alone, 22 listed companies in the humanoid robot sector have seen their stock prices jump by over 50%, and a staggering 56 out of 108 companies in the Wind-tracked robotics sector are up by more than 20%. Companies like Changsheng Bearing, Zhongda Leader, and Xiangyang Bearing have seen incredible gains, soaring by 165%, 129%, and over 70% respectively.
However, the recent weeks have brought a dose of reality. These high-flying humanoid robot concept stocks have started to experience a painful correction. Share prices have retreated, revealing the lack of solid earnings to back up the valuations. Xiangyang Bearing, for example, which is facing allegations of fraud by its actual controllers and even imprisonment, has plummeted over 30% this week. Changsheng Bearing is down 8.78%, Zhongda Leader has fallen 12.38%, and Zhaowei Machinery & Electronics has dropped 20% from its peak. Is this the “bubble” Zhu Xiaohu was talking about? Maybe.
Unsurprisingly, online reactions to Zhu Xiaohu’s “bubble” thesis have been mixed. Some netizens are calling him “self-smart” and “attention-seeking.” Others argue that “if they can fully replace nannies and complete all housework, then they will become household necessities,” drawing parallels to the early days of electric vehicles when skeptics were proven wrong. Some comments suggest that “people can’t earn money outside their cognitive range,” implying that Zhu’s skepticism about robots and AI models stems from his own limitations. Ouch, burn! Some even bluntly said, “Zhu Xiaohu is just old.” And then there’s the brutally honest comment: “Speaking in human language, he didn’t understand it, so he exited.” Of course, there are those who agree with Zhu, with one commenter stating simply, “Clearly, he’s running away at the top.”
To summarize Zhu Xiaohu’s core arguments, as compiled by Touzhong.com, it boils down to avoiding projects with high consensus and unclear commercialization paths. His investment philosophy is deeply rooted in commercial viability, emphasizing two key dimensions: market consensus and commercial potential. The worst-case scenario, according to Zhu, is “extremely concentrated market consensus with unclear commercialization,” citing large language models last year as a prime example. He claims GSR Ventures has never made money in such cases. Their winning investments, he argues, are typically in areas with low market consensus but clear commercialization prospects.
Let’s break down Zhu’s key points from the Touzhong.com interview:
This debate isn’t just between two investors; it reflects a broader tension in the Chinese tech scene right now. Are we on the cusp of a genuine robot revolution that will transform industries and daily life, or are we just caught up in another wave of overhyped tech promises fueled by cheap capital and FOMO (fear of missing out)?
To get another perspective, let’s turn to an interview with Xu Huazhe, co-founder of Starry Sea AI, another company in the embodied intelligence space. In an interview with the Southern Metropolis Daily on March 27, 2025, Xu offered a more nuanced view, acknowledging the potential for a bubble but emphasizing the underlying value of the sector.
Xu stated, “Embodied intelligence bubbles are better than metaverse bubbles.” His reasoning is that embodied intelligence is built on “solid, useful hardware foundations,” and even if a bubble forms and bursts, it will leave behind tangible results, unlike some other hyped sectors that collapse without any lasting impact. He believes that companies with strong fundraising capabilities and the ability to produce real-world results will be able to weather the cycles and capitalize on the next wave of market enthusiasm.
Starry Sea AI, founded in September 2023, is a relatively young player in the embodied intelligence field, but it’s already making waves. Interestingly, one of its four co-founders, Xu Huazhe, is a professor at Tsinghua University’s Institute for Interdisciplinary Information Sciences, while the other three have backgrounds in autonomous driving companies. This blend of academic and industry expertise is a common trait among many Chinese AI and robotics startups.
Starry Sea AI has been on a fundraising tear, closing a 300 million RMB (approximately $42 million USD) Series A round in late February 2025. This was their fourth funding round in just a year, bringing their total funding to nearly $100 million USD, with Ant Group leading two of the rounds. Ant Group’s continued investment is a strong signal of confidence in the sector, especially given Ant’s deep pockets and strategic interest in future technologies.
Xu’s entrepreneurial journey with Starry Sea AI is his first foray into the startup world. In the interview, he discussed Starry Sea AI’s business strategy, technology roadmap, and broader industry trends, including the influx of cross-industry players, the race to mass production, and the current funding frenzy.
Cross-Industry Players and Startup Dynamics:
Xu noted the increasing number of companies from diverse sectors entering the humanoid robot space, including car manufacturers, traditional industrial robot companies, home appliance makers, and even smartphone giants like vivo. He sees this as a positive sign for the industry.
“For car companies, the electric vehicle market is already very crowded. ‘Giants’ are looking for their next growth point. What is the next upgrade point in manufacturing? Everyone naturally thinks of robots, especially the now-hot embodied intelligence robots. In terms of use, embodied robots can work in car factories, connect with home appliances, and combine with mobile phones to act as mobile brains and bodies. So, cross-industry players have strong motivations.”
While acknowledging the increased competition, Xu also sees benefits for startups. “The good thing is we have more partners to play with, and we can accomplish things together. For example, we can’t manufacture MR (mixed reality) devices ourselves, but companies like vivo, which sees MR as part of its robot technology layout, mean we may not have to spend $30,000 to buy Apple’s Vision Pro in the future. Instead, we can work with partners in the embodied intelligence sector.”
He believes the influx of players validates the sector’s future potential. “More players entering means the embodied intelligence sector is improving. Everyone believes that this track has a future and prospects and can produce results. The more people pour in, the faster the sector develops, which is generally beneficial to the industry.”
However, Xu also recognizes the challenges for startups. “How small startups deploy their strategies is actually a difficult topic because startup resources are definitely not as abundant as those of ‘giants.’ Giants can easily come up with the money we’ve raised.”
His advice for startups? “First, be firm. When you are sure about something, act and iterate immediately to occupy users’ minds or seize application scenarios. Decisiveness is an advantage for startups. Large companies are often more hesitant, afraid of missing opportunities but also unwilling to jump in immediately before seeing things clearly. This gives us valuable time to survive.”
Xu emphasizes the importance of maintaining technological leadership, especially in the early stages. “Because in the current stage, while money is important, it is still difficult to find people in the market who can make embodied intelligence happen. There may be a maximum of 20 people in the whole country who can train VLA (vision-language-action models). This gives startups the possibility of extending their technological lead time as much as possible. Once the technology matures later, as long as you spend money, you can find people who can do embodied intelligence. At this time, money becomes more important. Obviously, spending a lot of money is not a strength of startups.”
He also stresses the need for startups to quickly achieve mass production. “Like building new energy vehicles, once the inflection point of mass production arrives, it will be difficult to get into the game again. Therefore, startups must reach the inflection point with other players as soon as possible, or reach the inflection point first.”
Mass Production and Pricing Strategies:
Xu believes the “mass production inflection point” is yet to come, and it’s hard to predict the exact scale. While 2025 is being touted as the “year of mass production” for humanoid robots, Xu sees this as more of a marketing narrative.
“When you really get into hardware development, you will find that it is not as difficult as imagined. It is more about clearly sorting out each chain, ensuring that each part is well-developed, and then piecing them together. Therefore, the arrival of mass production is technically achievable. Secondly, the outside world’s enthusiasm for humanoid robots also plays a certain catalytic role in achieving more mass production.”
He downplays the recent “low-price marketing” trend, like the 39,900 RMB (approximately $5,500 USD) starting price offered by Songyan Dynamics for its humanoid robot. He sees this as more of a marketing tactic to build brand awareness in the early stages, similar to Unitree’s 99,000 RMB (approximately $13,750 USD) starting price for its G1 robot, which actually sells for over 300,000 RMB (approximately $41,600 USD) to labs. He also points out that these lower-priced robots are smaller in stature, naturally leading to lower costs.
Hardware vs. “Whole Machine + Embodied Intelligence” and Vertical Integration:
Xu believes that while entering the humanoid robot hardware market is still feasible, the real challenge and opportunity lie in creating “whole machine + embodied intelligence” solutions. He sees “vertical integration” – controlling both hardware and AI – as the optimal approach.
“‘Whole machine + intelligence’ is indeed a better model at present. The reason is that, similar to the internet, entrance is also very important to the embodied intelligence industry. Whoever has the entrance is more comfortable. With the whole robot machine, there is a data entrance. Whoever gets the data can create better models, and better models bring better experiences to users. Purely doing embodied intelligence model companies can only buy entrance from others.”
He acknowledges the alternative path of focusing solely on AI models, like US-based Physical Intelligence, which aims to empower hardware companies with its AI. However, Xu believes this approach is less suitable for China, given China’s hardware manufacturing优势 (advantage) and the tendency of Chinese robot manufacturers to want to control the entire stack.
Starry Sea AI is pursuing a high degree of hardware self-reliance, aiming for 80-90% in-house development, outsourcing only chips (Nvidia or domestic) and some motor components in the early stages, with plans to increase in-house motor component production as well.
“Stockpiling Food” and “Preparing for Winter”:
Xu emphasizes a cautious, “winter is coming” mentality, learned from his co-founders’ experience in the autonomous driving sector. He stresses “stockpiling food” (maintaining a strong cash reserve) and avoiding excessive spending and rapid scaling, prioritizing survival and long-term sustainability over short-term hype and growth at all costs.
He also believes in the power of small, elite teams for innovation, citing DeepSeek as an example. He argues that large teams can become bogged down in communication overhead, hindering innovation speed. Starry Sea AI is intentionally keeping its team size relatively small (around 80 people), focusing on hiring top talent rather than rapidly expanding with average performers.
VLA Model Demos and Investor Education:
Xu notes the prevalence of VLA (vision-language-action) model demos in the industry, often used for fundraising and publicity. He distinguishes between demos aimed at the general public, often heavily edited and visually impressive, and demos aimed at industry insiders, who look for “one-shot” demos showcasing long-sequence action execution, language instruction following, dexterity, and generalization capabilities.
He believes sophisticated investors can see through superficial demos and understand the underlying technology. Starry Sea AI has focused on “educating” its investors about the different technical challenges of robot manipulation (hand-eye coordination, object interaction) versus locomotion (walking, balancing), setting realistic expectations.
Data Acquisition and Simulation:
Data scarcity is a major bottleneck for VLA model training. Starry Sea AI uses a combination of publicly available datasets and self-collected data. They have a data collection team in Beijing and are building a data collection factory in Suzhou, aiming to reach tens of thousands of hours of data this year.
Xu acknowledges the role of simulation data for data augmentation but cautions against over-reliance on it, as the gap between simulation and reality can lead to poor real-world performance, especially in physics-based tasks. He contrasts Starry Sea AI’s approach with Galaxy General Robot’s focus on simulation data, arguing that simulation is more suitable for simpler tasks like pick-and-place but less effective for complex manipulation tasks involving friction and real-world physics.
Business Allocation Between Beijing and Suzhou:
Starry Sea AI’s operations are split between Beijing and Suzhou. Suzhou is the hardware R&D and manufacturing base, with plans for a factory capable of producing tens of thousands of robots annually (though not necessarily aiming for that production volume immediately). Beijing, where the company is officially registered (due to investment from Shougang Fund), houses the core AI research team, leveraging proximity to Tsinghua University and its talent pool. They plan to expand their Beijing presence with a larger office in Yizhuang, focusing on bipedal humanoid robot hardware development.
“Elimination Round” in 2-3 Years:
Xu believes the humanoid robot funding environment has become much easier in early 2025, with investors exhibiting FOMO. He estimates that a minimum of $100 million USD in funding is needed for a humanoid robot company to reach a certain scale, primarily due to high R&D costs in human resources and computing power.
While acknowledging the current bubble-like conditions, Xu sees it as a “better” bubble than the metaverse bubble, due to the tangible hardware foundations. He predicts an “elimination round” in 2-3 years, where companies without real technological capabilities, commercial viability, and strong fundraising will be weeded out, with talent and resources consolidating around the leading players. He remains optimistic about the long-term potential of the sector, anticipating further breakthroughs and market enthusiasm in the years to come.
To further understand the nuances of this debate, let’s turn to an article by Qian Wenying, Secretary-General of the CEIBS Center for AI and Management Innovation, published on March 28, 2025, titled 今年具身智能热：爆发还是泡沫？ (This Year’s Embodied Intelligence Heat: Explosion or Bubble?). Qian offers a more academic and consumer-centric perspective on the embodied intelligence hype.
Qian’s article dives deep into whether the current embodied intelligence wave is a genuine technological revolution or just another capital-driven bubble. She examines the social value of embodied intelligence, China’s advantages in the industry, and the most likely paths to commercialization.
Qian analyzes the embodied intelligence trend from three key perspectives: the consumer, the industry observer, and the management scholar.
Consumer Perspective: Rigid Demand or “Nice-to-Have”?
Qian starts by defining embodied intelligence broadly, encompassing products that interact with the environment through perception, understanding, decision-making, and action, going beyond just humanoid robots. She includes smart home devices, wearables, service robots, and smart cars in this definition, questioning whether these products constitute “rigid demand” for consumers or are driven more by technological novelty and market hype.
She argues that while smart home devices like smart speakers and lights enhance convenience, they are “icing on the cake” rather than essential necessities. Smart home adoption rates remain relatively low globally (around 20% of households in 2024, according to Statista), with adoption being higher in first-tier Chinese cities but often driven by “tech novelty” rather than necessity.
Similarly, while service robots like robot vacuums have seen sales growth, their household penetration remains limited (around 30% globally, according to Euromonitor International). Wearables, while popular among fitness enthusiasts, are still optional for most consumers.
Qian points to several factors determining whether embodied intelligence products become rigid demands:
Industry Observer Perspective: True Explosion or Hype Bubble?
Qian questions whether 2025 is truly the “year of explosion” for embodied intelligence, despite the hype and industry buzz. She compares the embodied intelligence industry to the smart car industry, seeing it as more of a “combinatorial innovation” than a “disruptive technology innovation.” Smart car breakthroughs are not single-technology leaps but deep hardware-software integration. Similarly, embodied intelligence’s value lies in integrating sensors, AI algorithms, robotics, and automation, not just in one breakthrough technology.
She uses Tesla’s autonomous driving as an example of gradual development through deep learning and hardware, but notes that full autonomous driving adoption faces technical and regulatory hurdles. Embodied intelligence products face similar challenges in technology integration and ecosystem building. Consumer demand is driven by system-level product integration and “engineering innovation” that improves quality of life and work efficiency, not just isolated feature innovations.
This type of innovation, while potentially yielding strong short-term market performance, has a slower commercialization path than “disruptive innovation.” The “explosion year” requires not just tech breakthroughs but also supply chain collaboration.
Qian points out the complexity of the embodied intelligence supply chain, involving AI algorithms, sensors, IoT, robotic arms, transmission systems, and computing platforms. Coordination across these domains is crucial, and breakthroughs in each area are vital. However, the current supply chain faces uncertainties. AI computing and algorithm advancements are foundational, but translating them into products requires precision hardware and sensors. Current smart robots still face user experience issues in multi-tech integration.
From a market perspective, despite large corporate investments, supply chain collaboration is lacking, especially in China, where core technologies rely on foreign suppliers. This creates technological dependence in critical areas like chips and high-end sensors. The rise of domestic chipmakers and supply chain integration are key to embodied intelligence breakthrough. While companies push for development, the “explosion year” label is questionable.
Qian identifies key factors determining if a 2025 “explosion” is possible:
Qian concludes that 2025 is a crucial transition year, but full explosion requires overcoming tech, market, and capital hurdles. She suggests that the current “loud thunder, little rain” scenario, with capital inflow but excessive competition and fragmented supply chains, leads to inefficient competition. Drawing on past industry precedents, she predicts an “elimination wave” in the Chinese embodied intelligence industry in the coming years, with inevitable consolidation.
Management Scholar Perspective: Lab to Industry – How Far to Go?
Qian examines the path of embodied intelligence from lab to industry, drawing parallels to classic innovation diffusion patterns. She uses the technology maturity curve to analyze the current stage.
Revolutionary technologies start with small-scale lab demonstrations, initially adopted by tech enthusiasts and early adopter companies. Early market demand is limited, and technology may not fully meet consumer needs. This is the tech validation phase, characterized by high costs and low market acceptance. Market players, capital, and tech providers focus on tech demonstration and application experiments, with many features far from commercial viability.
Qian argues that embodied intelligence is currently in a typical “over-expectation period” on the technology maturity curve. Market hype outpaces actual application capabilities. Media and capital inflate expectations, exemplified by Unitree’s Spring Festival Gala performance. Public enthusiasm is high, but the technology isn’t ready for mass adoption.
The “valley of disillusionment” is a common phase for innovation. Overly high market expectations encounter real-world challenges. Embodied intelligence will face similar hurdles: immature tech, high costs, lack of industry standards, unclear consumer needs. This phase will involve a brutal “shakeout,” with only commercially viable technologies and companies surviving.
Qian outlines three stages of commercialization:
Insights and Reflections:
Qian concludes by reflecting on past tech revolutions, noting that many hyped innovations didn’t rapidly transform the world but underwent long periods of development. From smart speakers to toys to humanoid robots, embodied intelligence waves face similar challenges: expectation gaps between market hype and technological reality. Like iPhone hype, embodied intelligence is over-promoted, but real-world adoption and meeting consumer needs remain distant goals.
She questions whether the embodied intelligence “explosion year” has arrived or if it’s just a media/capital-driven illusion, potentially ending up like smart speakers – “tech toys” enhancing user experience without fundamentally changing lifestyles.
From a management perspective, Qian argues that commercialization must bridge the tech-market gap. As a complex, composite technology, embodied intelligence relies on hardware, algorithms, AI, and supply chain integration. She asks:
These questions point to whether the market can awaken from “tech illusion” and embrace genuinely valuable intelligent products.
Qian urges industry observers and consumers to rationally assess the market, asking if technology is truly driving industry change or if we’re in another “innovation bubble.”
From an industry perspective, companies must recognize that tech development is gradual, find suitable product positioning, avoid “hype cycles,” and navigate the “valley of disillusionment.” From a management perspective, supply chain innovation and competitive differentiation are crucial for industry success.
Qian’s key insights:
Ultimately, Qian concludes that whether embodied intelligence becomes the next big thing, the final fusion of technology and market will determine success.
So, there you have it, folks. A heated debate raging in China’s tech and investment circles about the future of humanoid robots. Is it a revolution or a bubble? The jury is still out. But one thing is clear: China is betting big on robots, and the world is watching to see if these dreams of walking, talking machines will become a reality, or just another chapter in the wild ride of tech hype and bust. Stay tuned, because this robot story is just getting started.
Explore how soaring cemetery prices in China are challenging ancient burial customs, leading to phenomena…
Uncover the dramatic story of Zhang Xiaoquan, China's iconic 400-year-old knife maker, now battling financial…
Discover Xiao Zhi AI, the open-source Chinese project taking the AI hardware world by storm.…
China's aging population isn't a burden, it's a massive opportunity. Learn how the 'Silver Economy'…
Witness the remarkable resurgence of China's restaurant industry in Spring 2025. This detailed report uncovers…
How Chinese businesses are strategically responding to escalating US tariffs. Explore their innovative strategies: market…