By ByteBot
China’s National Development and Reform Commission dropped a rare admission on November 27, 2025: a “humanoid robot-shaped bubble” is forming. Over 150 companies—more than half of them startups or firms pivoting from unrelated industries—have rushed into humanoid robotics, churning out “highly similar” models despite having almost no proven use cases. The warning carries weight because it comes from the same government that designated humanoid robotics as a strategic priority through 2030. It also carries weight because China has been here before.
If the government is warning about a bubble, you’re probably already watching it pop.
Every Bubble Signal Is Flashing Red
The humanoid robot market exhibits every classic bubble indicator. You’ve got 150+ companies entering an immature market where over half are startups or industry-switchers chasing hype. Li Chao, the NDRC spokesperson, admits humanoid robots “have yet to fully mature in terms of technological pathways, business models, and application scenarios.” That’s bureaucrat-speak for “we have no idea if this makes money.”
The production numbers tell the story: only 500+ humanoid robots work in actual production across BYD, Geely, FAW-Volkswagen, and Foxconn. That’s 500 units deployed versus 150+ companies producing them. Meanwhile, you’ve got impressive demos everywhere—AgiBot’s 66-mile marathon walk, Unitree’s dancing robots stealing the Spring Festival Gala, EngineAI’s robots learning complex choreography from brief demonstrations. Great PR. Unclear revenue.
When the ratio of companies to deployed products is 150 to 500, and the government that promoted the industry starts warning about “highly similar” products flooding the market, consolidation isn’t coming. It’s here.
China’s Bubble Playbook: Rinse and Repeat
This playbook is familiar. China’s bikesharing bubble (2017-2018) saw companies like Mobike, Ofo, and Bluegoo invest huge capital regardless of actual demand, producing massive overcapacity. The result: “bike graveyards” of damaged, unused bikes piling up, and multiple bankruptcies. Bluegoo collapsed in late 2017, signaling the broader crash.
The semiconductor sector (2020-2025) followed the same script. Chinese production capacity grew four times faster than global demand between 2015 and 2023. Chinese firms went from 19% to 33% of global wafer production capacity. The outcome: underutilized capacity, brutal price wars, squeezed margins, and deflation. RAND describes the pattern as “industrial involution” where provinces compete to build the biggest semiconductor cluster or AI hub, creating redundant capacity and inefficient allocation.
The pattern: government designates sector as strategic → capital floods in → companies rush to market → overcapacity develops → government issues warning (too late) → consolidation and failures follow.
Watch for the “Bluegoo moment” in humanoid robotics—the first high-profile failure that confirms the consolidation phase has begun.
Why Government Warnings Mark the Peak, Not the Beginning
Here’s what makes this warning particularly telling: it’s coming from the same NDRC that promoted humanoid robotics as one of six strategic growth industries through 2030. They’re simultaneously promoting and warning about the sector. That contradiction doesn’t prevent bubbles—it accelerates them.
Government warnings about tech bubbles historically arrive at or near the peak of hype cycles, not early enough to prevent losses. When Li Chao says authorities will “improve market entry and exit mechanisms, promote consolidation,” translate that to: expect failures and mergers. When he talks about supporting “essential research areas,” understand that others will lose funding.
For tech professionals and investors, the lesson is clear: don’t wait for official warnings to spot bubbles. By the time governments acknowledge bubble dynamics, it’s too late to sidestep losses. Learn to recognize the indicators yourself—excessive market entry, industry-switchers, product homogenization, unproven use cases.
The Broader AI Bubble: Agents, Coding, and GenAI
Swap “humanoid robots” for “AI agents” or “coding assistants,” and this story plays out globally. Gartner’s 2025 AI Hype Cycle places AI agents squarely at the “Peak of Inflated Expectations.” Sam Altman, OpenAI’s CEO, stated in 2025 that “an AI bubble is ongoing.” Ray Dalio said current AI investment levels mirror the dot-com bubble. Big tech is projected to spend $400 billion on AI in 2025, but revenue comes mainly from coding assistants and subscriptions—the same disconnect between investment scale and proven business models.
Daron Acemoglu, MIT economist and 2024 Nobel Prize winner, puts it bluntly: “These models are being hyped up, and we’re investing more than we should.” He’s talking about AI broadly, but the critique applies equally to humanoid robotics. Impressive demos, massive investment, unclear paths to profitability. If revenue doesn’t justify multi-trillion dollar valuations, “the financial foundation of Silicon Valley may be at risk.”
The humanoid robot bubble isn’t isolated. It’s one manifestation of broader AI hype cycles in 2025. Tech professionals should apply the same recognition framework across sectors: look for excessive investment without proven use cases, industry-switchers flooding in, and product homogenization.
Key Takeaways
- Don’t wait for official warnings to recognize bubbles—spot them early through indicators like excessive market entry (100+ companies in 50% from unrelated sectors), and product homogenization.
- China’s pattern repeats: government backing plus capital flood equals overcapacity and consolidation. Bikesharing had “bike graveyards.” Semiconductors had price wars and deflation. Humanoid robots will likely follow the same arc.
- The humanoid robot bubble mirrors global AI hype. AI agents, coding assistants, and GenAI investments exhibit similar dynamics: massive capital chasing unproven use cases, impressive demos that don’t translate to revenue, and growing warnings from industry leaders.
- Watch for the first major failure—the “Bluegoo moment”—to confirm when consolidation begins. NDRC’s talk of “promoting consolidation” signals that phase is approaching. Smaller firms will merge, exit, or fail. Funding will dry up for undifferentiated products.
- Apply the bubble recognition framework across any sector: if 100+ companies chase the same unproven use case while producing similar products, you’re in bubble territory. Recognize it before regulators do.
