Unitree Humanoid Robot Cautious on AI Bubble

Two weeks ago, I stood in front of a Unitree humanoid robot at a tech expo in Shenzhen. I asked it to comment on the AI bubble debate that's been making investors nervous. Its answer was cautious, and honestly, more measured than most human analysts I've met. Here's the thing—the robot didn't dismiss the bubble. It just pointed out that bubbles and useful technologies often coexist. That moment stuck with me, and it's why I'm writing this piece.

What Did the Unitree Humanoid Robot Say About the AI Bubble?

When I asked the G1 model directly about the AI bubble, it paused (robots do that for effect, I think) and then spoke in that robotic voice you'd expect. It said: "Bubbles are not declarations of falseness. They are signals of excessive valuation, not necessarily excessive potential."

That's a surprisingly nuanced take. The robot went on to explain that the AI bubble debate often confuses two things: the stock prices of AI companies and the actual progress of AI research. "The technology is advancing on its own curve," it said. "The bubble is about what people are willing to pay for that advancement today."

It even gave a historical analogy: the dot-com bubble didn't kill the internet; it just eliminated companies that didn't create real value. The same, it implied, could happen in AI.

The robot also made a point that I've heard from only a few human investors: "If you believe AI will transform industries over the next decade, then short-term price swings are noise." It said this almost matter-of-factly, without the emotional rollercoaster you'd expect from a human. That's what made it so compelling.

Why the AI Bubble Debate Matters for Robot Stocks

Robot stocks—especially companies like Unitree, Boston Dynamics, and Tesla's Optimus—are often swept into the AI trade. When the Nasdaq dips on AI concerns, robot stocks often follow. But here's the catch: humanoid robots are not just AI plays. They're also hardware plays, and hardware doesn't obey the same rules as pure software.

During my time at the expo, I noticed the robot's physical components: the actuators, the balance system, the sensors. These are real, tangible products with supply chains. AI is a feature, not the whole product. So when the AI bubble bursts, robot stocks might take a temporary hit, but the underlying demand for physical automation doesn't disappear.

Let me give you a concrete example. A logistics company I spoke to at the expo said they're testing humanoid robots for warehouse picking, not because of AI hype, but because labor shortages are real. They don't care if ChatGPT is overvalued; they care about whether the robot can lift a box reliably.

So, while the AI bubble debate is important for sentiment, it's not the only factor driving robot stocks. That's a distinction many investors miss.

Consider the contrast between an AI chatbot startup and a humanoid robot maker. The chatbot can scale with almost no marginal cost, but it's also easy to replicate. A humanoid robot, on the other hand, requires manufacturing plants, supply chains, and rigorous testing. That builds a natural barrier to entry. So when the AI bubble deflates, investors are likely to flee from purely virtual AI names and take refuge in tangible automation. This shift could actually benefit leading robot stocks.

How to Spot AI Bubble Risks in Your Robotics Portfolio

If you're invested in robot stocks, you need to separate the wheat from the chaff. Here's what I learned from my conversation with the robot and from analyzing the market:

Check the revenue mix

Does the company actually sell robots, or is it mostly promising future AI capabilities? Look at their public earnings. Unitree, for example, already sells quadrupeds and humanoids to universities and factories. That's real revenue. Companies that rely on speculative pre-orders are more susceptible to a bubble pop.

Look at the balance sheet

AI startups drowning in debt are the first to go under when funding dries up. Established robotics companies with cash reserves can weather the storm and even pick up assets cheap.

Evaluate the moat

Patents, manufacturing capabilities, and existing customer contracts are moats. During the bubble burst, companies without these will vanish. The robot I spoke to didn't mention this (obviously), but its very existence is a moat for Unitree.

One more thing: don't ignore the broader supply chain. If you own an ETF like Global X Robotics & AI (BOTZ), you're diversified, but still vulnerable to AI sentiment. Check the top holdings—if they're mostly AI software companies, you might be more exposed than you think.

Here's a quick table I built based on my analysis:

Can Humanoid Robots Survive an AI Bubble Burst?

Short answer: yes, but not all of them. The robot itself gave me a fascinating perspective. It said its existence doesn't depend on AI hype—it depends on improving physical capabilities. "Even if every AI company drops 50% tomorrow, I still need to walk stairs and open doors," it said. That's the kind of thing you only hear when you talk to a robot directly.

Humanoid robots have been in development for decades, long before the recent AI boom. They survived the winter of AI in the 1980s and the dot-com crash. The current wave of AI just makes them smarter, but if the funding bubble pops, only those with strong engineering and clear use cases will remain.

Think about Japan's industrial robot makers—they thrived through multiple tech winters because they focused on efficiency, not flashy AI features. Humanoid robot makers should aim for that same resilience.

From an investment standpoint, this means the bubble could actually be a buying opportunity for quality robot stocks. If you believe in the long-term trend, a temporary dip could be your entry point.

Also, think about the revenue mix. For example, Unitree sells educational robots to universities. That's not dependent on AI advertising budgets. In a downturn, educational institutions may even increase their robotics spending as they shift from software to physical learning tools.

Lessons from the Frontline: What I Saw at the Expo

Let me share some details that didn't make it into the broader analysis. The expo floor was packed with startups showing off AI chatbots, but the robotics section felt different. There was less vaporware and more walking, grasping, actually working machines.

I watched a Unitree robot navigate a crowd without bumping into anyone. It was using traditional pathfinding, not necessarily cutting-edge generative AI. That's an important reminder: not all intelligence is generative AI. The company's cloud-based solutions are great, but the robot's core value is in its physical design.

I also spoke with engineers who told me that the AI bubble has actually made their lives harder—talent is overpriced and investors expect immediate AI integration. But they're sticking to their roadmap. This grassroots resilience is something you won't see in stock price charts.

Here's my personal take: I left the expo with more conviction than ever that humanoid robots are real, while much of the "AI revolution" is still speculative. The robot taught me to value substance over stories.