The Humanoid Race Japan Started

In 2000, Honda’s ASIMO walked onto the world stage and made Japan the undisputed home of humanoid robots. Twenty-five years later, the headlines belong to Tesla, Figure and Chinese challengers such as Unitree. Japan did not lose its engineering. It lost the race to turn that engineering into a product.

The Head Start Nobody Matched

Honda began its walking-robot research in 1986 and unveiled the P2 in 1996, the first self-contained bipedal humanoid. ASIMO followed in 2000 and spent two decades as the face of Japanese technology, greeting visitors, conducting orchestras and climbing stairs on stage. Toyota, Kawasaki and Japan’s national research institute AIST built their own humanoids, and Japan’s industrial robot makers dominated factory floors worldwide.

Honda retired ASIMO from public appearances in 2022. The machines were remarkable, but they were built as showcases of mechanical control, not as products with a customer and a price.

Why the Lead Evaporated

1. Research, not a business model

ASIMO was funded as a corporate research and brand program. No team was asked to hit a unit cost, a payload target or a pilot-customer deadline. The new entrants started from the opposite end: Tesla’s Optimus, announced in 2021, and Figure’s robots are explicitly pitched at factory and warehouse labor, with a cost target attached.

2. Hardware-first versus AI-first

Japanese humanoids were choreographed: precise, pre-programmed and fragile outside a controlled setting. Today’s competitors treat the body as a platform for learned behavior, training on large amounts of simulation and demonstration data and borrowing directly from the large-model boom. Japan’s strengths in control theory mattered less once the bottleneck moved to software and data.

3. Cost and scale

China’s Unitree launched its G1 humanoid in 2024 at a list price of roughly US$16,000, a fraction of what research humanoids had cost. Chinese firms benefit from dense domestic supply chains for motors, batteries and sensors, and from government support that treats humanoids as a strategic industry.

4. Capital and risk appetite

US humanoid start-ups have raised funding rounds in the hundreds of millions of dollars. Japan’s large corporations tend to fund robotics from internal budgets, and its venture ecosystem has been far smaller. A project like ASIMO is easy to wind down when priorities shift.

What Japan Still Owns

The humanoid story is not only about the finished robot. Every humanoid needs precision reducers, actuators, bearings, sensors and motors, and this is where Japanese firms remain hard to replace.

Japan also remains the center of industrial robotics, accounting for a very large share of the world’s industrial robot output. The question is whether that installed base and know-how can be repackaged for general-purpose machines.

What It Means for Overseas Partners

• Component buyers — Western and Chinese humanoid makers already depend on Japanese reducers and precision parts. Direct supply agreements are a realistic entry point.
• Investors — the listed component suppliers offer exposure to humanoid growth without having to pick the winning robot.
• Deployment partners — Japan’s labor shortage makes it a natural testbed for service and logistics robots, and a place where foreign start-ups can find pilot customers.

The Takeaway

Japan did not fall behind in robotics engineering. It fell behind in treating the humanoid as a product, funding it like one and pairing it with modern AI. The next chapter is likely to be written by Japan as a supplier and a testbed more than as the maker of the headline robot, and that is where partners can find an opening today.