In May 2026, Japan Airlines began deploying humanoid robots on the tarmac at Tokyo’s Haneda Airport to handle baggage and clean aircraft cabins. The robots are roughly the size of a small adult, move on two legs, and cost around $13,500 to $16,000 each. They are made by Unitree Robotics — a Chinese company founded in 2016.

That last detail is the one worth sitting with. Japan is the country that built ASIMO — Honda’s famous two-legged robot that spent two decades walking around the world as the face of technological ambition. Japan has the deepest humanoid robotics research history of any nation. And yet, when one of its major airlines needed working humanoid robots for a real-world deployment, it turned to a Chinese manufacturer to supply them.

This article explains how that happened, what the JAL trial actually involves, and what the broader competitive picture looks like — fairly, without exaggerating either Japan’s decline or China’s dominance.

What Is Actually Happening at Haneda Airport?

The trial was announced officially by JAL Ground Service Co. and GMO AI & Robotics Corporation on April 27, 2026. It is described as Japan’s first demonstration experiment for humanoid robots at an airport — a phased, multi-year programme rather than a full commercial deployment.

The robots being used are Unitree G1 models. The G1 stands approximately 130cm tall, weighs around 35kg, moves at up to 4.4 mph, and operates for about two hours on a full charge. In the initial phase, the robots are being used to push cargo onto loading belts and assist with baggage handling on the tarmac — physically demanding, repetitive work that puts significant strain on human ground handlers. Cabin cleaning is planned as a later phase task.

The official reason JAL gives for the trial is straightforward: Japan’s aviation sector is facing a genuine labour shortage. Japan’s population declined for a fourteenth consecutive year in 2024 — not a blip but a structural trend that governments and corporations are now being forced to plan around. The working-age cohort currently makes up just 59.6% of the total population, a share projected to shrink by nearly 15 million people over the next two decades. Ground handling requires skilled, physically capable workers. Japan’s population is ageing and shrinking. International tourism is booming — a record 42.7 million visitors arrived in 2025, and Japan is targeting 60 million by 2030. The gap between available workers and operational demand is real and widening.

Watch: Japan Airlines Humanoid Robot Trial at Haneda

The Unitree G1 — What Kind of Robot Is It?

Unitree Robotics is a Chinese company founded in 2016, based in Hangzhou. It started by making quadruped robots — the four-legged kind — and expanded into bipedal humanoids with the H1 in 2023 and the G1 in 2024. The G1 caused significant attention in the robotics world when it launched at a starting price of around $16,000, substantially cheaper than comparable humanoid systems from Western companies.

The G1’s price point is significant. For context, Agility Robotics’ Digit — an American humanoid robot being deployed in Amazon warehouses — costs over $250,000. Boston Dynamics’ Atlas is not commercially available. Tesla’s Optimus is in limited production with no confirmed consumer price. Unitree made a deployable humanoid robot that costs roughly the same as a decent used car.

Japan’s Robotics History — Where It Leads and Where It Doesn’t

To understand the JAL trial clearly, it helps to understand what Japan actually is and is not in robotics — because the picture is more nuanced than headlines suggest.

Here is a statistic that reframes the whole narrative: Japanese manufacturers already account for approximately 70% of the global industrial robotics market, according to Japan’s own Ministry of Economy. That is not a country that has lost its robotics industry — it is a country that dominates one category of it while racing to catch up in another. The distinction matters.

Japan genuinely pioneered humanoid robotics research. Honda began developing bipedal walking robots in 1986 — fourteen years before ASIMO’s public debut in 2000. ASIMO was a landmark engineering achievement: it could walk, run, climb stairs, recognise faces, respond to voices, and interact with people in ways no robot had managed before. Japan’s AIST research institute developed the HRP series of humanoid robots from 2003 onwards, focusing on disaster response and industrial applications. Toyota, Kawasaki, and Fanuc built Japan’s industrial robotics sector into the most sophisticated in the world.

But ASIMO was retired in 2022. Honda said it was halting development to focus on more practical applications — care robots and mobility systems. The decision was pragmatic, not a failure, but it signalled a shift: Japan’s robotics industry moved toward high-precision components, actuators, sensors, and industrial systems rather than full humanoid platforms. Meanwhile, the commercial race for deployable humanoid robots — the kind you can actually send to do a job — accelerated in China and the United States.

Japan’s Actual Plan — The Numbers Behind the Strategy

The JAL trial is not an isolated experiment. It sits inside a much larger government-backed industrial strategy that most Western coverage has not connected to the airport story.

In March 2026, Japan’s Ministry of Economy, Trade and Industry formally presented its revised strategy through 2040, setting a target of capturing more than 30% of the global physical AI market. To understand what that means in money terms — McKinsey’s May 2026 analysis projects the global general-purpose robotics market will grow from less than $1 billion today to approximately $370 billion by 2040, expanding at around 70% annually. Japan is targeting roughly $111 billion of that market.

To back the target, Japan allocated ¥387.3 billion (approximately $2.6 billion) specifically for physical AI and domestic AI infrastructure in fiscal year 2026, within a broader ¥1.23 trillion AI and semiconductor package. In April 2025, a ¥150 billion subsidy for next-generation robotics R&D was also announced. This is not a wish — it is a funded national industrial strategy.

The strategic logic is clear once you see the full picture. Japan already controls approximately 70% of the global industrial robotics market according to its own Ministry of Economy. But industrial robots — the fixed-arm kind that weld car bodies and assemble electronics — are a slow-growth market, projected to expand at only around 3% annually through 2040. General-purpose humanoid robots, by contrast, are the fast-growth category. Japan wants to take its existing manufacturing depth and component expertise and pivot it toward the new category before China and the US lock up that market completely.

What the Humanoids Summit Tokyo Showed

On May 28-29, 2026, the Humanoids Summit held its first Tokyo edition at the Takanawa Convention Center — bringing together robotics companies from Japan, China, the US, and Europe. The event offered a clear picture of where things stand.

Japanese companies including Honda and Kawasaki were present and showed serious engineering. Kawasaki’s Kaleido 9 can lift 30kg loads, learn to operate cleaning tools, and supports remote control through VR headsets. Honda emphasised fine-motor control, durability, and manufacturing precision — areas where Japanese engineering remains genuinely world-class.

But by general consensus, the most attention went to Chinese companies. Booster Robotics and LimX Dynamics demonstrated robots that were capable, affordable, and — crucially — closer to being commercially deployable. High Torque’s Mini Pi Plus started at $5,500. The pattern observers noted was familiar: technology originally developed in Japan and the US, refined by Chinese manufacturers for faster and cheaper mass production.

It is worth noting that Hornyak’s quote represents one expert’s opinion, not an official assessment. TrendForce’s analysis offers a more measured reading: Japan is pursuing a different strategy — focusing on core components rather than full systems — while China and the US race to commercialise complete humanoid platforms. Both strategies have merit; they target different parts of the same supply chain.

Why Humanoid Shape Matters at Airports Specifically

One question worth asking is: why humanoid? Airports already use wheeled robots, conveyor systems, and automated vehicles. Why does JAL need something that walks on two legs?

The official answer from JAL’s press release is practical: airports were designed for humans. The spaces around aircraft, the angles of cargo loading, the layout of cabin interiors — all of it assumes a human-shaped operator. Wheeled robots struggle with stairs, uneven tarmac surfaces, aircraft door thresholds, and the tight spaces around fuselages. A robot with arms, hands, and legs that roughly match human dimensions can theoretically operate in the same physical environment without requiring the airport to be redesigned. That is the theory the trial is testing.

Whether it proves out in practice over the 2026-2028 trial period is an open question. Analysts at Barclays noted that robots still require significant human oversight despite rapid advances. The regulatory picture is also unresolved — Japan’s Ministry of Land, Infrastructure, Transport and Tourism has not yet addressed how aviation safety regulations apply to autonomous humanoid robots operating near aircraft.

Will Robots Take Ground Handler Jobs?

The job displacement question comes up immediately with any robotics story, and it deserves a straight answer rather than reassurance.

One data point that puts the longer-term picture in perspective: a Bank of America report cited by Fortune in April 2026 predicted that people will own more humanoid robots than cars by 2060. That is a projection, not a fact — but it signals where institutional investors see this heading. The near-term reality at Haneda is much more modest.

JAL’s official position is that the robots are intended to complement human workers rather than replace them — specifically to reduce physical burden on ground handlers doing repetitive, heavy work. The labour shortage framing is genuine: Japan genuinely does not have enough ground handlers and the gap is growing. In that specific context, robots filling roles that would otherwise go unfilled is not the same as robots replacing employed workers.

The longer-term picture is harder to call with confidence. If humanoid robots become capable and affordable enough to handle the full range of ground handling tasks, the economic pressure to reduce human headcount will be real regardless of current intentions. That tension exists in every industry where automation is advancing, and the aviation sector is no different.

My Take — Mr Wangdoo

The JAL Haneda trial is genuinely interesting — not because it is a dramatic turning point, but because it is a small, concrete, real-world example of something much larger playing out globally. The labour shortage driving it is not unique to Japan. Most developed economies face similar demographic pressures. What makes Japan’s situation distinctive is that it is arguably the country best placed culturally and technically to embrace humanoid robots, and it is having to import the commercial version of a technology it largely invented.

Whether that matters in the long run depends on where the value in humanoid robotics ultimately sits. If it is in the software and AI — teaching robots to do complex tasks in unstructured environments — then Japan’s hardware precision matters less. If it is in manufacturing scale and component supply, Japan’s position is stronger than the headlines suggest. Probably both will matter, which is why the next few years of deployments like the Haneda trial will be worth watching closely.

What I am not going to do is declare a winner. The trial started six weeks ago and runs until 2028. The robots are performing simple, repetitive tasks in controlled conditions. Anyone claiming to know how this story ends is getting ahead of the evidence.

Mr Wangdoo

Founder and Editor-in-Chief of Wangdoo.com. Independent tech journalist covering AI, EVs, gadgets, and emerging tech.