Starting in May 2026, humanoid robots built by Tokyo-based startup Gitai Japan Inc. will load luggage, move cargo, and clean aircraft cabins alongside human ground crews at Tokyo’s Haneda Airport. Japan Airlines announced the deployment in a joint press release on April 28, calling it the first use of humanoid robots in airport ground operations anywhere in Japan.
This is not a weekend showcase. JAL has committed to a three-year operational trial running through 2028, meaning the machines will face peak summer travel, typhoon-season disruptions, and the relentless grind of one of Asia’s busiest airports. The carrier’s executive officer Yoshiteru Suzuki framed the program as a sustained effort to address chronic labor shortages in ground handling, not a technology preview.
What the robots actually do
At a media event at Haneda, journalists watched Gitai’s humanoid units tow baggage carts through terminal spaces and position themselves at aircraft doors under staff supervision. Guardian correspondents on site described the robots as remote-operated or semi-autonomous systems, not fully independent replacements for human workers. Staff remained present to oversee each task, intervening when the machines encountered obstacles or needed repositioning.
The scope laid out in JAL’s bilingual release covers three categories: baggage handling, cargo transport, and cabin cleaning. The Japanese-language version uses the term “実証実験” (demonstration experiment) and confirms the “国内初” (first in Japan) designation. Both language versions align on timeline, location, and task scope, leaving little room for translation ambiguity.
Gitai, founded in 2016, has built its reputation on robots designed for extreme environments. The company previously tested robotic arms aboard the International Space Station in partnership with JAXA and NASA, proving the hardware could perform precise manipulation tasks in conditions far less forgiving than an airport ramp. That track record likely factored into JAL’s decision to partner with a firm whose machines have already operated under real-world pressure, not just in controlled labs.
Why Haneda, and why now
Japan’s airport ground-handling sector has been bleeding workers for years. The country’s Ministry of Land, Infrastructure, Transport and Tourism has repeatedly flagged staffing shortfalls at major airports, a problem that worsened after the pandemic gutted the aviation workforce and recovery in international passenger traffic outpaced hiring. Haneda, which handled over 87 million passengers in fiscal year 2024, sits at the center of that pressure.
Ground-handling work is physically punishing. Ramp agents lift bags weighing up to 23 kilograms thousands of times per shift, often in extreme heat or cold, on tight turnaround schedules. The job’s combination of low pay, high injury rates, and irregular hours makes recruitment difficult even in a labor market as tight as Japan’s. For JAL, robots that can absorb the most repetitive and strenuous tasks offer a way to keep operations running without competing for workers who increasingly have better options elsewhere.
The three-year window is what separates this from the short-lived robotics demos that have popped up at airports from Munich to Singapore. A trial lasting days or weeks lets a company generate favorable footage. A trial lasting through 2028 forces the technology to prove it can handle seasonal surges, equipment failures, and the thousand small complications that define daily ramp operations. If the robots cannot keep pace, three years is long enough for that failure to become visible.
The $15,400 question
The per-unit figure of roughly $15,400 (approximately ¥2.3 million) has circulated in secondary reporting and aligns with Gitai’s previously disclosed list pricing for its general-purpose humanoid platform. JAL’s own press materials do not publish a per-unit acquisition or operating cost, so the number should be understood as a manufacturer’s list price rather than an audited program cost. Total spending, including maintenance, software updates, on-site support staff, and integration with existing ground-handling systems, will almost certainly be higher.
For context, the annual cost of employing a single full-time ground handler in Japan, including benefits and overtime, typically exceeds ¥4 million. If the robots prove durable enough to operate across multiple shifts without proportional maintenance costs, the economics could tilt in their favor within the trial period. But that math depends entirely on performance data that JAL has not yet disclosed.
What is still missing
Several important gaps remain. JAL has not published quantitative performance benchmarks: no target for bags handled per hour, no acceptable error rate, no speed comparison with human crews. Without those numbers, outside observers have no baseline against which to judge whether the experiment is succeeding or quietly stalling.
No statements from ground-handling staff or labor unions have appeared in any of the available materials. The introduction of robots into daily ramp workflows raises practical questions about job displacement, retraining, and safety protocols on active aprons where aircraft, fuel trucks, and personnel share tight spaces. The absence of worker voices does not mean opposition exists, but it does mean the human side of this transition has not been addressed publicly.
Safety and liability frameworks are also unclear. The documents do not spell out how responsibility would be allocated if a robot collides with ground equipment, mishandles baggage, or causes injury. In tightly regulated aviation environments, those details typically require coordination between airlines, ground-handling subsidiaries, insurers, and airport authorities. Until such frameworks are visible, it is difficult to assess whether the trial is operating under bespoke waivers or existing rules adapted to new technology.
Post-2028 plans are similarly unaddressed. JAL has said nothing about whether a successful trial would lead to permanent deployment at Haneda, expansion to Narita or Kansai International, or licensing arrangements with other carriers. That silence is typical for early-stage programs, but it leaves the long-term strategic picture open.
What to watch as Haneda’s robots log their first real shifts
The verified facts are narrow but solid: real robots, built by a company with a spaceflight pedigree, will work real flights at one of the world’s busiest airports for three years. JAL has put its name and operational schedule behind the program, not just a press release.
The questions that will determine whether Haneda becomes a model for airports worldwide or a cautionary tale are all about execution. How many bags can the robots move per shift? How often do they break down? Do they reduce delays and damage incidents, or create new bottlenecks? And critically, how do human crews adapt to working alongside machines that share their ramp space?
Those answers will not come from another press event. They will come from the data JAL collects over thousands of flights, across seasons, through the kind of sustained operational stress that no demo can replicate. The trial starts in May 2026. The verdict will take years.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
