AIToday
RoboticsRobohubPublished: Oct 12, 2026, 19:01 JST

World Humanoid Robot Games 2026: laundry, not the 100m, tests robots

World Humanoid Robot Games 2026: laundry, not the 100m, tests robots

3 Key Points

  1. What happened

    At the World Humanoid Robot Games 2026 in Beijing, Tiangong Ultra ran the 100 metres in 8.64 seconds, faster than Usain Bolt's 9.58-second world record. But Wai-keung Fung of Cardiff Metropolitan University says no robot system can yet reliably handle the whole laundry process.

  2. Why it matters

    Laundry forces robots to cope with clothes that are inside out, twisted together, soft and constantly changing shape, so a robot that can sort and fold a messy basket and recover from mistakes may be a better sign of general-purpose robot intelligence than a fast run.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Context & Analysis

The World Humanoid Robot Games 2026 in Beijing put humanoid robots through sprinting, boxing, football and tasks inspired by factories, restaurants and emergency situations. Tiangong Ultra's 8.64-second 100 metres, quicker than Usain Bolt's 9.58-second world record, is the kind of result that makes the technology visible. Wai-keung Fung of Cardiff Metropolitan University argues that visibility is not the same as readiness.

Running fast is difficult, but the task itself is fairly clear: the robot knows where to go, the track is predictable and the finish line does not change shape. Laundry is the opposite. A shirt may be inside out, one sleeve may be hidden underneath another garment, and a pair of trousers may be twisted around a towel. Fabric is soft and constantly changing shape, so picking up one corner of a T-shirt moves the rest of it.

The difficulty is compounded because a physical skill does not transfer between machines the way an app can be installed on any device. A specialist laundry robot with rigid arms, overhead cameras and fabric-specific grippers shares neither the body nor the balance constraints of a humanoid with cameras in its head and five-fingered hands. Research projects such as Open X-Embodiment are exploring this problem by pooling experience from many different robots; the project has brought together more than one million real robot trajectories from 22 different robot types, with the hope that robots can learn useful patterns from one another rather than starting from zero every time.

FAQ
How fast did Tiangong Ultra run the 100 metres?
Tiangong Ultra completed the 100 metres in 8.64 seconds at the World Humanoid Robot Games 2026 in Beijing. That is quicker than Usain Bolt's 9.58-second world record.
Why is folding laundry so hard for robots?
A shirt may be inside out, a sleeve may be hidden, and trousers may be twisted around a towel, and fabric is soft and constantly changing shape. Humans adjust their grip automatically, but for a robot each of those small actions is a problem to solve.
Can one robot simply copy another robot's skill?
No, because physical skills do not work like an app that can be installed on any machine. Robots differ in arms, cameras, hands, joints, balance and sensors, so the same movement can produce a very different result; some knowledge can still transfer.

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleAlphaProof Nexus solves nine of 353 open Erdős problems