Chinese humanoid robots have posted running times that no human has ever matched.
At the World Humanoid Robot Games in Beijing, a robot built by Honor covered 100 metres in 9.39 seconds, according to the Telegraph, as reported by Techmeme. A separate Chinese machine, called Tiangong Ultra, ran 400 metres in 38.16 seconds — nearly five seconds faster than the winning time at the 2016 Rio Olympics.
The Telegraph notes the hardware is not exactly elegant: Tiangong Ultra has stumpy arms. These are purpose-built machines optimised for a track, not general-purpose helpers, and a record on a straightaway is a narrow test of what a robot can actually do.
Still, the timing is notable. The sprint records land alongside a financial milestone: China's Mech-Mind Robotics is set to open the order book for a US$300 million IPO, according to a report from AI Insider. That is investor money flowing into Chinese robotics at the same moment the sector is generating headline-grabbing demonstrations.
Not everyone in the region is convinced the spectacle is the point. In an editorial, the South China Morning Post argues that as China's humanoid robots make strides, practical use is the next frontier — a pointed reminder that stadium stunts and factory-floor usefulness are different problems.
That tension is the real story. A robot that runs fast is an engineering achievement; a robot that reliably does unglamorous work in warehouses, hospitals, or homes is a business. China is currently producing plenty of the first and asking capital markets to bet on the second.
Why it matters: the combination of record-setting demos and a $300 million public offering suggests China's robotics push is moving from research showcase to funded industry — with the question of what these machines are actually for still unanswered.