On June 16, 2026, Chinese tech outlet 93913 reported that China’s Ministry of Industry and Information Technology and the State-owned Assets Supervision and Administration Commission have launched a 2026 special action program for humanoid robots and embodied intelligence. The joint notice sets goals for large-scale real‑world deployments across industrial, service and emergency-response scenarios and targets “ten‑thousand‑unit” level rollouts by the end of 2026.
This article aggregates reporting from 2 news sources. The TL;DR is AI-generated from original reporting. Race to AGI's analysis provides editorial context on implications for AGI development.
Beijing’s new special action program makes explicit what was already implied by its industrial policies: China wants embodied intelligence and humanoid robots to be a volume business by the end of this decade. Moving from lab demos to “real‑scene training” in factories, logistics hubs, retail, healthcare and emergency response massively increases the amount of physical‑world data that can be used to train embodied AI models. That is exactly the ingredient current large language models lack: dense, long‑horizon interaction with the messy real world.
By mandating that provinces and central SOEs identify dozens of standardized scenarios and form consortia that pair end‑users with robot OEMs and algorithm providers, the government is effectively underwriting early demand for humanoid deployments. If even a fraction of the planned pilots reach “robot‑as‑a‑service” commercial scale, Chinese firms will accumulate a trove of embodied experience that can feed next‑generation world models and control stacks.
For the AGI race, this initiative strengthens the argument that some of the most interesting advances may come from the intersection of language models, vision and robotics, not text‑only systems. It also raises the competitive bar for Western companies that have treated humanoids as a long‑term bet rather than a near‑term industrial policy priority.


