UBTech has signed a three-year strategic agreement with Tianyi Telecom Terminal, the wholly owned device subsidiary of China Telecom, with targets that read less like a distribution deal and more like an industry plan. Over three years the two companies intend to push the home companion robot category past one million units, and bionic and wheeled humanoids past one hundred thousand units.
What the carrier brings
Tianyi Telecom Terminal exists to run terminals and terminal ecosystem partnerships, and it sits on the only nationwide retail and service network in China that already sells hardware to households at scale. Under the agreement it will wrap China Telecom’s own capabilities, including AI, quantum, satellite and security, into the robot as a product feature set, and then use its channel to push volume. The mechanics named in the agreement are specific: centralised procurement of robot categories, rental services, bundled AI entitlements sold with the device, and experience zones inside retail outlets.
The company describes itself as the implementing vehicle for the group’s cloud-network and AI strategy, and for its token-based business model, at the terminal layer. That phrase matters. Selling a robot with an AI subscription attached converts a one-off hardware margin into recurring service revenue, which is the same logic that turned broadband into smart speakers and set-top boxes in earlier carrier cycles.
What UBTech brings
UBTech supplies the hardware and the model stack, which it groups into four clusters: an embodied brain, a humanoid cerebellum for motion, a high-performance body, and swarm intelligence for multi-robot coordination. Its ten-thousand-unit class smart factory came online in the second half of this year and several product lines are in volume delivery, which is the precondition for a million-unit target being anything other than a press release.
For UBTech the strategic value is the channel rather than the order book. Humanoid and companion robots have a distribution problem that is harder than the engineering problem: consumers will not buy a category they have never seen operating, and specialist robot retail does not exist. A carrier with thousands of outlets, installed broadband households and a billing relationship solves that in one move.
The number that should be read carefully
One hundred thousand humanoids in three years is a large claim. China’s total humanoid output this year is measured in tens of thousands of units, most of them going to industrial pilots, research labs and exhibition halls. Home companion robots are a cheaper and more mature category, which is why the million-unit target sits there and the humanoid target is an order of magnitude smaller.
What is new is the shape of the bet. For three years the Chinese humanoid story has been told through factory deployments and government procurement. A state carrier committing retail shelf space, subsidised bundles and national after-sales service moves the argument to consumer demand, where the test is not whether the robot works but whether a household renews the subscription in year two.
That is also where the risk sits. Carrier channels are excellent at pushing hardware into homes and poor at sustaining products that fail there. If the companion robot turns out to be a smart speaker with legs, the installed base becomes the problem rather than the asset.
Editor’s note: This is an adapted translation of the original Gasgoo report. It has been trimmed and restructured for readability for an international business audience. The full original (in Chinese) is at https://www.gasgoo.com/robot/news/.