Why investors are betting billions on X Square Robot's embodied AI vision

Chinese start-up X Square Robot is building machines designed not for a single task, but for everyday life. Whether helping an elderly resident with daily tasks, working as a cleaner or sorting parcels in a busy warehouse, the company's ambition is the same: create robots that can understand the world around them, make decisions and complete a wide range of physical tasks without needing to be programmed for every single action.

That vision has captured investors' attention.

Just over two years after its founding in December 2023, X Square Robot has completed four consecutive funding rounds, culminating in a Series C that values the company at more than $2.8 billion. The backing comes from an influential mix of tech giants and venture capital firms, making it the only embodied AI (the kind that is integrated into physical systems like humanoid robots, robotic arms and autonomous vehicles) company in China to have secured lead-round investments at different stages from Alibaba, ByteDance, Meituan and Xiaomi. HongShan and IDG are also among its investors.

The fresh funding will be used to deepen research into embodied AI, strengthen its robotics platform and expand commercial deployments as the company pursues its long-term goal of building robots that can operate confidently in the physical world.

Unlike conventional robots that follow fixed instructions, X Square Robot's machines are designed to learn from experience. They see their surroundings, understand what is happening, decide what to do next and then carry out the task. More importantly, they continue learning every time they interact with people and places.

"Since day one, X Square Robot has focused on in-house development of foundation models, pursuing a challenging but necessary path," says Wang Qian, Founder and CEO of X Square Robot. "Today, our investments in embodied AI models, scalable, model-driven high-quality data pipeline system, and real-world deployment are beginning to deliver clear results."

Driving X Square Robot’s general-purpose embodied AI robots are its WALL series of foundation models. Rather than treating vision, language and movement as separate problems, the models combine them into a single system, enabling robots to understand instructions, recognise objects, plan their next move and react to changes around them.

The latest version, WALL-B, introduced in April, is another step towards that goal. By learning from real-world interactions instead of relying on carefully scripted instructions, the model allows robots to become more capable the longer they are deployed.

But building smarter robots is only half the challenge. Teaching them requires enormous amounts of real-world experience.

That is why X Square Robot has developed its own data pipeline, collecting information every time its robots perform tasks in homes, factories or warehouses. The data is cleaned, organised and fed back into the AI models, helping the robots improve continuously. Every delivery completed, every room cleaned and every object picked up becomes another lesson.

However, its biggest classroom is not a laboratory.

In Shenzhen and Beijing, X Square Robot has partnered with 58.com to launch an AI-powered cleaning service where robots work alongside professional cleaners in occupied homes. Instead of carefully staged demonstrations, the machines are navigating cluttered living spaces and responding to the unpredictability of daily life.

Under Its X Family Member Program, robots live with families for up to a month, exposing the robots to thousands of everyday situations that simply cannot be recreated in a testing facility.

X Square Robot is also working with a high-end elderly care provider to introduce robots that can deliver items, assist with cleaning, communicate with residents, carry out routine patrols and issue early-warning alerts.

Working with Jinbei Auto, X Square Robot is bringing its embodied AI systems on to automotive production lines, where robots perform tasks requiring consistent accuracy and precision while adapting to dynamic manufacturing environments.

In logistics, the company has teamed up with one of the leading logistics companies in China to introduce its robots into parcel feeding and sorting operations.

Each of these deployments feeds back into the company's larger ambition. Every successful task generates fresh data, making the AI models stronger and allowing the next generation of robots to become more capable.

"As AI moves beyond digital experiences into the physical world, progress will depend on close integration between models, data and robotics," says Wang. "We're building that foundation so embodied AI can become part of everyday life."