Eleven years ago, when Motovis was founded, we held one unshakable belief: AI is changing the future of humanity, and it will be embedded into all things in our world.
At the time, autonomous driving was still in its early stages of exploration, and "artificial intelligence" remained a distant concept for most people. But we had already seen the direction — embed AI into the real world, and let intelligence drive physical entities. The automobile is the ideal scenario in which artificial intelligence meets physical entities and reaches massive-scale application. This was the starting point of Motovis: Embedded AI for Automobiles — a leader in embedded AI for autonomous driving.
Over the past eleven years, with intelligent driving as our focus, we have paved the mass-production path for embedding artificial intelligence into physical entities, delivering cumulative shipments of more than 3 million units. We have built full-stack capabilities spanning algorithm and model design — compute deployment optimization — data closed loop, and become one of the few leading technology companies in the automotive industry that independently owns the full stack of algorithms from perception to planning and control. From the earliest CNNs to today's VLA (Vision-Language-Action) models and world models, our algorithm team has always stood at the very frontier of the industry.
Looking back on these eleven years, what we have done goes beyond embedding AI algorithms into cars. The algorithmic capabilities we have built are turning the automobile into a physical entity of embodied intelligence. We empower intelligent vehicles to perceive the outside world in real time, to possess a physical body through sensors and actuators, to interact with the external environment in real time, and to evolve proactively through that interaction. Multi-modal fusion perception systems integrate heterogeneous data from cameras, radar and other sensors in real time, forming a three-dimensional understanding of the world. Through VLA, our systems can grasp physical causality and common sense, enabling intelligent driving to truly "read" the real world.
We let AI exist and evolve within the automobile as an intelligent agent — not merely giving the car an intelligent brain, but making it a physical entity with a body, able to interact with its environment and close the loop from perception to action. From Embedded AI to Embodied AI — this is more than the evolution of Motovis's technology roadmap; it reveals the vast application scenarios that Motovis's algorithms can reach.
We believe that Physical AI must address the core challenge of the algorithm–compute–data trinity. What Motovis has accumulated over eleven years is precisely this: core algorithms fully validated in large-scale commercial deployment, and massive volumes of data naturally accumulated through mass production.
On the core strategic frontier of Physical AI, we already possess four core algorithmic capabilities, each validated across millions of real-world scenarios:
4D Spatial Perception: Real-time perception of multi-sensor heterogeneous data, precisely reading the spatial logic of the physical world.
World Model and Dynamic Prediction: Anticipating the motion of surrounding objects and projecting how physical scenes will evolve.
Automotive-Grade, Highly Reliable On-Device Deployment: Landing intelligent algorithms on physical hardware with high efficiency, low latency and high safety.
A Complete Data Closed-Loop Iteration System: Enabling algorithms to evolve continuously in real-world scenarios while dramatically reducing the cost of data acquisition.
These four core algorithmic capabilities are the most essential and scarcest hard-core assets of the Physical AI era, and the foundation on which we build our core competitiveness for the future. Our algorithms will also be deployed in more Physical AI scenarios, empowering every industry.
No matter how cutting-edge a technology may be, if it cannot be put to work, it remains little more than an exhibit in a laboratory.
Motovis has always held fast to two things: the advancement of AI algorithms, and commercial deployment. Neither is dispensable — without algorithmic breakthroughs, there is no product competitiveness; without commercial validation, technology has no closed loop.
Today, Motovis's intelligent driving solutions have been validated in enormous volumes of real-road scenarios. We have shipped more than 3 million units, serving many of China's leading automakers — Chery, GAC, BAIC, Changan, FAW, Dongfeng, Shaanxi Automobile and Sinotruk — as well as two of the world's top three automakers by global sales volume. Intelligent vehicles equipped with Motovis's products are driving not only in every province of China, but also across Europe, the Middle East, Southeast Asia, Australia and New Zealand, Africa and South America, making intelligent driving accessible to users around the world. This is not merely a commercial scorecard; it is a large-scale, real-world validation of our full-stack technology system. Algorithmic breakthroughs empower product deployment, and mass-production validation feeds back into technology iteration — this is Motovis's proven path of AI algorithm evolution.
Eleven years marks both a moment of reflection and the start of a new expedition.
We are standing at a brand-new inflection point: business scale continues to expand, technology barriers continue to strengthen, and commercial deployment continues to accelerate. Embodied AI is not only empowering automobiles — it is stepping into the tangible physical world, letting intelligence perceive all things, understand the laws that govern them, and drive physical entities.
This is the tremendous opportunity the era has given us, and the historic mission of Motovis.
Meanwhile, Motovis's IPO process is advancing rapidly. Listing is not the end, but a brand-new beginning — it will inject stronger capital momentum into technology R&D, talent development and market expansion. This is not the goal of a few, but the shared journey of everyone at Motovis.
The achievements of the past belong to every member of Motovis; the glories of the future must be created by our own hands. We thank every partner for your perseverance, dedication and creativity, and every investor and customer for your trust and companionship.
With one heart and one direction, let us turn technological ideals into industrial reality, empower Physical AI across every industry, and witness the next decade together.
Let us forge ahead — together.
Dr. Yu Zhenghua, CEO of Motovis
August 2026
```