China's GigaAI Advances World Model Technology With Manufacturing Data Advantage

Gate News message, April 15 — Chinese AI startup GigaAI is rapidly closing the gap with the U.S. in world model technology, a core capability for physical AI, by leveraging vast industrial manufacturing data and government-backed data collection systems. According to reporting from the South China Morning Post on April 14, China has moved beyond the research phase to achieve commercial deployment advantages through access to large-scale, structured data from manufacturing facilities and robot operations.

World models simulate 3D environments and physical laws in virtual space to train physical AI systems such as robots and autonomous vehicles. GigaAI recently secured 1 billion yuan (approximately $190 million) in new funding, followed by an additional round of similar scale within weeks. The company claims its latest model, GigaWorld-1, outperforms Google and Nvidia-affiliated models in visual quality, adherence to physical laws, and 3D accuracy. GigaAI is collaborating with electric vehicle makers including Nio, Xpeng, and BYD on vision-based autonomous driving systems and is reportedly generating annual revenues in the tens of millions of yuan range.

In the U.S., investment competition remains intense. WorldLabs, founded by AI researcher Fei-Fei Li, and AMI Labs, associated with Yann LeCun, each secured approximately $1 billion in funding during the first quarter of 2026. Google DeepMind is working with Waymo to apply world models to autonomous driving, while Tesla is using the technology to train its Optimus humanoid robot. Among Chinese tech giants, Alibaba's Amap mapping service formalized world model research earlier this year, and Tencent released an open-source model capable of generating 3D environments from single images or text.

Industry observers note that while China's manufacturing data advantage is significant, commercialization still requires time to validate. Key challenges include precisely replicating the complex physical variables of the real world and ensuring safe deployment in robot and autonomous vehicle services. Safety and cost efficiency are expected to be the ultimate competitive factors in commercial-stage deployment.