Two Funding Rounds Secured in 40 Days, Embodied Data Practitioner Enters the Arena
Investment news outlet learned on August 12th that ScaleForce, an Embodied AI data infrastructure company, has raised two funding rounds within 40 days from investors including top domestic embodied AI industry players, Hengxu Capital, and Capital Today.
Established less than three months ago, ScaleForce has already secured a multi-million-yuan data order from its lighthouse customer, TaShi Zhihang. It has also established collaborations with multiple world model companies, leading embodied intelligence hardware manufacturers, and industry solution providers. Founder Guo Jiangliang believes that the competition in embodied intelligence has entered its second phase, which focuses on data and intelligence. ScaleForce aims to enable data to flow through the entire lifecycle of physical AI, fostering infinite intelligence within the physical world.
The immense popularity of embodied intelligence contrasts sharply with a severe shortage of high-quality, real-world interactive data needed for training. Unlike large language models that can scrape training material from the internet, embodied AI requires multimodal physical interaction data involving vision, touch, joint trajectories, object mechanics, and environmental temporal alignment, which must be collected in the real world. The industry faces challenges such as uncontrollable data quality, insufficient scale, and poor data generalization across different robotic platforms.
ScaleForce addresses these issues with its core product, the MatrixOS physical AI operating system. It features the ADA (Action-Data Alignment) data generalization engine, which reportedly increases cross-platform task success rates from 65% to 92%. Its GDP (Global-Deep-Proactive) data quality engine employs various analytical methods to optimize data collection, achieving an 80% usable conversion rate. The company has built a global, scalable data production network with over 95% automation, utilizing human-centric collection paradigms and sub-millisecond multi-sensor synchronization kits.
MatrixOS functions as a comprehensive data "production system" for embodied intelligence, connecting the physical world for data acquisition, processing, and management, then delivering it to models and applications. ScaleForce is building what it claims will be the world's largest and lowest-cost high-quality data production and distribution network.
The company has secured its position in the emerging embodied AI data industry chain by providing tailored scene data to hardware makers, real physical interaction data to world model companies, and task-specific data closed loops to industry clients. Its partnership network includes world model firms like ZhiZai WuJie, Huawei's Ascend ecosystem, and academic institutions like Peking University and Beihang University. International expansion is also underway, starting in Southeast Asia.
The founding team brings deep industry experience. Founder Guo Jiangliang was a founding member of Baidu Intelligent Cloud and former Technology VP of AInnovation. Chief Scientist Alex, a Peking University PhD, has held core roles at Meta AI and Huawei Noah's Ark Lab. The team's combined expertise in technology, product, and commercialization forms the foundation for ScaleForce's rapid progress.
With 2026 considered the "first year of embodied data," and 2027-2029 viewed as a critical window for the scaled commercialization of humanoid robots, the demand for robust data infrastructure is poised for significant growth. As companies like Unitree approach IPOs, the next phase of competition in embodied intelligence will heavily rely on data, marking a crucial time window for the embodied data sector.
marsbit08/12 02:46
