Six Months, $14M: China’s Industrial AI Robots Surge

Six months after founding, Sunrising AI has raised over $14 million to bring embodied AI to automotive assembly lines.

Sunrising AI, founded in April 2025 by Zhang Tao, a former Alibaba AutoNavi technical director, and Li Shengbo, a Tsinghua University professor, has quietly assembled a high-conviction capital base. The latest round, led by IDG Capital and Oriental Fortune Capital with participation from EFORT Intelligent Equipment, 01VC, DaTai Capital, and L2F Ventures, brings total funding to more than RMB 100 million in under half a year. The money is being funneled into core technology, productization, and commercialization as the team scales beyond proof-of-concept deployments.

At the heart of Sunrising’s pitch is a high-smoothness neural network architecture designed for industrial manipulation on wheels—an emphasis on smooth, precise action in structured yet complex environments. The company relies on reinforcement learning rather than traditional imitation learning to improve precision and success rates on the factory floor. That choice matters: reinforcement learning potentially enables robots to adapt to variations in final-assembly tasks, but it also raises questions about data efficiency and safety in live production settings.

To address the chronic shortage of real-world data in factory environments, Sunrising leans heavily on simulation. It has built a GOPS platform to modularize model development, accelerating experimentation and scalable deployment across different production lines. In practice, that means developers can swap in new manipulation modules or task policies without reworking entire systems—a critical capability if the company hopes to deploy across dozens of plants with different line layouts.

Sunrising AI has already moved from theory to field tests, with multiple automakers as partners and initial proof-of-concept deployments in production environments. The roadmap is ambitious: enter at least 10 automotive manufacturers and deploy more than 1,000 robots within three years. If achieved, the scale would position China’s industrial robotics sector as a more self-sufficient spine for automakers, reducing reliance on imported automation solutions during a period when supply chain resilience is under intense scrutiny.

From a practitioner’s perspective, the strategy highlights several tradeoffs and risk factors for global manufacturers eyeing China as a robotics supplier or partner. First, the shift from imitation to reinforcement learning signals a desire for more adaptable automation, but it heightens the demand for robust validation and safety controls before large-scale rollout. Second, the GOPS platform’s modular approach should lower integration costs and time-to-value, yet it also raises questions about IP sharing, data governance, and compatibility with existing plant floor systems. Third, the emphasis on simulation data underscores a broader trend in China’s robotics ecosystem: the pursuit of data-rich development environments to accelerate capability without waiting for every real-world cycle. That, in turn, can compress lead times for vendors but may require careful cross-checking of simulation-to-real transfer performance.

Supply chain disclosures reveal a Chinese market where capital is flowing into AI-driven automation with a tilt toward domestic platform ecosystems. The work being done by Sunrising aligns with a broader push to embed advanced AI in manufacturing processes, potentially reshaping how automakers source robot cells, teleoperate maintenance, and standardize line configurations. Yet execution remains the wild card. Real-world deployments across multiple OEMs will test the platform’s true scalability, the reliability of reinforcement-learning policies in high-stakes assembly tasks, and the ability to maintain cross-plant consistency as production rhythms shift.

If Sunrising proves out, global manufacturers should watch not only the robots themselves but the tooling around them—the GOPS platform, the data pipelines, and the ability to translate simulation gains into durable on-floor performance. A Chinese robotics ecosystem that can deliver hundreds to thousands of turnkey robotic cells at scale would redraw vendor expectations and potentially tilt procurement toward domestic AI-enabled automation in automotive lines.

Sources

Inverse Matrix Technology Raises Over $10 Million to Advance World Models

SUNRISING AI Raises Over $14 Million Within Six Months of Founding