“Data Is the Battleground for Physical AI” | Understanding
https://www.youtube.com/watch?v=4LPWH2hCIUkSKAI Intelligence CEO Jaecheol Lee Appears on Business & Industry YouTube Channel Understanding: Presenting a Vision for Data Infrastructure in the Physical AI Era Through Synthetic Data-Based Robot Learning
SKAI Intelligence CEO Jaecheol Lee appeared on the business and industry YouTube channel Understanding to discuss the key challenges facing the Physical AI industry and the critical role of data technologies in overcoming them.
Lee introduced Physical AI as the next generation of artificial intelligence that enables robots and autonomous systems to perceive, understand, reason about, and act within the physical world.
Unlike generative AI, which primarily learns from digital data such as text and images, Physical AI must understand a wide range of real-world physical characteristics, including object geometry, material textures, friction, reflectivity, weight, assembly tolerances, and constantly changing operating environments.
However, acquiring such data in real industrial settings is both costly and time-consuming, and collecting sufficient data for rare, hazardous, or exceptional scenarios is often impractical. Lee identified this data bottleneck as one of the biggest barriers to the commercialization of Physical AI.
SKAI Intelligence addresses this challenge through Digital Twin and Synthetic Data technologies. By accurately recreating factories and manufacturing environments in virtual space, robots can first learn from diverse operating conditions and exceptional scenarios before being deployed in real-world environments.
The company further improves AI performance through a Real–Sim–Real data loop, continuously feeding real-world operational data back into the digital twin to refine AI models and establish an industrial AI training framework optimized for production environments.
This approach safely reproduces hazardous or difficult-to-capture scenarios while enabling robots to repeatedly learn from a wide variety of conditions before deployment, making it one of the core technologies driving the commercialization of Physical AI.
Leveraging NVIDIA Omniverse-based digital twin environments and high-fidelity synthetic data generation technologies, SKAI Intelligence provides advanced AI training infrastructure for industrial applications.
Its technological capabilities have been recognized globally through its participation as a Main Contributor to the Physical AI white paper jointly published by ABB Robotics and NVIDIA.
The interview also explored the growth potential of the Physical AI industry and the opportunities it presents for Asian manufacturing. Lee referenced SoftBank Group Chairman Masayoshi Son's projection that Physical AI could generate economic value equivalent to 20% of global GDP by 2030.
Considering the significant share of global manufacturing represented by Korea, China, and Japan, Lee suggested that Asia is well positioned to become a global hub for the Physical AI industry.
He further emphasized that the next three years will be a critical window for Korean companies to secure global leadership by combining strong manufacturing capabilities with advanced data technologies.
According to Lee, future competitiveness will depend not only on developing AI models but also on how quickly companies can accumulate industrial data and build the data infrastructure necessary to utilize it effectively.
SKAI Intelligence supports the entire industrial data lifecycle—from collecting and storing distributed manufacturing data to structuring it for AI and generating synthetic data where real-world data is insufficient—enabling AI and robot training at scale.
Going forward, the company plans to accelerate AI transformation across industries by strengthening its position as a leading data infrastructure provider for the Physical AI era through Digital Twin and Synthetic Data technologies.
SKAI Intelligence's vision and technologies for the Physical AI era can be viewed in CEO Jaecheol Lee's interview on the business and industry YouTube channel Understanding.
Related Video
Jaecheol Lee, CEO of SKAI Intelligence | Understanding
[YouTube Link]