– Professor Young-Jae Jang of KAIST, Jeonbuk Province, Jeonbuk National University, Sungkyunkwan University
– Nurturing Jeonbuk as a global hub for the physical AI industry
[Herald Economy = Reporter Bon-Hyuk Goo] KAIST announced on the 28th that it has jointly won the ‘Physical AI Core Technology Demonstration’ pilot project, organized by the Ministry of Science and ICT, in collaboration with Jeollabuk-do, Jeonbuk National University, and Sungkyunkwan University. Additionally, this consortium plans to participate in a full-scale demonstration project expected to be worth a total of 1 trillion won in the future.
In this project, KAIST led the research planning on the theme of ‘Collaborative Intelligence Physical AI,’ while Jeonbuk National University and Jeollabuk-do plan to jointly conduct research based on this and create a collaborative intelligence physical AI industrial ecosystem in Jeollabuk-do. The pilot project will start on September 1 this year and will be conducted over the next five years, with the intention of establishing Jeonbuk as a global physical AI hub.
KAIST is responsible for the development of foundational research technology, the creation of a research environment through the establishment of a test bed, and further industrial expansion. The project’s overall supervisor at KAIST, Professor Young-Jae Jang from the Department of Industrial and Systems Engineering, has been leading research on collaborative intelligence physical AI since 2016.
‘Physical AI’ refers to the latest artificial intelligence technology where AI understands concepts of time and space, allowing physical devices such as robots, autonomous vehicles, and factory automation devices to perform tasks without human intervention.
Specifically, collaborative intelligence physical AI involves technology where numerous robots and automation equipment work together to achieve goals in a factory environment and is recognized as a key foundation for implementing ‘dark factories’ (unmanned factories) in the semiconductor, secondary battery, and automobile manufacturing industries.
Furthermore, unlike existing manufacturing AI, it does not require vast amounts of past data. It adapts quickly to constantly changing manufacturing environments through real-time simulation-based learning, overcoming the limitations of dependency on past data and is considered the next-generation technology.
Currently, the global AI industry is led by large language models (LLM) that simulate language intelligence. However, physical AI must surpass language intelligence to include spatial intelligence and virtual environment learning, necessitating the organic combination of hardware and software such as robots, sensors, and motors. As a manufacturing powerhouse, Korea is in a favorable position to build this ecosystem and has a chance to lead global competition.
KAIST actually topped MIT and Amazon to win with its collaborative intelligence physical AI case study at the world’s largest industrial engineering conference, INFORMS, in April 2025. This is seen as proof of the global competitiveness of physical AI technology.
Professor Young-Jae Jang stated, “Winning this large-scale national project is a result of the recognition of KAIST’s accumulated collaborative intelligence physical AI research capabilities both domestically and internationally over the past decade,” and added, “This will be a turning point for Korea’s manufacturing industry to establish a globally leading ‘physical AI manufacturing innovation model.’”
KAIST President Kwang-Hyung Lee emphasized, “KAIST is playing a leading role in the practical industrialization of national strategic technology beyond academic research,” and “Taking this achievement as a foundation, we will collaborate with Jeonbuk National University and Jeollabuk-do to develop Korea into a leading global physical AI innovation hub.”