Proposes Four Major Policies at the National Assembly Debate on the 23rd
Byeng Dong - Youn, CEO of onepredict, is presenting strategies for manufacturing AX polarization and the diffusion of AI Native Factories at the 'Policy Debate on Manufacturing AI Transformation and On-Site Diffusion' held at the National Assembly Members' Office Building on the 23rd.
"Manufacturing AI policies must now go beyond supporting simple technology adoption and implementation for individual companies, and move toward providing public infrastructure that encompasses education, computing and data, and on-site verification."
Byeng Dong - Youn, CEO of the manufacturing AI specialist onepredict, made this statement on the 23rd at the 'Policy Debate on Manufacturing AI Transformation and On-Site Diffusion' held at the National Assembly, suggesting that to strengthen the competitiveness of domestic manufacturing, the focus of manufacturing AI (AX) policies must be boldly shifted from existing research and development (R&D) and one-off verification to 'actual on-site operation and diffusion'. This debate was co-hosted by National Assembly members Kim Won-i and Choi Hyung-du, and attended by officials from the National Assembly, government, industry, and academia.
Through a presentation titled 'The Problem of Polarization and Policy Proposals for Manufacturing AX Transformation', CEO Youn diagnosed that domestic manufacturing faces a complex crisis including a declining working-age population, a shortage of skilled labor, the restructuring of global supply chains, and the pursuit of China's manufacturing sector. He explained that since existing equipment investments or simple process automation alone are difficult to simultaneously solve various problems in manufacturing sites such as productivity, quality, safety, and supply chains, the entire operation method of manufacturing plants must be transformed to be AI-centric.
In particular, he pointed out that even if there is an abundance of data at manufacturing sites, there is a lack of refined data that can be utilized for AI training, and data by process and equipment is disconnected from each other. He analyzed that the AX gap between companies is widening, compounded by the 'volatilization of tacit knowledge' where the experience and judgment criteria of skilled workers disappear without being left as data, a shortage of AI professionals and computing resources, unclear problem definition, and difficulties in verifying the return on investment (ROI).
CEO Youn warned, "Manufacturing AI is not a universal technology that all companies can immediately utilize just because it is developed and supplied," and "The gap between some companies that possess data, GPUs, professional personnel, and verification environments, and those that do not, can soon lead to the polarization of competitiveness for the entire South Korean manufacturing industry."
He presented the 'AI Native Factory' as a next-generation manufacturing operation model to solve these problems. The AI Native Factory goes beyond applying different AI solutions to individual processes, and is a future-oriented factory that connects manufacturing data, foundation models, task-specific AI agents, and equipment control systems into a single operational structure. He explained that AI agents in each area, such as quality, production, maintenance, safety, energy, and supply chain, exchange information in real-time, and higher-level AI agents coordinate decision-making and execution considering the goals of the entire factory, which can develop into an autonomous manufacturing environment where the factory senses data, infers situations, and autonomously performs necessary actions.
As measures to prevent and resolve manufacturing AX polarization, CEO Youn proposed four major policies: providing AX education infrastructure, providing public GPU support for AX computing and data infrastructure, digitizing tacit knowledge to assetize the on-site know-how of skilled workers, and providing AX verification infrastructure that supports on-site verification and operational diffusion. The intent is that the government should not stop at supporting the introduction of technology by individual companies, but should build public infrastructure that manufacturing companies can utilize jointly.
Specifically, he emphasized 'AX education infrastructure' that establishes upskilling and reskilling courses for incumbent workers by job category such as production engineering, quality control, and equipment operation, and builds a regular consulting system centered on regional Techno Parks; 'public GPU support' that builds joint GPU centers within industrial complexes and provides subscription-based vouchers; 'digitalization of tacit knowledge' that standardizes on-site know-how and unstructured data into AI training data; and 'AX verification and operation infrastructure' that moves away from being PoC-centered and supports maintenance and operation costs for multiple years.
CEO Youn emphasized, "Manufacturing AI competitiveness is not determined merely by how far ahead a few leading companies go," and "The AX transformation of the entire South Korean manufacturing industry is possible only when small and medium-sized enterprises can also access professional personnel, computing resources, manufacturing data, and verification environments, and continuously utilize verified AI in actual factory operations."
Meanwhile, onepredict is a manufacturing AI specialist company that supports decision-making across all aspects of factory operations, including production, quality, and maintenance, based on manufacturing data, and is promoting the implementation of AI Native Factories based on an AI factory operating system.
...
...
...
...
Byeng Dong - Youn, CEO of onepredict, is presenting strategies for manufacturing AX polarization and the diffusion of AI Native Factories at the 'Policy Debate on Manufacturing AI Transformation and On-Site Diffusion' held at the National Assembly Members' Office Building on the 23rd.
"Manufacturing AI policies must now go beyond supporting simple technology adoption and implementation for individual companies, and move toward providing public infrastructure that encompasses education, computing and data, and on-site verification."
Byeng Dong - Youn, CEO of the manufacturing AI specialist onepredict, made this statement on the 23rd at the 'Policy Debate on Manufacturing AI Transformation and On-Site Diffusion' held at the National Assembly, suggesting that to strengthen the competitiveness of domestic manufacturing, the focus of manufacturing AI (AX) policies must be boldly shifted from existing research and development (R&D) and one-off verification to 'actual on-site operation and diffusion'. This debate was co-hosted by National Assembly members Kim Won-i and Choi Hyung-du, and attended by officials from the National Assembly, government, industry, and academia.
Through a presentation titled 'The Problem of Polarization and Policy Proposals for Manufacturing AX Transformation', CEO Youn diagnosed that domestic manufacturing faces a complex crisis including a declining working-age population, a shortage of skilled labor, the restructuring of global supply chains, and the pursuit of China's manufacturing sector. He explained that since existing equipment investments or simple process automation alone are difficult to simultaneously solve various problems in manufacturing sites such as productivity, quality, safety, and supply chains, the entire operation method of manufacturing plants must be transformed to be AI-centric.
In particular, he pointed out that even if there is an abundance of data at manufacturing sites, there is a lack of refined data that can be utilized for AI training, and data by process and equipment is disconnected from each other. He analyzed that the AX gap between companies is widening, compounded by the 'volatilization of tacit knowledge' where the experience and judgment criteria of skilled workers disappear without being left as data, a shortage of AI professionals and computing resources, unclear problem definition, and difficulties in verifying the return on investment (ROI).
CEO Youn warned, "Manufacturing AI is not a universal technology that all companies can immediately utilize just because it is developed and supplied," and "The gap between some companies that possess data, GPUs, professional personnel, and verification environments, and those that do not, can soon lead to the polarization of competitiveness for the entire South Korean manufacturing industry."
He presented the 'AI Native Factory' as a next-generation manufacturing operation model to solve these problems. The AI Native Factory goes beyond applying different AI solutions to individual processes, and is a future-oriented factory that connects manufacturing data, foundation models, task-specific AI agents, and equipment control systems into a single operational structure. He explained that AI agents in each area, such as quality, production, maintenance, safety, energy, and supply chain, exchange information in real-time, and higher-level AI agents coordinate decision-making and execution considering the goals of the entire factory, which can develop into an autonomous manufacturing environment where the factory senses data, infers situations, and autonomously performs necessary actions.
As measures to prevent and resolve manufacturing AX polarization, CEO Youn proposed four major policies: providing AX education infrastructure, providing public GPU support for AX computing and data infrastructure, digitizing tacit knowledge to assetize the on-site know-how of skilled workers, and providing AX verification infrastructure that supports on-site verification and operational diffusion. The intent is that the government should not stop at supporting the introduction of technology by individual companies, but should build public infrastructure that manufacturing companies can utilize jointly.
Specifically, he emphasized 'AX education infrastructure' that establishes upskilling and reskilling courses for incumbent workers by job category such as production engineering, quality control, and equipment operation, and builds a regular consulting system centered on regional Techno Parks; 'public GPU support' that builds joint GPU centers within industrial complexes and provides subscription-based vouchers; 'digitalization of tacit knowledge' that standardizes on-site know-how and unstructured data into AI training data; and 'AX verification and operation infrastructure' that moves away from being PoC-centered and supports maintenance and operation costs for multiple years.
CEO Youn emphasized, "Manufacturing AI competitiveness is not determined merely by how far ahead a few leading companies go," and "The AX transformation of the entire South Korean manufacturing industry is possible only when small and medium-sized enterprises can also access professional personnel, computing resources, manufacturing data, and verification environments, and continuously utilize verified AI in actual factory operations."
Meanwhile, onepredict is a manufacturing AI specialist company that supports decision-making across all aspects of factory operations, including production, quality, and maintenance, based on manufacturing data, and is promoting the implementation of AI Native Factories based on an AI factory operating system.