[AI Autonomous Manufacturing, Opening the Future] "From Mass Production to the Era of AI-driven Ultra-Precision"

Chaejsik Choi, KAIST Professor: "The Key to Korea's Manufacturing Transformation Is AI"
"High Value-Added Production Must Be Achieved Through AI"
A Survival Formula for Small and Medium-Sized Enterprises Beyond Large Corporations

Professor JaeSik Choi and Professor JaeCheol Kim from KAIST AI Graduate School are explaining a case where AI was used to improve performance in manufacturing sites. Photo by JongMin Baek, Tech Specialist

Korea's manufacturing industry stands at a crossroads of a 'major transformation.' A massive tectonic shift has begun, rendering the old formulas for success ineffective. The focus is shifting from cost reduction through mass production to maximizing efficiency through 'ultra-precision.' It is now considered a crucial task to lay the groundwork for utilizing artificial intelligence (AI) in manufacturing processes, not only for large corporations but also for small and medium-sized enterprises, in ways that require less human intervention.

Chaejsik Choi, a professor at the KAIST AI Graduate School and CEO of the AI automation solutions company Inizy, stated in a recent interview with Asia Economy that the only way for Korea's entire manufacturing sector to survive is through 'ultra-precision' powered by AI. He advised that a true paradigm shift in manufacturing can only occur if the changes brought about by AI innovation in manufacturing extend beyond large corporations to also include small and medium-sized enterprises.

◆Industrial paradigm shift from 'bulk' to 'ultra-precision'= He offered a clear-eyed analysis of the reality facing the entire Korean industry, saying, "The era of 'bulk' production, competing on quantity as we did in the past, is over." He emphasized the need to transform the industry's structure to focus on producing high-precision, high-value-added products that cannot be made by just anyone, such as high-bandwidth memory (HBM), and to sell these at premium prices.

Professor Chaejsik Choi of KAIST is explaining the use of AI in production sites in an interview with Asia Economy. Photo by Jongmin Baek, Tech Specialist

Professor Choi cited cases in which Inizy, the company he founded, contributed to improving diesel quality at SK Energy and reducing quality variation in cold-rolled steel sheets at KG Steel. These are representative examples of how 'ultra-precision' can be achieved through AI. He explained that ultra-precision, which cannot be attained with engineering know-how alone, can be reached through AI learning and processes. For example, AI is used to detect changes in the physical properties of automotive steel plates or steel pipes during processing.

He noted that even small differences made possible by AI can increase added value. Professor Choi said, "The price of bulk ammonia can differ by up to 100 times compared to high-quality products. AI is necessary to create such differences."

Implementing manufacturing AI can also prevent 'technology discontinuity.' Professor Choi pointed out, "The retirement of a single 'artisan' in a small or medium-sized company that lacks a systematic system can immediately lead to the loss of the company's core technology." AI can accumulate and systematize the artisans' tactile sense and know-how as data, serving as a 'digital apprenticeship system' to preserve and transfer a company's most valuable assets.

◆The front line of change: challenges for small and medium-sized enterprises= Professor Choi explained that the wave of massive change presents an even greater challenge and opportunity for small and medium-sized enterprises, which form the backbone of our industry. He said, "Upgrading the entire industry is impossible without innovation among the many small and medium-sized enterprises that make up the supply chain," explaining why AI is even more essential for these companies.

High initial costs and a lack of experts are significant barriers for small and medium-sized enterprises looking to adopt AI. Professor Choi suggested 'collaborative ecosystems' and 'user-friendly AI' as solutions. He said, "There are limits to what individual companies can achieve on their own," and proposed a 'collective approach' in which the government or associations lead the creation of data platforms that small and medium-sized enterprises can jointly utilize, or provide affordable subscription-based AI solutions. He emphasized, "AI should not be the main player; skilled factory workers should be the main players," and added, "AI solutions that can be used as easily as smartphone applications, without complex coding, need to be distributed in order to take root on the factory floor."