SK hynix is accelerating manufacturing innovation by positioning its “Autonomous Fab” strategy as a key response to surging demand in the artificial intelligence (AI) era.
Speaking at the NVIDIA GTC 2026 panel discussion titled “Building the Future of Manufacturing,” Do Seung-yong, Vice President and Head of Digital Transformation (DT) at SK hynix, highlighted the dual challenge of expanding production capacity while managing increasing manufacturing complexity.
“As AI-driven semiconductor demand rises rapidly, production capacity cannot scale at the same pace due to structural limitations,” Do explained. To address this, SK hynix is expanding capacity both in Korea and globally, including its investment in Indiana, U.S.
However, he emphasized that building new fabs and stabilizing mass production takes time, making it essential to improve efficiency in existing production lines in parallel.
Manufacturing complexity is also intensifying. The growing share of high-value, customized products such as high-bandwidth memory (HBM) has increased operational difficulty, making it more challenging to balance quality, cost, and speed. As a result, traditional experience- and rule-based automation is reaching its limits.
To overcome these challenges, SK hynix is developing an Autonomous Fab targeted for 2030. This next-generation manufacturing model enables factories to learn from data and make decisions autonomously, significantly reducing the time from design to mass production.
The company’s strategy is built on three pillars: Operational AI, Physical AI, and Digital Twin.
Operational AI serves as the “brain” of the factory, transforming engineers’ expertise into data-driven decision-making. This has reduced processing time for tasks such as equipment maintenance and defect analysis by more than 50%.
Physical AI functions as the execution system, advancing automation into areas traditionally dependent on human labor. By integrating AI into wafer transport systems such as overhead hoist transport (OHT), and deploying vision-based robots and autonomous mobile robots (AMRs), SK hynix expects to improve logistics efficiency and safety while reducing parts inventory by approximately 30%.
Digital Twin provides a simulation environment, recreating real fabs in a virtual space using NVIDIA Omniverse. This allows the company to validate production flows, material movement, and factory layouts in advance, while enabling simulation, AI training, and operational optimization without interrupting actual production.
By integrating these three pillars, SK hynix aims to build a faster and more flexible next-generation manufacturing system and strengthen its global competitiveness in the semiconductor industry.
