Digital twins now sit at the center of SK Telecom’s manufacturing effort after the company applied the technology to semiconductor facilities operated by SK hynix using NVIDIA Omniverse libraries. The work focuses on large production settings that contain complex systems and extensive operational data. SK Telecom said the technology has been adjusted to function in manufacturing environments that require high levels of coordination and scale.

The company said it completed a proof of concept for semiconductor factory deployment last year and intends to introduce commercial use in stages. That timeline aligns with SK hynix’s target of building an autonomous factory model by 2030. The virtual system recreates production spaces and equipment, allowing manufacturers to test process changes and equipment placement before physical adjustments take place.

SK Telecom also developed what it calls agentic modeling technology using the NVIDIA Agent Toolkit to handle manufacturing information used inside virtual factory environments. The system processes varied inputs, including equipment layouts and physical site structures, while supporting tasks linked to data preparation, scene adjustments, and operational efficiency during development and use.

At the same time, the company said it strengthened its platform through NVIDIA Omniverse libraries to improve the handling of large OpenUSD-based three-dimensional scenes. The changes aim to improve loading performance, execution speed, and hardware resource use. SK Telecom said this approach supports dependable and expandable virtual environments even in manufacturing sites that contain substantial amounts of data, including semiconductor production facilities.

SK Telecom said its work with NVIDIA confirmed that manufacturing systems can move beyond visualization alone and function as platforms capable of interpreting and improving large-scale three-dimensional production data. The company also said it plans to widen its role in physical AI technology across industries, while expanding business efforts aimed at enterprise and public-sector customers through its broader AI capabilities.

📊 What This Means (Our Analysis)

The development highlights how virtual factory environments can become more practical when companies improve the way industrial data is handled and tested before real-world changes happen. By tying validation work to phased commercialization and an autonomous manufacturing roadmap, the effort points to a more structured path for industrial AI systems that depend on accurate operational modeling.

The broader importance rests in the move from visual representation toward systems designed to process and improve manufacturing information at scale. SK Telecom’s emphasis on stability, performance, and larger industrial use suggests growing attention to physical AI systems that can support demanding production settings without separating infrastructure, modeling, and operational workflows.

📌 Our Take: The next phase will likely be defined by how virtual factory systems become embedded into day-to-day industrial operations.