Key Takeaways

  • Samsung Group and SK Group are expected to announce up to 2,000 trillion won in long-term investments tied to South Korea’s industrial strategy
  • The initiative reinforces South Korea’s push to expand semiconductor capacity and AI data-center infrastructure
  • Global research points to rising AI compute demand, sustained chip-equipment spending, and increasing data-center power needs

South Korea’s industrial landscape is shifting again, and this time the scale is unusual even by the country’s own standards. Samsung Group and SK Group are preparing to unveil as much as 2,000 trillion won of investment commitments over the next decade as part of President Lee Jae Myung’s strategic roadmap. The report, originating from Korea Economic Daily, surfaced just ahead of a scheduled presentation at the presidential office where Samsung and SK leadership are expected to lay out their spending plans.

These numbers arrive in a global environment already primed for rapid AI expansion. According to Gartner, global generative AI spending is projected to reach $200 billion by 2028, up from an estimated $19.4 billion in 2024. This trajectory has become a primary target for many governments, and South Korea is positioning itself as a foundational supplier of the compute, memory, and data-center infrastructure needed to support that rise.

The investment scope is large enough that even earlier reporting seems modest in hindsight. Maeil Business Newspaper had previously said Samsung alone might exceed 1,000 trillion won in spending. Now the combined figure across Samsung Group and SK Group is expected to reach as much as 2,000 trillion won over ten years. These are the kinds of totals that shift global manufacturing maps.

Officials say the initiative will align with President Lee’s Three Mega Projects for the Great Leap Forward. Industry, science, climate, and land ministries will outline government policy steps before Samsung and SK take the stage. The timing is deliberate. President Lee had previewed that very unusual figures would be announced on June 29, hinting at a coordinated strategy that spans fabrication, packaging, and AI infrastructure.

At the facility level, Samsung Electronics and SK Hynix are expected to build between four and five semiconductor fabs each in the Gwangju area. Samsung plans to add chip packaging plants in South Chungcheong province, while SK Hynix will expand NAND production in North Chungcheong province. It is a geographically distributed buildout across regions already familiar with large-scale manufacturing.

The country’s emphasis on AI data centers fits into broader trends. IDC has estimated that global AI spending will surpass $500 billion by 2027, and this aligns with South Korea’s allocation of 550 trillion won for AI data-center investment by 2029. That is part of a wider 1,350 trillion won national plan that also includes 800 trillion won earmarked for semiconductor fabs. These figures were highlighted in recent government disclosures and reiterated by industry observers.

To support these facilities, the infrastructure requires substantial power. The International Energy Agency has noted that data centers, AI, and crypto could sharply increase electricity demand through 2026. This underscores why South Korea’s plan emphasizes regional power grid upgrades and reinforced energy delivery. Without it, the fabs and data centers simply cannot operate at required levels.

On the manufacturing side, SEMI has projected that global fab equipment spending will stay elevated as manufacturers scale advanced nodes and memory for AI-heavy workloads. Samsung Electronics and SK Hynix are positioned directly in that stream, which gives the country leverage in an increasingly competitive semiconductor environment. This equipment is central to long-term competitiveness because memory and compute remain the primary bottlenecks for AI training and inference.

While physical infrastructure drives national industrial narratives, technical standards shape how organizations deploy the resulting AI capabilities. ISO and NIST have both released frameworks that companies are adopting as they scale. ISO has formalized ISO/IEC 42001 for AI management systems, and NIST has published its AI Risk Management Framework 1.0. These models are gaining traction precisely because enterprises need ways to operationalize AI responsibly as the infrastructure around them grows.

South Korea’s push is not emerging in isolation. Bloomberg recently reported that Samsung and SK Hynix are central to a reported $880 billion national AI buildout focused on chips, data centers, and broader digital infrastructure. That figure is smaller than the new potential investment totals but aligned conceptually. Nations are treating semiconductors as strategic assets, and South Korea is doubling down on its strengths.

The presentation scheduled for 2 p.m. gives companies room to detail timelines and facility configurations. Key industry focus will remain on whether these fabs lean heavily into advanced nodes, as well as how quickly the AI data-center infrastructure can come online given the energy constraints flagged by multiple analysts.

The broader direction is clear: Samsung Group and SK Group are preparing to execute one of the largest industrial expansion plans announced this decade. The scale reflects the pressure and opportunity in AI markets globally. It also underscores that for South Korea, semiconductor capacity, packaging, memory, and data centers are not just business priorities. They are national strategy.