Samsung Pledges $648 Billion to Transform South Korea Into an AI Superpower

On June 28, 2026, Samsung Group chairman Lee Jae-yong stood before South Korean President Lee Jae-myung at a government briefing in Seoul and announced the single largest corporate investment commitment in the country’s history: 1,000 trillion won, roughly $648 billion, to be deployed across South Korea over the next decade.

The number is almost too large to contextualize. It is greater than the combined GDP of about 140 countries. It dwarfs the total US semiconductor CHIPS Act funding by nearly a factor of ten. And it will be directed almost entirely at the infrastructure and component layers that the global AI industry cannot exist without — chip factories, AI data centers, next-generation batteries, and the specialized substrates that sit between silicon and system.

The Plan, Broken Down

Samsung’s 1,000 trillion won commitment is not a single allocation but a coordinated pledge across multiple Samsung Group entities, each targeting a different layer of the technology stack.

Samsung Electronics is the largest contributor, with its semiconductor division committing to a nationwide expansion of chip production facilities. Approximately 60 trillion won will flow to the company’s established Yongin complex, already one of the largest semiconductor manufacturing hubs in the world. But the expansion is explicitly structured to distribute economic activity beyond the capital region. New facilities will be built in Chungcheong, Yeongnam, and Incheon — a geographic spread that reflects both President Lee Jae-myung’s regional development priorities and the practical need for redundant manufacturing capacity in an era of geopolitical supply chain risk.

Samsung Display has pledged 100 trillion won over the decade. This commitment arrives as display technology becomes increasingly entangled with AI hardware — from the high-refresh-rate panels in AI-accelerated edge devices to the specialized screens used in AI workstations and server room monitoring infrastructure.

Samsung SDI, the group’s battery arm, is expanding its Cheonan manufacturing plant for next-generation battery production. The investment addresses a growing constraint in the AI hardware ecosystem: the energy storage systems that power both edge AI devices and the uninterruptible power supplies that data centers require.

Samsung Electro-Mechanics will boost production of high-value substrates at its Sejong facility. This is a less-discussed but strategically critical investment. Advanced package substrates — the interconnect layers that physically link chips to circuit boards — have become a major bottleneck in AI accelerator production. NVIDIA’s B200 GPUs, which power the current generation of large-scale training clusters, require substrate technology that only a handful of manufacturers worldwide can produce at scale. Samsung’s Sejong expansion directly targets that constraint.

More Than 350 Trillion Won for AI Infrastructure

The single largest slice of the 1,000 trillion won commitment is earmarked for AI infrastructure, primarily data centers. Over 350 trillion won — roughly $227 billion at current exchange rates — will flow into the data center ecosystem across South Korea over the next ten years.

This figure places Samsung’s data center ambitions in the same league as the hyperscaler build-outs announced by Microsoft, Google, and Meta in the US. The difference is that Samsung’s investment creates domestic South Korean capacity rather than expanding offshore operations — a priority that President Lee Jae-myung has made central to his economic agenda since taking office.

The data center commitment also addresses a structural weakness that the AI boom has exposed in the Korean economy. Despite being home to the world’s largest memory chip manufacturers — Samsung Electronics and SK Hynix together control approximately 75% of the global HBM market — South Korea has lagged in building the large-scale inference and training infrastructure that could let Korean companies run AI workloads domestically rather than routing them through US-based cloud providers.

Strategic Timing

The announcement came three days after Chairman Lee Jae-yong met privately with President Lee at the presidential office on June 25 to align on the final scope of the plan. The sequencing was deliberate: the administration wanted to formally associate the investment commitment with the president’s industrial policy agenda, and Samsung wanted government coordination on regulatory approvals and infrastructure permitting that will be required across five provinces.

The geopolitical backdrop makes the timing even more significant. South Korea has spent the past two years navigating an extraordinarily difficult position in the US-China semiconductor conflict. As a major supplier to Chinese chip customers and a critical partner for US AI companies, Samsung has faced pressure from both sides. The 1,000 trillion won announcement effectively declares that Samsung’s primary strategic commitment is to South Korea itself — building domestic capacity that cannot be held hostage by trade restrictions imposed by any external power.

That calculus has become more urgent following the tightening of US export controls on advanced semiconductors in early 2026, which created new uncertainty about Samsung’s ability to serve Chinese customers with leading-edge products.

The AI Memory Angle

No discussion of Samsung’s investment ambitions is complete without addressing HBM. High-bandwidth memory has emerged as the defining constraint in AI accelerator performance, and Samsung has spent much of 2025 and early 2026 fighting to reclaim the dominant market position it briefly lost to SK Hynix.

Part of the 1,000 trillion won envelope will go toward expanding Samsung’s HBM4 and HBM4E production capacity. These next-generation memory architectures — with higher bandwidth, lower power consumption, and improved thermal performance — are what NVIDIA, AMD, and Google’s TPU teams will require for the GPU and accelerator generations planned for 2027 and beyond.

Samsung’s technical roadmap calls for sampling HBM4 to strategic customers in late 2026 and ramping volume production in 2027. The Yongin expansion investment is directly linked to ensuring that roadmap can be executed at the volumes the AI hardware market now demands.

Regional Development as Industrial Policy

The geographic distribution of Samsung’s investment reflects a political dimension that should not be overlooked. President Lee Jae-myung campaigned heavily on reducing the economic concentration in the Seoul metropolitan area that has characterized South Korean economic development since the 1970s. The decision to extend Samsung Electronics’ semiconductor investments to Chungcheong, Yeongnam, and Incheon — rather than consolidating everything at existing facilities in Gyeonggi province — directly serves that political priority.

In exchange, Samsung gains something valuable: streamlined permitting, preferred access to government-subsidized power infrastructure, and a domestic political coalition that will advocate for Samsung’s business interests in trade negotiations. It is industrial policy in the most literal sense — government and conglomerate aligning mutual interests around a shared infrastructure investment.

What It Means for the Global AI Race

Samsung’s 1,000 trillion won plan is the largest national-level AI infrastructure commitment by a single private company announced to date, and it arrives at a moment when the AI investment landscape is already historically unprecedented.

The world’s hyperscalers collectively plan to spend over $700 billion on AI infrastructure in 2026 alone. The US CHIPS Act, the EU’s Chips Act, Japan’s semiconductor subsidies, and China’s $295 billion AI infrastructure build-out are all accelerating in parallel. Samsung’s Korean commitment adds another national-scale pillar to what is becoming a genuinely global mobilization of capital around AI infrastructure.

Whether the investment generates the economic returns Samsung is projecting depends on factors that no one can model accurately: the trajectory of AI capability development, the evolution of hardware architectures, and whether the geopolitical environment stabilizes enough to allow cross-border technology supply chains to function efficiently.

What is certain is that South Korea is betting its industrial future on the AI hardware stack — and that Samsung is the mechanism through which that bet is being made.