South Korea’s chip investment plan, unveiled by President Lee Jae Myung at a televised event alongside the chiefs of Samsung and SK Hynix, is the most ambitious the country has ever attempted. AP via US News reports the total package exceeds $576 billion over several years, spanning chip fabrication, AI data centres, and related infrastructure.

At the core of the plan is a semiconductor manufacturing hub in South Korea’s southwest. Samsung and SK Hynix will invest a combined 800 trillion won ($518 billion) alongside suppliers to build two new chip fabrication sites each in the region, Reuters via Yahoo Finance reports. That $518 billion figure covers the chip-fab component alone; the broader $576 billion total includes AI mega-projects beyond the fabrication sites.

The southwestern city of Gwangju and South Jeolla province will invest a further 5 to 20 trillion won in the chipmaking projects. An additional 81 trillion won is earmarked for a chip packaging cluster in the Chungcheong area near Seoul, according to Reuters.

South Korea’s Chip Investment and the Regional Race

‘We must secure the core elements of AI faster than any other country,’ Lee said at the announcement. ‘Semiconductors, physical AI, and AI data centres are the triple axis for a great leap forward.’ The framing was deliberate: Taiwan, China, and Japan are all committing heavily to domestic semiconductor capacity as AI-driven demand for chips intensifies globally.

Lee also announced plans for AI infrastructure hubs outside Seoul, where most of the country’s advanced manufacturing is currently concentrated. He described the project as a matter of ‘survival,’ framing regional rebalancing as both an economic and a social imperative. ‘Now, we must break this long-standing cycle of discrimination and marginalisation,’ he wrote in an earlier statement, ‘not only for the sake of justice and equity, but also to ensure sustainable and inclusive growth.’

Ahead of the public event, Lee met separately with SK Group Chairman Chey Tae-won and SK Hynix CEO Kwak Noh-jung to discuss semiconductor and AI competitiveness, the Korea JoongAng Daily reports. The private session covered both investment plans and AI data centre development before the announcements were made official.

The Valuation Stakes for Samsung and SK Hynix

The two companies have become proxies for the global AI trade. CNBC reports that SK Hynix and Samsung together account for more than 40% of South Korea’s benchmark Kospi index, binding the index’s performance tightly to global AI semiconductor demand.

SK Hynix crossed the $1 trillion market capitalisation threshold in May, with shares up approximately 250% since the start of the year as of that point, driven by surging orders for high-bandwidth memory chips used in AI servers. Investing News Network reports that by 28 May SK Hynix closed with a market cap of approximately 1,631 trillion won ($1.08 trillion), while Samsung Electronics stood at approximately 1,750 trillion won ($1.16 trillion). Micron Technology crossed the $1 trillion threshold around the same time.

The customer base underlines why the stakes are so high. Samsung and SK Group count Nvidia among their buyers. US tech giants including Google, Amazon, and Meta have said they will spend $650 billion on AI technology this year, sustaining the demand that has driven a global semiconductor shortage and pushed component prices higher. Apple and Microsoft raised prices on some devices last week, citing higher component costs.

South Korea’s ambition, however, runs ahead of some structural realities. The U.S. International Trade Administration notes that the country’s semiconductor ecosystem remains less developed in system semiconductors, fabless chip design, AI chips, advanced packaging, and core materials and equipment, despite its global leadership in memory and large-scale manufacturing. To address that gap, the South Korean government’s K-On-Device AI Semiconductor Technology Development Project plans to invest approximately $664 million between 2026 and 2030 to strengthen indigenous AI chip capabilities.

Whether $576 billion is enough to close those gaps against rivals who are moving just as fast is the question the next few years will answer.

Shares: