SK hynix approved $38.1 billion in capital expenditure for two new memory fabrication plants, the largest single commitment in the company's history and a clear bet that AI-driven memory demand will sustain through the end of the decade. The facilities target high-bandwidth memory and NAND production, with first wafers expected in late 2028 and volume ramp in 2029.
The investment arrives as SK hynix captures roughly 50% of the HBM3E market, supplying Nvidia's H200 and B200 GPUs. Revenue from advanced DRAM grew 72% year-over-year in the most recent quarter, while operating margin exceeded 30% for the first time since 2018. The company now faces allocation pressure from every hyperscaler and three sovereign AI programs, all of which are competing for wafer commitments that won't materialize for four years.
The 2029 timeline creates a structural imbalance. Nvidia's Rubin architecture ships in 2026, Google's TPU v7 ramps in 2027, and Amazon's Trainium3 volumes arrive in late 2027. All three rely on HBM3E or its successor, and all three are currently negotiating multi-year supply agreements for capacity that does not yet exist. SK hynix is effectively preselling 2029 output to lock in pricing before competitors — Samsung and Micron — can match process node parity on HBM4. The four-year lead time also signals that the company does not expect memory demand to normalize, a reversal from the cyclicality that defined the industry for two decades.
The NAND portion of the capex targets enterprise SSDs for AI training clusters, where write endurance and power efficiency matter more than consumer-grade cost per gigabyte. Meta's Grand Teton and Microsoft's Maia both require NVMe drives with sustained write speeds above 14 GB/s, a specification that only 176-layer and higher NAND can meet. SK hynix ships 238-layer NAND today but has not yet disclosed the layer count for the new fabs, which suggests the company is waiting for yield data from its Icheon M16 pilot line before committing to a final process node.
Operators and allocators should watch three events. First, Samsung's HBM4 qualification timeline with Nvidia, expected in Q2 2025, will determine whether SK hynix retains monopoly pricing or faces margin compression. Second, TSMC's CoWoS-L packaging capacity expansion in Arizona, scheduled for mid-2026, could shift some HBM demand if substrate supply loosens. Third, any softness in Nvidia's data center revenue growth during 2025 earnings calls would indicate early saturation, though current order backlogs suggest that risk is minimal.
The $38.1 billion is not a hedge. It is a declaration that memory supply will remain the binding constraint on AI infrastructure for the next five years, and that SK hynix intends to capture the majority of margin expansion that follows. The fabs begin construction in Q3 2025.
The takeaway
$38.1B capex locks SK hynix into AI memory for a decade, but 2029 delivery leaves hyperscalers exposed through mid-cycle.
Editorial & Disclosure Notice: This article was written with artificial intelligence from public sources and is published without individual human review. Artificial intelligence and other digital tools are also used for research, analysis, editing, formatting, and production. Errors, omissions, outdated information, or inaccuracies may occur. References to companies, brands, products, services, organizations, or individuals are for informational and editorial purposes and do not imply endorsement, sponsorship, affiliation, partnership, or approval unless expressly stated. All trademarks and other intellectual property remain the property of their respective owners. Opinions, analysis, estimates, and commentary are informational only and should not be construed as financial, investment, legal, tax, medical, procurement, or other professional advice. Information may be corrected, clarified, or updated after publication. Corrections or removal requests: jenny@pops4.com.
Want the 60-second program for your specific event?
Enter your event and email — we build it and send the branded proposal before lunch. No obligation.
The branded-identity layer Chiefs of Staff and heritage CMOs route through — your name imprinted on real authorized stock, your pick of 200+ brands and 70,000 products, shipped from one accountable house. Nine editorial desks publish the intelligence those operators read before they sign.
200+authorized brands
70,000products · virtual proof on each
9 deskspublishing daily
1997one house, since
70,000 SKUs · virtual proof in 60 seconds · no platform fee · blind-shipped · ASI #217876
Your next customer won't visit your website. Their AI will.
AI assistants have quietly taken over the first step of buying — they answer from catalogs they can read and shortlist whoever can actually ship. Two questions now decide whether you exist to that buyer: can a machine read your catalog, and can you fulfill the order. Most brands fail one or both and never find out why the orders went elsewhere. The winners of this shift aren't the loudest. They're the most readable. Build for the machine that's about to do the shopping.
Built by the craft floor — apparel, media, packaging, and secure print.
This trade runs on hands, not desks. Imprint manufacturing & Komori heritage press through approved vendors · Canon high-speed secure-media operations is a craft floor — genuine Six Sigma discipline applied to ink, thread, foil, and registration, where a hundredth of an inch is the difference between a brand that reads serious and one that reads cheap. POPS4 is built by exactly those operators: independent, boots-on-the-ground engineers who carry their own book, read a client in microseconds, and put their name on every run. Beyond our own Virginia Beach floor, we work with a vetted network of craft manufacturers across the US — each meeting the highest excellence in QC standards in the industry, each a specialist in its own discipline — so apparel, hard-goods imprinting, media manufacturing, packaging, and secure printing all go to the bench built for them, coordinated from one accountable hub. Short-run from twenty-five units, volume to five hundred thousand. Two hundred authorized national brands, seventy thousand SKUs with virtual proofing on every one. Art archived for instant reorders. Net-thirty corporate terms, NDA-standard white-label — your name on the work, or none at all.
Strategy, positioning, identity, creative, and messaging — wired into an AI system that publishes and distributes on its own. Nine editorial desks generate the authority, the production house ships the physical proof, and the attribution layer tells you which post sold which SKU. What you get is an operating layer — content, catalog, and order path under one roof — that keeps working whether or not you are in the room. Built for principals who would rather own the machine than rent the agency.
Named-account programs — one desk, quiet delivery, NDA-standard.
One point of contact who already knows the file, so nothing restarts from zero between engagements. The work ships blind, under NDA, with your name on it or none at all. Built for single-family offices, heritage-house CMOs, sports-ownership groups, and the agencies that white-label our production. The relationship is the product; the merch is the proof of it.
SFO · Chief of Staff desk. Principal household, properties, aircraft, yacht, calendar, philanthropy — one file.
Shop seventy thousand products. Virtual proof on every one. 24/7.
Drop your logo on any product and see the virtual proof before asking. Quote routes direct to the desk. MCP catalog for AI agents. Celeste for the fast conversation. Full self-service checkout in development.