SK Hynix, Nvidia Seal Multiyear AI Memory Pact as HBM Market Heads for $54.6 Billion
Updated
Updated · The Motley Fool · Jun 25
SK Hynix, Nvidia Seal Multiyear AI Memory Pact as HBM Market Heads for $54.6 Billion
3 articles · Updated · The Motley Fool · Jun 25
Summary
June 6 marked SK Hynix and Nvidia’s formal move into a multiyear co-development deal that covers both supply and design of next-generation AI memory, deepening SK Hynix’s role beyond that of a component vendor.
70% of global memory output is expected to be absorbed by AI data centers in 2026, driven by longer AI context windows, higher memory per server and rising inference demand that make HBM the tightest part of the chip supply chain.
SK Hynix already holds an estimated 57% to 62% of the HBM market and 60% to 70% of HBM4 volume allocated to Nvidia’s Vera Rubin platform, extending a lead built with HBM3E into the HBM4 generation.
Bank of America projects the HBM market will jump 58% to $54.6 billion in 2026, while SK Hynix says the segment can grow at a 30% annualized rate through 2030.
That AI-driven capacity shift is squeezing other electronics markets: IDC forecasts smartphone shipments could fall as much as 5% and PC sales as much as 9% as memory supply is redirected to data centers.
Will the AI boom sacrifice consumer tech innovation and affordability for data center dominance?
Is the AI memory 'supercycle' a bubble, destined to burst into a catastrophic industry glut after 2027?
Is the AI hardware race creating an energy crisis that will ultimately throttle its own growth?
NVIDIA–SK hynix Alliance Reshapes $50B HBM Market: AI Supercycle, Supply Chain Shifts, and the Race for Next-Gen Memory
Overview
In June 2026, NVIDIA and SK hynix formed a strategic alliance to transform semiconductor manufacturing using advanced AI and simulation platforms. This partnership strengthens SK hynix’s role in the AI supply chain by focusing on the development of fab digital twins—detailed 3D factory scenes built with NVIDIA Omniverse™ and OpenUSD. These digital twins are essential for visualizing and optimizing complex manufacturing processes, supporting autonomous operations, and enabling AI systems to analyze data and automate tasks. By combining their strengths, both companies aim to drive innovation and efficiency in next-generation AI technologies and chip production.