AI Semiconductor Supercycle to Run Through 2028 as HBM Demand Expands Beyond Memory

The Quick Take

• AI-driven semiconductor supercycle is forecast to extend to at least 2028 according to industry research
• Demand is broadening from high-bandwidth memory (HBM) into CPUs, substrates, equipment, and components
• AI technology is becoming embedded in daily life, with 2027 consumer trend forecasts showing AI-native behavior shifts

Korea's semiconductor research community is forecasting an AI-driven supercycle extending to at least 2028, driven by expanding demand beyond the initial high-bandwidth memory buildout phase. Research from Semicon Research Lab indicates that while HBM dominated the first wave of AI hardware spending, the beneficiary pool is now widening to include central processing units, advanced packaging substrates, manufacturing equipment, and electronic components — a signal that the AI hardware cycle is maturing into a broad-based infrastructure buildout. Korean semiconductor names, particularly Samsung Electronics and SK Hynix in HBM, and HANA Microdisplay and Wonik IPS in equipment, stand to benefit from the expanding demand scope.

The extension of the AI semiconductor supercycle carries significant implications for Korean tech sector valuations and for the global semiconductor equipment and materials supply chain. Korean firms occupy critical positions in advanced packaging and HBM production that are difficult to replicate quickly, giving them structural pricing power in a sustained upcycle. The second source's framing of AI as becoming as pervasive as air — referenced in the 2027 Trend Korea forecast — signals that AI-driven consumer device demand is set to drive a second wave of semiconductor unit volume growth complementing the enterprise AI infrastructure wave currently driving HBM and GPU demand.

Investors should watch Samsung Electronics' HBM market share trajectory against SK Hynix, which has led in early HBM3E supply to NVDA and AMD. The equipment spending commitments from TSMC and Samsung at their advanced packaging facilities will indicate whether the substrate and component supply chain expansion is proceeding at the pace needed to meet 2026-2028 demand projections. The macro variable that determines whether the 2028 supercycle thesis holds is the pace of hyperscaler AI infrastructure spending: if Microsoft, Google, Amazon, and Meta maintain or accelerate their AI capex commitments through 2026, the Korean semiconductor supply chain benefits directly and durably.

Synthesized from 2 sources.