On May 26–27, 2026, Micron Technology and SK Hynix each saw their market capitalizations surpass $1 trillion as investors piled into memory makers powering AI workloads. UBS lifted its Micron price target to $1,625 and SK Hynix jumped over 11%, reflecting surging demand for high‑bandwidth memory used in Nvidia’s AI GPUs.
This article aggregates reporting from 4 news sources. The TL;DR is AI-generated from original reporting. Race to AGI's analysis provides editorial context on implications for AGI development.
Micron and SK Hynix joining the trillion‑dollar club within 24 hours is a clean signal of where markets think AI’s chokepoints really are. We’ve spent two years obsessing over GPU vendors; this re‑rating says memory — especially high‑bandwidth memory for training and inference — is just as strategic as the compute die itself. When HBM makers command oil‑producer style pricing power, you know the bottleneck has moved further down the stack.
For the AGI race, that means capacity growth is going to be governed as much by how fast we can manufacture and package advanced memory as by Nvidia’s GPU roadmap. If Micron and SK Hynix can fund aggressive capex out of trillion‑dollar equity valuations, the industry may be able to keep scaling training runs without hitting an immediate cost wall. That’s good news for labs pushing context length, multimodality and agentic workloads.
But it also concentrates systemic risk. Two or three firms now sit at the fulcrum of AI infrastructure, with deep exposure to geopolitical shocks, export controls and process‑node hiccups. Any disruption in HBM supply would ripple straight into model training schedules, potentially doing more to slow frontier progress than headline regulation.


