High-Bandwidth Memory (HBM)
+ FollowHigh-bandwidth memory, or HBM, uses stacking and advanced packaging to improve data-transfer efficiency and is primarily deployed in AI accelerators and high-performance computing. Recent AI demand has prompted SK hynix and other companies to expand capacity and transform their businesses. Capacity bottlenecks, slowing price increases, moves by AI server makers to reduce reliance on NVIDIA, a South Korean antitrust investigation and a U.S. patent case have added uncertainty. HBM affects chip performance, demand for advanced packaging and the memory industry's business cycle, making supply, demand, competition and regulatory changes important to watch.
Key Moments
13 selectedThe full history is split into 13 equal periods by event count, taking the most-covered story from each. Coverage rises and falls with the news cycle, so sampling period by period keeps the most recent events from taking everything.
- 2026-09-01 CXMT Begins HBM3E Trial Production, Eyes Mass Output Within Weeks 2 reports
- 2026-08-24 Micron Warns AI Compute Growth Is Outrunning HBM Bandwidth 2 reports
- 2026-08-04 NVIDIA Weighs Cutting Rubin Ultra HBM Configuration 4 reports
- 2026-07-27 SK Hynix Posts Record Q2 Profit on AI Memory Boom 7 reports
- 2026-07-15 SK Hynix ADR Surge Sparks Rebound in AI Memory Stocks 3 reports
- 8 more key moments Sign up to see the full context
Event Timeline
164 related eventsSubscribe to Mark Radar Weekly
Every Friday, the week's strongest signals in your inbox. Unsubscribe anytime.