TSMC founder Morris Chang made his first public appearance in more than a year on January 29, 2026, dining with Nvidia chief executive Jensen Huang just hours after Huang arrived in Taipei.
Elon Musk's ambition to build a massive in-house semiconductor manufacturing complex has drawn fresh caution from Nvidia CEO Jensen Huang, who stressed that advanced chip production ranks among the most technically demanding industrial undertakings globally—far beyond the reach of capital spending alone.
US semiconductor equipment maker KLA said growing AI infrastructure demand is accelerating investment in advanced logic chips, high-bandwidth memory (HBM), and advanced packaging, driving higher process control intensity, while supply constraints and cost pressures remain ongoing risks.
SanDisk delivered a strong fiscal second quarter in 2026, reflecting accelerating demand across data centers, industrial applications, and consumer electronics, driven by the ongoing AI infrastructure buildout. The US flash memory maker reported revenue of US$3.03 billion, up 61% year over year, while net profit surged 7.7-fold to US$803 million, according to its January 29, 2026, earnings release.
AI server architectures are rapidly shifting toward ultra-high-density designs, keeping advanced CoWoS packaging capacity persistently tight. The primary bottleneck stems from continuously expanding chip sizes, which introduce warpage and thermal stress challenges while driving ABF substrate demand to multiply, creating structural supply pressure and underpinning a recovery in 2025 operations for Taiwan's three leading IC substrate makers.
The recent surge in DRAM prices is widening pricing headroom across the semiconductor supply chain, prompting Chang Wah Electromaterials (CWE), a major semiconductor materials supplier and distributor, to plan price increases for epoxy molding compound (EMC) starting in March 2026.
China's two largest technology companies, Alibaba and Baidu, have launched initial public offering (IPO) processes for their semiconductor design subsidiaries, a move widely seen as part of Beijing's broader push to strengthen domestic chip self-sufficiency and secure local AI computing capacity under tightening US technology restrictions.
Recent reports from mainland China indicate that authorities have approved the import of an initial batch of Nvidia's H200 chips. The move highlights yet another turn in the intensifying US–China technology rivalry.
As artificial intelligence (AI) applications expand, the rising scale and density of server computing have placed a premium on system stability. AI servers, characterized by high costs, extreme power consumption and significant heat generation, face the risk of substantial losses from system outages. This has heightened the importance of baseboard management controllers (BMCs), which provide real-time monitoring of voltage, temperature and system status, driving steady growth in Aspeed Technology's operating performance.
Memory demand remains strong, intensifying the industry's supply-demand imbalance. Contract prices for the first quarter of 2026 are rising sharply, with major South Korean memory suppliers releasing new quotations showing DDR5 price hikes of up to 80%, while DDR4 prices are holding at roughly 50% increases. Parts of the server memory supply chain are still awaiting official price settlements in February 2026.
Taiwan's semiconductor test supply chain is experiencing strong growth driven by increased demand for AI and HPC chips, industry sources report. Close collaborations with foundries and IC design houses have helped local firms meet advanced testing requirements, boosting the competitiveness of domestic chips.
More coverage
