Micron Technology (MU) enters a new trading week under pressure after a sharp pullback on Friday, yet the memory chip maker remains a central player in the artificial intelligence investment narrative. Shares closed at $724.66, down 6.6%, following a week that saw the stock hit an intraday high of $818.67 before reversing course. The decline ended a streak of record closes and left the stock roughly 3% below its May 8 level.
The selloff was not isolated to Micron. The broader market stumbled as the 10-year Treasury yield jumped to 4.601% on Friday, a move that Interactive Brokers senior economist José Torres described as a "bear-steepening yield surge" that "ignited turbulence on Wall Street." The S&P 500 fell 1.2%, the Nasdaq Composite dropped 1.5%, and the Dow Jones Industrial Average lost 537 points. Rising oil prices added to the risk-off tone. Nvidia (NVDA), a key AI chip bellwether, slid 4.4% ahead of its earnings report due after Wednesday's close.
Despite Friday's setback, Micron shares have surged approximately 154% year-to-date, reflecting Wall Street's enthusiasm for AI-driven demand for memory chips. The company is seen as a proxy for AI data-center spending, which fuels sales of DRAM and high-bandwidth memory (HBM) used alongside AI accelerators. In March, Micron reported fiscal second-quarter revenue of $23.86 billion, up from $8.05 billion a year earlier, with non-GAAP earnings of $12.20 per share. CEO Sanjay Mehrotra called memory "a strategic asset" for customers as AI adoption expands.
Micron added to its product momentum last week by announcing it has begun sampling 256GB DDR5 server memory modules to key partners. The modules target data centers running AI workloads, with a focus on higher capacity, bandwidth, and power efficiency. Raj Narasimhan, head of Micron's cloud memory unit, highlighted these three factors as critical for AI efficiency.
The competitive landscape also offers potential tailwinds. Reuters reported that SK Hynix is nearing a $1 trillion market capitalization on AI-related demand for both conventional memory and HBM. Meanwhile, Samsung Electronics faces a possible 18-day strike starting May 21 if wage negotiations fail. Analysts suggest that rivals SK Hynix and Micron could capture some demand if Samsung's output is disrupted, though a prolonged strike might strain the broader chip supply chain.
Market sentiment around AI-related stocks remains driven by fear of missing out, according to Fabien Yip, market analyst at IG in Sydney. However, Friday's selloff served as a reminder that momentum can reverse quickly. Investors will be watching closely when Micron executives speak Wednesday at the J.P. Morgan Global Technology, Media and Communications Conference in Boston. The session, set for 8:40 a.m. EDT, may provide insights on supply trends, AI server demand, and second-half pricing.
The bull case for Micron is not without risks. Memory is inherently cyclical, with profits soaring during tight supply and falling when production ramps up. A Wall Street Journal analysis warned that heavy capital spending by Micron, Samsung, and SK Hynix could lead to oversupply. It also noted that more efficient AI algorithms or weaker-than-expected adoption could reduce future memory demand.
For now, the fundamental story remains intact: AI systems require increasing amounts of memory, and Micron is well-positioned to supply it. But after a week that demonstrated the stock's volatility, the market's willingness to sustain premium valuations for a cyclical semiconductor play will be tested in the days ahead.
