Micron stock rose 3% on Monday as investors positioned for earnings due Wednesday, with the shares on pace to hit a fresh record. The move puts Micron Technology in the center of the memory trade again, and it leaves holders facing a familiar setup: a fast run into results, with expectations already elevated.
Melissa Weathers and $1,500
Melissa Weathers of UBS raised her price target on Micron to $1,500 per share. In her note, she said demand for DRAM is still set to vastly outpace supply growth in the coming years, driven by more memory-intensive AI workloads.
That call gives the stock’s latest move more weight than a simple pre-earnings drift. If demand keeps outrunning supply the way Weathers expects, pricing power can stay firm longer than the market would normally assume, especially in a part of the chip market where memory has become a bottleneck in AI.
360% before Wednesday
Micron stock is up roughly 360% in the past six months. That scale of gain means much of the rerating has already happened before Wednesday’s report, so the burden on the earnings release is less about proving the story exists and more about showing it can keep going at this pace.
Last month, Micron Technology, Samsung Electronics and SK Hynix all reached $1 trillion valuations for the first time, a marker of how deeply memory demand has been repriced across the group. On Monday, SK Hynix passed Samsung Electronics as South Korea’s largest company after raising unit prices amid rising costs from tighter supply.
Thursday highs in memory
On Thursday, Micron hit an all-time high, while Sandisk and Western Digital also reached records. That streak shows how broadly the memory trade has extended beyond one name, with investors treating the group as a direct read on AI-related demand rather than a narrow company-specific move.
Wednesday now becomes the test. Micron’s report will show whether AI inference demand is still powerful enough to support the kind of memory pricing that has already pushed the stock, UBS’s target and the rest of the group to extremes. The unresolved issue is how much of the surge comes from AI inference itself versus a wider bet on tighter memory supply.
