
Micron Technology gained sharply after UBS increased its price target from $535 to $1,625, implying another major valuation expansion. The move followed Micron’s entry into the trillion-dollar club and came amid continued strength in AI-related stocks. UBS tied its outlook to rapidly growing demand for high-bandwidth memory used alongside GPUs in AI data centers. HBM supports faster processing of large datasets and lower power consumption for large language models. Micron management indicated that HBM production capacity for 2026 is already sold out, contrasting with memory’s historical boom-and-bust cycle. The article notes that memory scarcity may be becoming as important as GPU availability as hyperscalers invest heavily in AI compute capacity.
"The catalyst behind Micron's surge was a new price target from UBS analyst Timothy Arcuri, who raised his target from $535 to $1,625. That's not a typo. The increase implies another doubling from Micron's already elevated valuation after the stock crossed the $1 trillion mark. The stock is up 832% year-to-date."
"UBS based much of its thesis on exploding demand for high-bandwidth memory , or HBM, which has become essential infrastructure for AI data centers. HBM is the specialized memory packed next to GPUs from companies like Nvidia ( NASDAQ:NVDA ), and it helps large language models process huge datasets at faster speeds while reducing power consumption."
"Micron management recently said its HBM production capacity for 2026 is already sold out. That matters because memory historically has been a boom-and-bust business driven by oversupply. AI may be changing that equation."
"Surprisingly, the bottleneck in AI infrastructure isn't just GPUs anymore. Memory is becoming equally scarce. Training advanced AI models requires enormous bandwidth, and every hyperscaler - from Microsoft ( NASDAQ:MSFT ) to Amazon ( NASDAQ:AMZN ) and Meta Platforms ( NASDAQ:META ) - is spending aggressively to expand AI compute capacity."
Read at 24/7 Wall St.
Unable to calculate read time
Collection
[
|
...
]