DanielRomero

The cost of an $NVDA AI factory will soon 5x
➢ Jensen said that the cost of an AI factory per GW used to start at $20B to $30B
➢ Now, it is closer to $50B to $60B
➢ Soon, it could reach $80B to $100B
Memory has a lot to do with that

Jensen isn’t pulling any punches on $AMD and $INTC

$NVDA Rubin is officially in full-rate production
The semiconductor supply chain rally can continue

Can $PENG AI factory platform compete when DSX Sim Omniverse already exists?

$IREN made an appearance in an $NVDA presentation

$MU just crossed $1000

Mizuho Securities analyst Vijay Rakesh maintains Advanced Micro Devices $AMD with a buy rating, and adjusts the target price from $515 to $615

JP Morgan still expects DRAM and NAND ASP growth to decelerate sharply from late 2026 to early 2027
$MU

NVIDIA wants to eat $AMD and $INTC lunch
AI agents have made CPUs important again
$NVDA and Microsoft are expected to unveil the first Windows PCs powered by Nvidia chips next week
The devices are expected to include Microsoft Surface models and PCs from other manufacturers like Dell. They will reportedly be announced around Computex in Taiwan and Microsoft’s Build conference
Nvidia is entering the Windows PC CPU market with Arm-based technology, putting it in direct competition with Qualcomm and challenging the long-time dominance of Intel and AMD in Windows laptops
Microsoft is also expected

JP Morgan still expects DRAM and NAND ASP growth to decelerate sharply from late 2026 to early 2027
A significant contrast with UBS’s latest projections
$MU

Korean memory stocks are seeing aggressive target hikes
NH Investment & Securities raised its target for Samsung Electronics (005930.KS) to 490,000 won, from 310,000 won, citing stronger memory conditions and improving foundry / System LSI expectations
SK Securities raised its targets for Samsung Electronics (005930.KS) to 500,000 won and SK Hynix (000660.KS) to 3 million won, arguing that both stocks remain undervalued after restoring prior target P/E assumptions
Nomura also reportedly raised its target for Samsung Electronics (005930.KS) by 74% to 590,000 won, and its target for SK Hynix (00

$AMD: Mizuho raised its target to $515 from $415, citing rising CPU demand from agentic AI workloads

$MU Price Target Updates
- > Mizuho raised Micron to $1,150 and Sandisk to $1,825
- > D.A. Davidson raised Micron to $1,500
- > Barclays raised Micron to $1,175 and Sandisk to $2,300, while upgrading Sandisk to Overweight.

Morgan Stanley says the setup for legacy memory is getting tighter than expected
DDR4 pricing could rise another 20% in Q3 as supply keeps shrinking, with the bank now modeling a 19%-20% supply-demand gap in 2H26 and 18%-20% gaps in 2027-2028
Other industry coverage:
TrendForce says DDR4 will remain in short supply through 2026, even after Micron’s Virginia Fab 6 ramp
$MU move is mostly an internal capacity reallocation, not a real increase in total LPDDR4/DDR4 capacity. TrendForce expects DDR4 to represent only around 7% of Micron’s DRAM output in 2026, with that share declining further in th

Samsung started shipping HBM4E samples to customers
The chip is over 20% faster than its previous HBM4 product and uses Samsung’s latest 1c DRAM, plus its own 4nm foundry base die
Anthropic also named Samsung as a strategic infrastructure partner, and Samsung was highlighted for its logic chip capabilities, not just memory
Current HBM share as of Q4 2025:
- > SK hynix: 57%
- > Samsung: 22%
- > $MU: 21%

$10B over 5 years for quantum computing investments seems like a major conviction leap for the industry
$IBM generates around $15B in FCF per year. Dedicating roughly 13% of annual FCF to this new technology is massive

$DELL said its customer count surpassed 5,000, with growth across neocloud, sovereign, and enterprise customers
This was a fairly bullish call for $PENG
Trading at 26x forward P/E, with revenue growth accelerating
Not cheap, but the next prints should be good

$DELL +38% now in after-hours
$NVDA is pumping every company in the supply chain
Jensen single-handedly created millions of new millionaires

As mocked as it is, the bottleneck trade will stay forever
You only need to project the next 100 years to realize that AI is going to bottleneck everything
The first bottlenecks appeared in the areas most directly exposed. That was memory and power access
Then it became substrates, MLCCs, logic fabs, lasers, even copper foil and power semis
It’s only a matter of time before AI and robotics/physical AI accelerate the rate of progress to the point where every physical asset becomes shortaged
Maybe I’m wrong, and this is clearly a very early thought, but the everything shortage is a natural conse