7

Most people are learning about memory stocks through AI news or their favorite financial influencers, who themselves only recently discovered the sector. Here’s my edge: I’ve actually bought memory directly from these suppliers. During my time at a Fortune 50 tech company, I worked as a Memory Operational Commodity Manager, giving me firsthand knowledge of how the industry operates and how the major semiconductor suppliers think. That direct relationship with their largest clients and understanding of their supply chain constraints, is something you can’t get from Twitter threads or financial media. Recently, major tech firms are now offering to pay for SK Hynix’s equipment capex. If you understand semiconductor economics, you know this is absolutely unprecedented. It signals desperation for supply and is massively bullish for the memory supercycle ahead. By subscribing, you’re getting analysis from someone who actually lived inside this industry not just someone reading about it. That’s the difference between surface-level commentary and real conviction. Join the LRMI community today! $MU $DRAM $NVDA $SMH $SNDK $WDC $SKHYNIX lrmi.ai/p/we-dont-own-enough…

Some food for thought… What if Intel $INTC just completely got rid of their consumer business and went all in on DCAI and Foundry, just like how Micron $MU completely cut off their consumer line and has fully went all in on DC demand

$NVDA the private credit firm?

INTEL $INTC: Why the stock is down -23% after-hours The clearest explanation is a sharp reset to near-term earnings and 2027 cash-flow expectations, despite a good Q3. Intel beat its own Q3 outlook, but guided to falling revenue, substantially lower margins and roughly halved adjusted EPS in Q4. The call then extended the PC weakness and supply constraints into 2027. 1. Q4 guidance overwhelms the Q3 beat Q3 revenue was $16.5 billion, only $0.2 billion above Intel’s July outlook, while non-GAAP EPS of $0.52 beat that outlook by $0.14. For Q4, management’s October 29 guidance implies: - Revenue down approximately 6% QoQ, to the $15.5 billion midpoint from $16.5 billion. - Non-GAAP gross margin down 610 basis points QoQ, from 44.1% to 38%. - Non-GAAP EPS down approximately 52% QoQ, from $0.52 to $0.25. 2. The PC deterioration is severe, and extends beyond one quarter Intel expects client computing and physical AI revenue to decline more than 20% QoQ in Q4, following a 9% sequential decline to $8.1 billion in Q3. Customers are reducing inventory as memory shortages and higher system costs weaken consumption. The Q&A made the duration more concerning: The inventory correction is concentrated in Q4 and Q1, potentially stretching into Q2 2027. Management expects the 2027 PC unit market to fall approximately 10%, and explicitly said it does not expect the memory situation to improve in 2027. This makes the weakness harder to dismiss as a temporary inventory adjustment. 3. The margin decline is not just the Israel factory incident Management estimates the Fab 28 shutdown will cost approximately $300 million of Q4 revenue and 200 basis points of Q4 gross margin, largely affecting server products. The CFO expects a similar revenue impact in Q1 2027, although most wafer write-offs should occur in Q4. Even adding back the estimated factory impact, Q4 non-GAAP gross margin would be approximately 40%, still 410 basis points below Q3. Management identified lower client factory loading and product mix as additional pressures; purpose-built silicon carries below-corporate-average margins. 4. The call pushed positive free cash flow further out Asked directly about 2027, CFO Dave Zinsner said it was probably not a year in which Intel would generate positive free cash flow, because several manufacturing investments are occurring together. Management expects 2027 gross capital spending to rise strongly from its $20 billion 2026 plan, with equipment purchase orders already up nearly 40% sequentially. Intel also said Q4 would reflect a full quarter of shares from August’s $23 billion equity offering. The implication is a longer wait for the turnaround to translate into cash generation per share. 5. Strong AI demand cannot immediately offset the damage The positive is real: Q3 DCAI revenue grew 78% YoY to $7.3 billion, and management expects mid-teens sequential growth in Q4. But Intel remains constrained by wafers and substrates in the first half of 2027, with additional capacity becoming available in the second half. The CFO would not commit to sequential DCAI growth in Q1. The damaging combination is a collapsing PC order outlook, lower factory utilization, constrained ability to monetize server demand, and another investment-heavy year without positive free cash flow. Those disclosures give a much stronger explanation for a sharp selloff than the headline Q3 results alone.

INTEL $INTC: "Strong demand across hyperscale, enterprise and neo clouds Purpose Built Silicon revenue more than doubled YoY Agentic workloads driving increased customer interest across product roadmap"

GOOGLE $GOOG $GOOGL: Show me the CapEx! +114% Y/Y Growth

MICROSOFT $MSFT: CapEx increased 30% to $45.2 billion "Roughly 3/4's of CapEx we're for short-lived assets, primarily CPUs & GPUs" Microsoft spent roughly $33.9 billion on just new CPUs and GPUs alone!!

MICROSOFT $MSFT: "Microsoft’s OpenAI investment boosted Q1 FY2027 net income by $207 million, or $0.03 per diluted share. The larger effect on earnings growth came from comparing that gain with a substantial investment loss a year earlier."

GOOGLE $GOOG $GOOGL: Latest TPU Systems v7 Example

GOOGLE $GOOG $GOOGL: I'd say AI is generating some good returns

GOOGLE $GOOG $GOOGL: Google Cloud: +103% Q3 Y/Y revenue growth Gemini: ~32B Tokens per minute, up 40% Q/Q This is massive

$ORCL is heading back to $200

$SKHY The demand + share buy backs seem so attractive

Samsung will as well

SK HYNIX $SKHY: SK hynix is describing a pricing-led memory upcycle with demand exceeding supply through 2030 and more than half of 2027 and 2028 capacity covered by long-term agreements, making sustained pricing power versus an already-anticipated peak the key trading question.