πŸ”¬ AI capex growth to $3-$4 trillion annually by 2030 will benefit companies controlling key AI buildout components: photonics, silicon photonics, laser chokepoints, epiwafer supply chains, neoclouds, and memory

high convictionFirst mentioned Nov 25, 2025

AI capex growth to $3-$4 trillion annually by 2030 will benefit companies controlling key AI buildout components: photonics, silicon photonics, laser chokepoints, epiwafer supply chains, neoclouds, and memory.

Evidence & details
Evidence
  • +Jensen Huang projection of AI capex reaching $3-4 trillion annually by 2030
  • +List of companies and their specific roles in AI infrastructure
Catalysts
  • ⚑Exponential increase in AI capex
  • ⚑Shift to silicon photonics and CPO
  • ⚑Expansion of neoclouds and memory demand
Risks
  • βˆ’Execution risk for small companies
  • βˆ’Competition from larger players
  • βˆ’Technological disruption

Chart

Post Timeline Β· 1 posts

new thesishigh

AI capex spend is expected to go to "$3 to $4 trillion annually" by 2030 from $NVDA Jensen Huang projections. You're not bullish enough. And it might be a good idea to stay exposed + own the keys of the AI Kingdom: -> $AXTI controls the materials buildout with photonics. -> …

Thesis: AI capex growth to $3-$4 trillion annually by 2030 will benefit companies controlling key AI buildout components: photonics, silicon photonics, laser chokepoints, epiwafer supply chains, neoclouds, an…

view on X β†—AXTI Β· SOI Β· SIVE Β· IQE Β· NBIS Β· MU Β· SNDK
Current Price
$1.64K
via cornerstones
Return from entry
β€”
entry Nov 25, 2025
Thesis Lifecycle
πŸ“
updatedMar 6, 2026

NAND flash price hikes by Samsung and likely others far exceed analyst expectations, implying massive upside for memory companies.

source β†—
πŸ“
updatedMar 7, 2026

Market misinterprets the gold rush metaphor; real alpha comes from bottlenecking the shovel sellers by investing in materials suppliers critical to semiconductor manufacturing.

source β†—
πŸ“
updatedMar 7, 2026-1 conv

Bottlenecks in supply chains, specifically in photonics (substrates/EML lasers) in Asia, have historically produced large returns, and identifying similar bottlenecks can generate alpha. Examples include Nittobo, Macronix, Nanya Glass, and US players SNDK and LITE.

source β†—
πŸ“
updatedMar 10, 2026+1 conv

NAND price surge benefits memory companies due to inelastic supply and strong demand from hyperscalers.

source β†—
πŸ“
updatedMar 10, 2026-1 conv

NAND price hikes and structural memory demand will lead to strong operating income for memory companies, contrary to consensus that rate hikes are unsustainable.

source β†—
πŸ“
updatedMar 13, 2026+1 conv

Memory prices are expected to remain elevated due to unbroken hyperscaler demand outstripping supply growth, leading to explosive operating margin increases for memory makers.

source β†—
πŸ“
updatedMar 15, 2026

Memory cycle structural change due to AI leads to sustained high profits despite future price declines, making memory companies like SK Hynix, SNDK, and MU undervalued.

source β†—
πŸ“
updatedMar 16, 2026

Memory names are bullish due to structural demand and high operating profits, with SK Hynix potentially making $400B in operating profit by 2028, indicating room for re-rating.

source β†—
πŸ“
updatedMar 16, 2026

Memory sell-off due to Helium concerns was overblown; fundamentals strong with NAND/DRAM hikes beyond estimates and company statements of no material effect.

source β†—
πŸ“
updatedMar 16, 2026

Memory names are critical to AI buildout; fears of disruption from Iran conflict are overblown, and long-term operating income will drive prices.

source β†—
πŸ“
updatedMar 20, 2026-1 conv

AIXA is like ASML for photonics and captures a large capex cycle this year, while beneficiaries include bottleneck price-hikers like SNDK, EML like LITE, and substrates like AXTI.

source β†—
πŸ“
updatedMar 23, 2026

Hyperscaler capex flowing upstream benefits photonics sector companies, leading to fundamental re-rating and outperformance regardless of macro.

source β†—
πŸ“
updatedMar 25, 2026

Sandisk's $1B+ stake in Nanya suggests legacy memory bottlenecks are bigger than market thinks, bullish for memory companies.

source β†—
πŸ“
updatedApr 7, 2026+1 conv

Multi-year bottlenecks in transformers (HPS.A), NAND (SNDK), and EML/OCS (LITE) create near-guaranteed revenue and margin expansion due to backlogged orders into 2028.

source β†—
πŸ“
updatedApr 10, 2026-1 conv

Forward earnings compression with SNDK and AAOI is a buying opportunity as shorts fuel the rally due to misunderstood growth prospects and TAM expansion.

source β†—
πŸ“
updatedApr 13, 2026

AAOI and SNDK are large beneficiaries of photonics and memory supercycles, making them poor short targets despite hedge fund actions.

source β†—
πŸ“
updatedApr 30, 2026+1 conv

SNDK reported blowout Q3 earnings and guided Q4 well above estimates, confirming strong momentum in memory companies driven by NAND price hikes.

source β†—
πŸ“
updatedMay 6, 2026-1 conv

CPO market growth is in a supercycle; names like AAOI and SNDK are high-beta correlated to CPO and should not be shorted.

source β†—
πŸ“
updatedMay 9, 2026+1 conv

Recent semiconductor developments indicate bullish opportunities across multiple tickers: Intel benefits from Apple shift, Micron from memory tightness and price increases, TSM from revenue growth, NBIS from compute demand, LPK from glass substrate acceleration, and SanDisk from NAND price rise.

source β†—
πŸ“
updatedMay 24, 2026

AI capex growth to $3-$4 trillion annually by 2030 will benefit companies controlling key AI buildout components: photonics, silicon photonics, laser chokepoints, epiwafer supply chains, neoclouds, and memory.

source β†—