The Super Vietnamese Cycle of AI Will Last 15 Years. We Are in Year 3

The supercycle of AI is the biggest investment theme of our generation. Bigger than mobile. Bigger than cloud. A 15-year structural shift will reshape every industry in the global economy. Hyperscalers have just committed $725 billion in capital expenditure for 2026, nearly double last year. Microsoft, Google, Amazon, and Meta each spend over $100 billion individually. 👉 This is not speculation. I have mapped out the entire supercycle into four phases so you know exactly where we are and where the asymmetric opportunities lie.

Phase 1: Already Running (2023 to 2025) The foundational layer is complete. $AMD increased 78% in 2025, $NVDA rose 39%, and $INTC just announced explosive Q1 results, pushing the Philadelphia Semiconductor Index above 10,000 for the first time. Chips still power every stage but entry points for the current generation have disappeared, and risk/reward has been compressed.

• $NVDA, $AMD, $ARM, $INTC, $AVGO, $MU, $GLW
• Semiconductors, Memory & Storage, Photonics/Optoelectronics
• Platform complete. Still growing but priced for it. Phase 2: Peak Building (2025 to 2027) The stage where most investors have just awakened. $CEG bought Calpine to become the largest private power producer in the US with 55 GW. $GEV increased over 200% in one year. $VRT co-designed the cooling system for NVIDIA’s Rubin architecture. $GLW rose 74% since the start of the year thanks to fiber optic demand. SMR nuclear reactors are a breakthrough with $OKLO, $SMR , and $BWXT positioning to supply power directly to data centers. Still potential for growth but the obvious plays have moved.
• $CEG, $GEV, $VRT, $VST, $TLN, $ANET, $GLW, $MOD, $EQIX $OKLO, $SMR, $BWXT, $NNE
• Power/Grid, Cooling, Networks, Nuclear/SMR Peak Building.
• SMR nuclear reactors are the “dark horses.” Phase 3: Positioning Window (2026 to 2028) Where AI escapes the data center and enters the physical world. Most will arrive late. Tesla is converting Fremont into an Optimus factory, investing $25 billion, with mass production targeted for the second half of 2026. Rocket Lab announced record revenue of $602 million with an order backlog of $1.85 billion. $LUNR increased 47% since the start of the year with $943 million in contracts. Valkyrie drones by $KTOS are chosen for the Marines. The positioning window is opening right now.
• $TSLA, $RKLB, $LUNR, $KTOS, $AVAV, $PATH, $ISRG $MP, $FCX, $ALB, $ASTS
• Robotics/Automation, Space/Defense/Drones, Rare Earths
• This is where asymmetric risk/reward exists. Phase 4: The Final Frontier (2028+) The game is over. Microsoft’s capital expenditure is $190 billion. Alphabet’s $190 billion. Amazon’s $200 billion. Meta’s $145 billion. Google Cloud’s backlog exceeds $460 billion. They are building the rails for AI software dominance and AGI. Quantum tech is still early but $IONQ and D-Wave are laying the groundwork. Platforms controlling the software layer will win the entire supercycle.
• $MSFT, $GOOGL, $AMZN, $META, $ORCL, $IONQ
• Dominating AI Software, AGI Infrastructure The thesis lasts a decade.
• Accumulate when weak. Main Lessons
• Phase 2 is confirmed (hyperscaler capital expenditure of $725 billion)
• Phase 3 is where smart money is positioning right now: Robots, space, defense, nuclear
• SMR deals from 2026 to 2028
• Most will shift into these stocks late by 12 months The 15-year supercycle. Not a trade. Phase 1 has run. Phase 2 is priced in. Phase 3 is where you want to be.