Commentary: Where Smart Money Pivots Amid Tech and Global Conflicts
Listen to the full version

As global geopolitical tensions pivot to the conflict between the U.S. and Iran, and U.S. equities endure a brutal risk-off correction, global equity markets are experiencing severe whiplash. The critical question for investors in April 2026 is: Where is the money going?
The current retreat of global capital resembles peeling an onion — layer by layer, from the outside in. In the artificial intelligence sector, application stocks rallied first and are now the first to fall, with many software companies plummeting from their peaks. Infrastructure stocks remain the core of the onion. Nvidia has managed to stay relatively high, but historical parallels to the 2001 dot-com bubble are alarming.
Unlock exclusive discounts with a Caixin group subscription — ideal for teams and organizations.
Subscribe to both Caixin Global and The Wall Street Journal — for the price of one.
- DIGEST HUB
- Global capital shifts from AI apps to HALO assets like Walmart/oil majors and emerging markets (South Korea, Brazil) amid US-Iran tensions.
- Conflict accelerates energy transition to renewables/nuclear; post-IPO tech needs strategic capital like sovereign funds.
- AI winners: full-stack firms (Google, Alibaba); China upgrades supply chains for quantum/AI/aerospace; space eyes leasing models.
1. In April 2026, amid escalating U.S.-Iran conflict and a risk-off correction in U.S. equities, global markets face severe volatility, prompting investors to question capital flows [para. 1].
2. Global capital retreat mirrors peeling an onion: AI application stocks, which rallied first, are now plummeting, while infrastructure like Nvidia holds but evokes 2001 dot-com parallels [para. 2].
3. Michael Burry warns Nvidia's long-term procurement commitments echo Cisco's 2001 burden, risking liabilities if AI apps fail profitability [para. 3][para. 4].
4. U.S. capital shifts to HALO investments (Heavy Asset, Light on Obsolescence), favoring Walmart and oil majors over AI; exiting dollar assets boosts emerging markets like South Korea (memory chips) and Brazil (tariffs) [para. 5].
5. New tech IPOs face post-six-month crucible; China's AI stocks volatile due to lock-ups, with selling pressure post-expiry [para. 6][para. 7].
6. Only firms attracting strategic capital, e.g., Middle Eastern sovereign funds, achieve higher valuations, highlighting capital savvy [para. 8].
7. U.S.-Iran conflict differs from 2003 Iraq War: NATO sidelined, drones challenge U.S. superiority; markets price stagflation akin to 2020 crash [para. 9][para. 10].
8. Strait of Hormuz vulnerability accelerates energy transitions to wind, solar, nuclear; supply chain diversification; China's biomanufacturing reduces petrochemical dependence [para. 11].
9. AI shifts to token price wars; China surpasses U.S. consumption, prioritizing speed, stability, cost over model prestige [para. 12][para. 13].
10. Full-stack dominators (chips, cloud, models, apps) like Google and Alibaba win via scale-subsidized costs and iteration [para. 14].
11. Aerospace nears standardization/leasing shift; space costs high from non-standard parts—reusable rockets first, then engine leasing like aviation [para. 15][para. 16][para. 17].
12. Scaled financing to slash payload costs, enabling commercial spaceflight [para. 18].
13. China's future industries (quantum, embodied AI, aerospace) focus supply chain upgrades for incremental value via motors, sensors, lasers [para. 19][para. 20].
14. Chinese local governments excel with rapid, targeted policies for industrial iteration [para. 21].
15. Amid fragmentation, capital seeks tangible utility, resilient infrastructure against conflict and recalibration [para. 22]. Li Feng, FreeS Fund partner, authors; views not Caixin's [para. 23][para. 24].
(Word count: 498)
- Nvidia
- Nvidia remains relatively high as an AI infrastructure stock amid market whiplash, but faces warnings from Michael Burry. Its massive long-term procurement commitments echo Cisco's 2001 dot-com bubble trap, potentially becoming liabilities if AI applications fail to prove profitability. (48 words)
- Cisco
- In 2001, during the dot-com bubble burst, Cisco's peak-demand orders became crushing liabilities, per Michael Burry's warning. This parallels Nvidia's AI procurement risks if applications fail to prove profitable.
- Walmart
- Walmart is cited as a HALO (Heavy Asset, Light on Obsolescence) investment—a low-valuation, heavy-asset safe haven. Amid U.S. equities' risk-off correction and AI stock retreats, capital is flocking to Walmart and oil majors. (38 words)
- Google is cited as an undisputed winner in the AI race, controlling the full four-layer stack: chips, cloud infrastructure, large models, and frontend applications. It leverages internal scale to subsidize cloud costs and enable rapid technological iteration. (48 words)
- Alibaba
- Alibaba represents full-stack dominance in the AI race, controlling chips, cloud infrastructure, large models, and frontend applications. It leverages internal scale to subsidize cloud costs and enable rapid technological iteration, positioning it as a winner amid market shifts.
- FreeS Fund
- Li Feng, author of the article, is a founding partner at FreeS Fund.
- PODCAST
- MOST POPULAR





