Foxconn has ’immense confidence’ in growth momentum due to AI, chairman says

Foxconn said it has "immense confidence" in growth as AI demand drives major cloud providers' capex above $700 billion this year, with potential to reach $1 trillion next year. The company also reported first-quarter profit up 19% year over year, beating expectations, and guided 2024 capex to rise 30% from last year's T$174 billion as it expands AI server capacity. Shares are up 19% year to date, though still well behind Taiwan's 54% gain.

Analysis

The market is still underestimating how concentrated the AI buildout is becoming in a small set of capital-intensive suppliers. If hyperscaler capex continues to compound toward the implied $1T level, the first-order winners are not just GPU vendors but the picks-and-shovels around server assembly, power delivery, liquid cooling, and packaging; that supports a second leg of earnings upgrades for the most integrated contract manufacturers. The key nuance is that Foxconn’s scale gives it operating leverage, but it also makes it a price-taker if customers push for faster deployment and lower unit costs as supply catches up.

For NVDA, this is supportive but not incrementally surprising: the market already prices strong demand, so the bigger implication is that supply chain throughput, not end-demand, remains the bottleneck. That tends to favor firms with allocation priority and system-level integration over pure component exposure. A less obvious beneficiary is AAPL indirectly: if AI server demand is pulling capacity, managements across electronics ecosystems will protect high-margin or strategic programs first, which can tighten availability for lower-priority consumer electronics and support pricing discipline.

The contrarian risk is that the current narrative extrapolates hyperscaler spending before monetization catches up. A one-year lag between capex and revenue realization would force either a spending pause or a margin reset, which would hit the entire AI supply chain simultaneously. Over 3-6 months, the biggest reversal catalyst is not demand collapse but digestion: if lead times normalize and inventory builds, the market may rotate out of manufacturing beneficiaries into software and application-layer names that can show faster payback.

The relative underperformance of Foxconn versus Taiwan equities suggests the AI supply-chain trade is crowded at the index level but still selectively mispriced within the ecosystem. This creates an opening for pairs that isolate capex beneficiaries from broader Taiwan cyclicality. The best risk/reward is in names with visible order books and low customer concentration, not in the broad beta expression of the theme.

Key Decisions for Investors

• Long NVDA on 3-6 month horizon, but only on pullbacks; use a tiered entry because the stock likely needs continued capex validation to re-rate higher, with upside tied to supply-chain checks rather than headline demand.
• Pair trade: long contract-manufacturing / AI infrastructure exposure, short broader consumer-electronics beta over 1-2 quarters; the thesis is that AI-related volumes will crowd out lower-priority hardware and preserve pricing power for capacity-constrained suppliers.
• Avoid chasing Foxconn-style beneficiaries after a sharp move; if entering, prefer call spreads rather than stock to capture upside from AI capex confirmation while limiting downside if hyperscaler spending pauses.
• For AAPL, stay neutral-to-slightly long as a defensive quality hedge rather than an AI momentum trade; the setup is better as capital reallocation from crowded AI hardware into cash-generative megacaps if the supply chain shows signs of digestion.
• Set a 4-8 week watchlist trigger on hyperscaler capex commentary; if next round of guidance softens, rotate out of AI manufacturing beneficiaries and into software/platform names with faster revenue realization.