3 Artificial Intelligence (AI) Stocks Hedge Funds Are Buying Hand Over Fist

13F “smart money” flows in Q2 show hedge funds buying Amazon, Alphabet, and Taiwan Semiconductor as the AI/data-center buildout ramps. Amazon reported 37% YoY AWS growth, while North America sales rose 16% and International revenue increased 15%; Alphabet’s Google Cloud revenue jumped 82% YoY. The article frames this as confirmation for continued AI-driven cloud and chip demand, with no explicit valuation or earnings-per-share figures cited.

Analysis

The crowded ownership pattern matters more than the headline growth numbers: these are now consensus AI compounders, so the next leg of outperformance will likely come from operating leverage, not multiple expansion. For AMZN and GOOG, the key mechanism is that cloud acceleration should eventually improve free-cash-flow conversion if capex intensity plateaus; if capex keeps outrunning revenue, the market will punish the stocks even if growth stays strong. That makes the next 1-2 earnings prints the critical horizon, not the next few trading sessions.

TSM is the cleanest second-order beneficiary because it captures the spending curve regardless of which chip architecture wins, but it also inherits the most underappreciated risk: any AI capex pause by hyperscalers would hit foundry utilization before it shows up in end-demand commentary. The market tends to overpay for “pick-and-shovel” certainty; the contrarian risk is that TSM’s moat is real, but the stock can still de-rate if investors decide the AI supply chain is moving from scarcity to digestion. A move in the next 6-18 months will likely be driven by packaging capacity, advanced-node pricing, and whether cloud growth re-accelerates or normalizes.

The consensus is also missing that 13F buying is backward-looking and often reflects risk management rather than a fresh fundamental edge. If these names gap on continued hedge-fund ownership, the better signal is whether option-implied expectations and sell-side estimates keep rising; if not, the trade becomes a crowded-quality trap. Falsifiers: any guide-down in cloud growth, a visible capex slowdown, or evidence that AI revenue monetization is lagging infrastructure spend.

Key Decisions for Investors

• Prefer a relative-value long TSM / short NVDA pair over the next 1-3 months: TSM captures AI volume without the same product-cycle and competitive-pricing risk; risk/reward improves if hyperscaler capex stays elevated and GPU pricing normalizes.
• Add to AMZN only on post-earnings consolidation, not pre-event chasing: the stock needs confirmation that AWS growth is converting into FCF expansion; if cloud growth holds but margins compress, upside is limited to low-teens percent, while a miss could re-rate it sharply.
• Use GOOG as a quality hedge inside a tech book, but keep sizing modest until the next cloud print: if Google Cloud growth decelerates from the current pace, the market will likely compress the multiple faster than for AMZN because expectations are already elevated.
• For TSM, buy pullbacks rather than breakouts and use a 3-6 month horizon: the stock should outperform if AI capacity remains tight, but any sign of hyperscaler capex digestion or export-control headlines is a reason to trim quickly.
• No aggressive options chase here; if you want exposure, consider AMZN or GOOG call spreads 1-2 quarters out only after the next earnings revision cycle confirms durability. The risk/reward is better after implied volatility resets than before it.