Amazon, Alphabet, and Microsoft All Delivered Soaring Cloud Growth Thanks to AI. Here's the 1 I'd Buy Right Now

Q2 results show all three hyperscalers are accelerating AI-driven cloud demand—AWS sales +37% YoY to $42.2B (fastest in 18 quarters), Azure +43% YoY, and Google Cloud +82% YoY to $25B—while capex ramps materially (Amazon to $220B; Alphabet capex forecast to $200B). However, free cash flow turned negative for Amazon (-$7.6B) and Alphabet (-$5.8B) on higher capex, while Microsoft generated $19.6B FCF and expects to stay free-cash-flow positive in fiscal 2027. The article flags Microsoft as the value outlier at ~21x next-year earnings versus ~25–26x for Alphabet and Amazon.

Analysis

The signal is less “AI demand is real” and more “capital intensity is now the gating variable.” That favors the hyperscaler with the best self-funding profile, because it can keep buying capacity without forcing a re-rating on balance-sheet strain or FCF dilution. In that frame, MSFT deserves a premium relative to AMZN and GOOG only if investors believe it can convert spend into durable operating leverage faster than peers; otherwise, the multiple gap should narrow toward the cheapest FCF-adjusted name, not necessarily the fastest-growing cloud line.

Second-order beneficiaries sit outside the article: GPU/accelerator vendors, networking, and power/cooling chains should keep seeing demand even if hyperscaler stock reactions fade. The bigger loser is any software or IT budget bucket that gets crowded out as cloud providers prioritize infrastructure over downstream monetization; if AI workloads remain concentrated in a few hands, pricing power migrates to the infrastructure layer first, not the application layer. That dynamic also means AWS and Google Cloud can keep taking share while still underperforming on shareholder returns if capex absorbs the gross profit.

The contrarian risk is that the market may be over-trusting the notion that “positive AI demand” automatically justifies open-ended capex. Over the next 1-3 months, watch for any capex guide raise without a commensurate revenue or operating income inflection; that is usually when stocks de-rate. Over 6-18 months, the thesis breaks if MSFT’s FCF turns negative or if AMZN/GOOG show enough FCF conversion to remove the valuation advantage.

Key Decisions for Investors

• Long MSFT / short AMZN or GOOG on a 1-3 month horizon: MSFT has the cleaner FCF profile and should hold its multiple better if AI capex remains elevated; use any post-earnings pullback to enter, and cover if AMZN/GOOG demonstrate sustained FCF inflection.
• Buy MSFT on weakness, but size it as a quality-growth compounder rather than a momentum trade: upside is relative multiple resilience, not a near-term blowout in earnings; thesis is invalidated if FY27 free cash flow guidance turns negative.
• Watch-list long NVDA (or semis/networking basket) versus the hyperscaler equity basket: the capex spend is a demand transfer to infrastructure suppliers, and the trade should work as long as data-center buildouts stay ahead of monetization.
• Avoid chasing AMZN/GOOG after strength unless next-quarter guidance shows capex plateauing or FCF recovery; otherwise the risk/reward is capped by ongoing cash burn and higher scrutiny on ROIC.