AMZN, GOOGL, MSFT Attractively Valued? David Bailin's AI Bull Case

The market is still pricing these platforms as if AI is mainly an EPS timing issue, when the bigger effect is a multi-year reinvestment cycle that can widen moats. For AMZN, MSFT, and GOOGL, the first-order hit is visible in cash flow optics, but the second-order gain is that they can finance the infrastructure race internally while smaller competitors have to rent capacity at worse economics. That tends to pull share toward the incumbents and, downstream, benefits constrained suppliers in power, networking, and data-center infrastructure more than the article implies.

The key distinction is between accounting drag and economic return. If AI workloads lift cloud attach rates, enterprise spending, and ad/search monetization, the payoff shows up with a lag; if not, the same capex will start to look like margin dilution and multiple risk. Over the next 1-3 months, the market will care less about narrative and more about capex commentary, cloud growth inflection, and whether depreciation is outpacing revenue leverage.

Contrarian risk: consensus may be underestimating how long the spend cycle can stay elevated, but it may also be overestimating near-term monetization. The thesis weakens if cloud growth decelerates, AI utilization remains internal rather than customer-facing, or if management teams start signaling capex normalization. In that case, the winners shift from the hyperscalers to the picks-and-shovels layer, while the big platforms underperform on FCF revisions before the long-term AI story is disputed.

The cleanest setup is not to chase the names after a bullish commentary print, but to use weakness to express the view that scale wins. If investors keep treating AI as a cost item, the opportunity is in buying the balance-sheet winners on drawdowns and avoiding lower-quality software names that need AI to justify their valuation without having the distribution to monetize it.