Microsoft: Not Cheap, But Shares Can Still Move Higher

MSFT remains one of the cleanest “own the picks and shovels” beneficiaries of the AI buildout, but the market is still underestimating how much of the value accrues to the infrastructure stack rather than the model layer. The company’s capex intensity is also a moat signal: if management can keep funding datacenters and AI capacity without impairing returns, smaller cloud peers are forced into a brutal choice between margin erosion and underinvestment. That should support MSFT’s share of enterprise AI spend even if overall IT budgets stay flat.

The more interesting second-order effect is competitive compression. Every incremental dollar MSFT pushes into AI infrastructure raises the hurdle rate for hyperscaler rivals and for software vendors trying to defend pricing without owning compute. Over the next 6-18 months, this likely widens the gap between “AI-enabled” and “AI-monetized” franchises, with MSFT sitting closer to the latter; vendors dependent on Azure capacity, enterprise productivity workflows, or partner ecosystems should see improved attach rates, while standalone software names risk value leakage.

The main risk is not execution, but expectation saturation. After a strong run in AI-related capex and cloud narratives, the stock can still underperform if investors begin discounting another 12-18 months of spend before incremental margin inflection shows up. A reversal likely needs either a meaningful deceleration in cloud growth, evidence that AI monetization is lagging capex by more than a few quarters, or a broad factor rotation back into cyclicals/value that compresses mega-cap duration-like multiples.

Contrarianly, the consensus may be too focused on the quality premium and not enough on the payback math: if capex continues to step up faster than free cash flow conversion, the market may eventually reward less-exuberant AI exposure elsewhere. That argues for owning MSFT as a core compounder, but not chasing it indiscriminately here; the better trade may be relative value versus lower-quality software and cloud names that are more exposed to AI hype without the balance-sheet firepower to sustain the buildout.