2 Tech Stocks Down More Than 20% That I Think Are Coiled Springs Right Now

Microsoft and Oracle are presented as attractive pullback opportunities despite AI spending concerns, with Microsoft Cloud revenue up 29% to $54 billion, Azure up 40%, and Copilot seats above 20 million. Oracle’s cloud infrastructure revenue rose 84% to $4.9 billion, multicloud database demand surged 531%, and remaining performance obligations jumped 325% to $553 billion, supporting raised fiscal 2027 revenue guidance of $90 billion. Both stocks are down sharply from highs—Microsoft 23% and Oracle 45%—but the article argues fundamentals remain strong and valuations look compelling.

Analysis

The market is treating AI capex as a margin problem, but the more important second-order effect is that cloud scarcity is still the binding constraint. If Microsoft is capacity-constrained while demand remains elastic, then near-term revenue is being capped, not impaired; that creates a setup where incremental data center spend can convert directly into revenue acceleration once supply catches up. The same dynamic likely benefits the entire AI infrastructure stack, but especially power, networking, and hyperscale suppliers that can remove bottlenecks faster than software competitors can displace share.

Oracle’s setup is more fragile but also more asymmetric because the equity is now functioning like a levered call option on utilization of fixed infrastructure. The debt load is only a problem if contracted demand fails to convert into actual throughput; the key tell is whether remaining performance obligations keep compounding faster than interest expense and depreciation. If that holds, the market may be mispricing a refinancing story as a solvency story, which usually resolves sharply once investors see the capex peak approaching.

Consensus is probably underestimating how sticky enterprise AI spend becomes once workflows are embedded in identity, security, and database layers. Copilot adoption is not just a product metric; it raises switching costs across Microsoft’s stack and makes future monetization more defensible than a pure application-layer AI vendor. For Oracle, the contrarian angle is that contracted backlog matters more than headline leverage in a capacity-constrained phase, and the stock may re-rate well before free cash flow visibly inflects.

The main risk is not competition, but a demand-air-pocket in AI infrastructure spending over the next 2-4 quarters: if customers digest current deployments or financing conditions tighten, the market will punish the most levered buildout names first. That said, the downside in Microsoft looks more time-based than thesis-based, while Oracle’s downside is tied to execution and funding spread risk. A weaker macro tape could delay multiple expansion, but it would not erase the long-duration optionality embedded in both names.

Key Decisions for Investors

• Add to MSFT on weakness with a 3-6 month horizon; the risk/reward favors owning the highest-quality AI beneficiary while the market is still discounting capex as permanently dilutive. Use a staged entry rather than full size given sentiment is still fragile.
• Initiate a pairs trade: long MSFT / short a software basket or high-multiple SaaS names over 1-2 quarters. The thesis is that infrastructure winners with proven monetization will outperform software names whose AI disruption risk is still mostly narrative.
• Buy ORCL only as a smaller, higher-beta satellite position or via call spreads expiring 6-12 months out. The stock can rerate quickly if backlog converts, but leverage makes it vulnerable to any slowdown in cloud demand or a credit spread shock.
• Prefer long NVDA and selected power/networking suppliers over ORCL if the goal is AI capex exposure with less balance-sheet risk. The cleaner expression is to own picks-and-shovels beneficiaries rather than the levered financing vehicle.
• Set a tactical risk trigger: reduce ORCL exposure if backlog growth decelerates materially or if management language shifts from 'capacity constrained' to 'demand normalization.' That would be the earliest signal that the market is right to focus on leverage rather than growth.