2 Magnificent Artificial Intelligence (AI) Stocks Down 16% to 28% to Buy Hand Over Fist

Microsoft is described as trading at a valuation discount (20x vs the S&P 500’s ~21.7x forward earnings) despite strong momentum, including Copilot generating ~$37B in annual recurring revenue (+123% YoY) and Azure growing ~40%, alongside revenue up 18% and EPS up 23% in the latest quarter. Nvidia is pitched as even cheaper on forward multiples (15.4x next-year earnings vs ~21.9x for the benchmark) while AI capex is expected to exceed $1T next year, with Wall Street projecting 82% FY2027 and 41% FY2028 growth. Overall, the article frames the setup as a time-sensitive “buy on the dip” opportunity for both names as markets re-rate AI winners before 2026.

Analysis

The market is effectively asking whether AI spend is still a growth engine or becoming a balance-sheet tax. For NVDA, the key issue is not demand existence but conversion of that demand into durable EPS revisions; as long as hyperscaler capex keeps stepping up, the stock should remain the cleanest lever on the AI buildout. The second-order winner is the adjacent supply chain — TSM, AVGO, and networking names — while the first-order losers are AI-adjacent software names that are spending heavily on copilots without clear monetization.

MSFT looks cheaper for a reason: the market is discounting the risk that AI contributions are being offset by rising infrastructure intensity. If Azure and Copilot continue to grow, MSFT can re-rate, but the burden of proof is now higher because any slowdown in cloud growth or margin dilution would keep the multiple pinned near market. The likely catalyst window is the next 1-2 earnings prints, not a 12-month narrative shift; if capex guides from the mega-cap buyers soften, the AI trade could de-risk quickly.

Contrarian take: consensus may be underestimating how long AI capex remains elevated, but it may also be overestimating how quickly that spending accrues to software monetization. The most important falsifier for NVDA is a sequence of guide-downs in hyperscaler spending or gross margin pressure; for MSFT, it is Azure growth decelerating into the low-30s or Copilot failing to move the needle on ARPU. This is more a relative-value and earnings-revision setup than a broad market call.

Key Decisions for Investors

• Go long NVDA on any 3-5% pullback into the next hyperscaler capex update cycle; 1-3 month horizon, with upside driven by forward EPS revisions, stop if gross margin or booking commentary disappoints.
• Pair trade: long NVDA / short AMD for a 2-4 month relative-value trade; thesis is that NVDA captures the next dollar of AI spend more directly, while AMD remains more exposed to share-gain expectations already embedded in the stock.
• Treat MSFT as a selective add, not a chase: buy only on confirmation of sustained Azure acceleration in the next quarter; if Azure slips below the mid-30% growth range, reassess as the valuation gap to the S&P may persist.
• Watch for signs of AI capex digestion from MSFT/AMZN/GOOGL earnings; if capex growth flattens while NVDA stays expensive, consider reducing long beta and rotating into TSM/AVGO instead.