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

The article argues both Microsoft and Nvidia are trading at valuation discounts to the S&P 500, highlighting Microsoft at ~20x forward earnings versus the index’s ~21.7x and Nvidia at ~15.4x next-year earnings versus ~21.9x. Microsoft’s latest-quarter revenue rose 18% and diluted EPS increased 23% YoY, while its Copilot AI business is cited at $37B in annual recurring revenue growing 123% YoY and Azure at 40% growth. For Nvidia, next-year AI-related spending is expected to exceed $1T with Wall Street projecting 82% growth in FY2027 and 41% in FY2028, suggesting potential upside if sentiment/positioning catches up before 2026.

Analysis

The key market mechanism here is not “AI is good,” but that the market is still underpricing the durability of AI capex and the monetization path for the two names with the most direct claim on it. For Microsoft, the setup is a quality-growth rerating trade: if cloud and productivity AI keep compounding, the stock should gravitate toward a premium multiple versus the index, which would also support other large-cap software names as investors re-rate duration. The risk is that AI feature adoption is real but monetization is less elastic than the headline growth rate implies, capping multiple expansion.

For Nvidia, the second-order winner is the broader semiconductor and infrastructure stack: networking, memory, foundry, and power/cooling vendors benefit as long as hyperscalers keep pulling forward spend. The loser is any AI customer trying to defend margins while capex stays elevated; that pressure can show up months later in guidance cuts from cloud and software platforms that are forced to absorb higher infrastructure costs. The near-term catalyst remains the next round of capex commentary, not the current quarter’s earnings print.

The contrarian view is that both stocks may look “cheap” only because the market is already discounting peak AI enthusiasm and elevated rate sensitivity. If long-duration multiples do not expand, the stocks can still work on earnings alone, but the easy multiple upside is smaller than bulls assume. The thesis is falsified if Azure growth decelerates materially, or if hyperscaler capex guidance flattens and Nvidia’s forward growth estimates start coming down instead of up.

Key Decisions for Investors

• Long MSFT into the next earnings window via a 3-6 month call spread. Target a rerating toward a mid-20s forward multiple; cut if Azure growth or AI monetization comments disappoint.
• Long NVDA on any post-rally consolidation using 6-9 month calls or a call spread. The payoff is driven by second-half capex guidance; invalidate if hyperscaler spending commentary softens or export restrictions tighten.
• Pair trade: long NVDA / short AMD for the next 2-3 quarters. The cleaner AI platform moat and better demand visibility should keep NVDA outperforming unless AMD shows an unexpected supply or share-share inflection.
• Relative value: long MSFT / short QQQ or XLK into earnings. This isolates quality growth rerating while reducing index beta if the market de-risks.
• Watchlist alert, not a trade yet: semicap and infrastructure names such as ANET, AVGO, and TSM on any evidence that Nvidia demand is translating into broader supply-chain orders.