Antonio Menditto
**From AI Hype to Earnings Reality** Last week offered a useful reminder that markets don't move in straight lines. The **Nasdaq fell 4.6%**, the **S&P 500 lost 1.9%**, while several AI leaders—including **NVIDIA (-8.6%)**, **Alphabet (-8.3%)** and **Apple (-4.8%)**—experienced sharp pullbacks despite no meaningful deterioration in the long-term AI story. To me, this isn't a sign that the AI cycle is ending. It's a sign that the market is becoming more disciplined. Over the past two years, a significant part of equity returns has come from **multiple expansion**—investors were willing to pay increasingly higher prices for future growth. Today, that dynamic is changing. Valuations remain demanding. The Nasdaq trades at roughly **33x trailing earnings and around 24x forward earnings**. Those multiples can certainly be justified, but only if earnings continue to grow fast enough to support them. This is why companies like **Microsoft** have become so interesting. Microsoft continues to invest tens of billions of dollars in AI infrastructure, Azure and Copilot. In the short term, these investments put pressure on margins, leading some investors to question whether returns will materialize quickly enough. But if AI adoption accelerates across enterprises, Microsoft is arguably one of the best-positioned companies to convert those investments into recurring, high-margin revenue over the coming years. The same logic applies across the sector. **Micron's latest results confirmed that demand for AI memory remains exceptionally strong**, yet semiconductor stocks still corrected. That's what happens when expectations become extremely high: good news is no longer enough—companies must consistently exceed expectations. I also found it interesting that several analysts now describe the market as moving from **FOMO (Fear of Missing Out)** to **earnings discipline**. Investors are no longer buying every AI-related company indiscriminately. Instead, they're rewarding businesses that combine strong execution, sustainable earnings growth and reasonable valuations. Outside technology, I continue to like sectors with long-term structural tailwinds, particularly **healthcare**, where companies such as AstraZeneca, Roche and Medtronic offer resilient business models, and selected **European industrial and infrastructure companies**, which generally trade at more attractive valuations than many U.S. mega caps. For long-term investors, I believe the key takeaway is simple: **The AI revolution hasn't changed. The market has.** Going forward, stock selection will likely matter more than ever. The winners won't necessarily be the companies with the most exciting stories—they'll be the ones that consistently turn innovation into earnings and cash flow.
Not investment advice. The author may have financial interests in the mentioned instruments.
null
.