Is It Really Smart to Buy Stocks Right Now? Here's Warren Buffett's Best Advice.

The article flags stretched U.S. equity valuations, citing the Buffett indicator at an all-time high of 238% and the Shiller CAPE near dot-com levels circa 2000. It warns that the current AI cycle mirrors the dot-com era, with the “Magnificent Seven” expected to spend well over $1T on capex in 2025-2026. While Buffett advises long-term ownership and buying wonderful companies at reasonable prices, the near-term setup is described as resembling a “gambling mood,” implying higher risk of pullbacks from elevated prices.

Analysis

This is less a sell signal than a dispersion signal. When the market is priced for perfection, the next leg is usually a widening gap between cash-generative compounders and long-duration stories that need uninterrupted multiple expansion to justify today’s prices. That favors BRK.B, KO, and selectively GOOGL: all three can compound without needing heroic top-line assumptions, and BRK.B has a built-in volatility bid through its cash optionality when sentiment cracks.

The vulnerable pocket is the AI supply chain and any index basket that is implicitly short valuation discipline, especially QQQ/SMH-adjacent exposures. If hyperscalers keep spending but monetization lags, the first break is likely not earnings collapse but capex scrutiny: order growth slows at the margin, gross margins compress in the semicap ecosystem, and the market starts discounting a lower terminal growth rate for NVDA and its suppliers. That is a 1-3 month repricing risk; the full 6-18 month risk is a regime shift where AI is treated like prior infrastructure booms and multiples compress even if revenues keep rising.

Contrarian take: the consensus is too eager to equate expensive with imminent downside. In a liquidity-supported market, valuation can stay elevated longer than fundamentals say it should, especially if rate cuts or softer inflation extend duration assets. The better framing is that the market is not uniformly expensive; it is expensive in the crowded, narrative-heavy names and still reasonable in durable cash compounders. Falsifiers: a clean disinflation/rates rally that lifts mega-cap growth breadth, or another quarter of AI spend that converts into measurable monetization and operating leverage rather than just capex.

Key Decisions for Investors

• Pair trade: long BRK.B / short QQQ for 3-6 months. Thesis is lower-beta cash optionality versus index multiple risk; target 8-12% relative outperformance if breadth narrows. Invalidate if QQQ breadth improves and rate cuts re-ignite multiple expansion.
• Relative value: long GOOGL / short NVDA on a 6-12 month horizon. GOOGL has better self-funding and optionality if AI monetization is real; NVDA is more exposed to capex digestion. Falsify if NVDA guidance re-accelerates and hyperscaler orders stay linearly up.
• Overweight KO versus the broad market as a defensive ballast. It should hold up better if the market derates from the top down; this is more a portfolio defense than a high-conviction alpha trade.
• Buy 3-6 month QQQ put spreads only on a vol spike or after a failed breakout in the index. Use as a hedge against valuation compression rather than an outright crash bet; risk/reward is best when implied vol stays below realized vol.
• Set an alert on SMH if AI capex commentary turns from expansion to optimization. That would be the earliest signal that the market is transitioning from 'spend at any price' to return-on-capital discipline.