If a Stock Market Crash Is on the Way, Warren Buffett Says This Is the Best Thing Investors Can Do for Their Portfolio

The S&P 500 and Nasdaq have faced volatility after record highs as investors weigh rising bond yields and stubborn inflation; whether the sell-off becomes a larger correction remains uncertain. The article warns that AI's transformative potential does not guarantee strong returns for every company, recalling the Nasdaq's 78% decline from March 2000 to October 2002. It highlights Buffett's emphasis on durable competitive advantages, diversification, reasonable valuations, and staying invested; a hypothetical $10,000 invested in the S&P 500 near its dot-com-era peak grew to nearly $55,000 over roughly 25 years.

Analysis

The key transmission channel is not “AI versus no AI,” but the return on incremental AI capital. Persistent higher yields raise the hurdle rate on long-duration growth assets, while data-center depreciation and power/network costs can absorb more of the hyperscalers’ cash flow if monetization lags. That could pressure supplier expectations—including NVIDIA—even as near-term orders remain strong. Watch capex growth against disclosed AI-related revenue and cloud margins, rather than headline spending alone.

The contrarian point is that long-run index compounding does not resolve near-term entry-price or concentration risk. The article’s historical example supports staying invested, but is not evidence that today’s market valuation or AI earnings expectations are protected from a rates shock. Conversely, treating all AI exposure as a bubble misses that established platforms can fund investment from existing businesses.

Near term, yields and positioning can dominate fundamentals; over 1–3 months, earnings commentary on capex returns and cloud margins is the more useful catalyst. Over 6–18 months, power availability, infrastructure costs, and customer willingness to pay will distinguish durable beneficiaries from capacity suppliers exposed to a spending pause. Berkshire Hathaway may offer a relative-quality hedge, but it is still an equity and not a guaranteed rates or drawdown hedge.

Key Decisions for Investors

• Avoid a blanket short of AI. For a tactical relative-value hedge, consider long BRK.A versus short NVDA only if yields keep rising and AI capex-return evidence weakens; size to matched market risk, not equal dollars. Reassess if hyperscalers show improving cloud margins alongside sustained AI demand. This is a relative trade, not a claim that Berkshire is immune to a selloff.
• Use upcoming AMZN, GOOG, META, and MSFT results to track capex growth against cloud revenue, margins, and any quantified AI monetization. If spending accelerates without improving returns, reduce concentrated AI-infrastructure exposure; if returns improve, that weakens the bearish case for NVDA.
• Do not materially time broad-market exposure from this article alone. Rebalance concentration and liquidity risk rather than selling diversified holdings solely on volatility; no broad directional trade is warranted without valuation, positioning, and yield-path data.
• Falsifiers and alerts: a sustained decline in bond yields would ease valuation pressure; rising hyperscaler cloud margins and explicit evidence of customer-paid AI workloads would challenge the capex-ROI concern. Conversely, upward capex revisions paired with weaker margins or a spending pause would strengthen it.