The Brutal Math

In September 2026, roughly 30% of the Nifty 50's value is concentrated in a handful of AI-linked mega-cap stocks. That's higher concentration than the dot-com bubble itself. Price-to-sales ratios for leaders like Nvidia and Broadcom touched 30x—levels that historically preceded sharp drawdowns. Tech volatility has resurged to bubble-era levels. By every rear-view-mirror metric, this looks precarious.

But here's the twist: earnings expectations have been revised upward for 2026, not downward. Analysts now expect 32% earnings growth this year—double their forecast from January. That's because AI infrastructure spending is actually driving return on investment. Microsoft disclosed ₹3,378 in AI capex in Q3 alone. Meta is spending 50% more on data centers than it did in 2025. These aren't wasted investments; they're being monetized through subscriptions, cloud services, and advertising.

The dot-com bubble burst because companies with zero revenue and no path to profitability were trading at absurd valuations. Nvidia, Meta, and Microsoft have revenue. They have margin expansion paths. The question isn't whether AI is transformative—it clearly is—but whether current stock prices reflect that transformation or overextend it.

Where the Consensus Breaks

The Bull Case: Earnings growth justifies valuations. A new technology cycle (like electricity, the internet, mobile) can sustain high multiples for years. Concentration risk is overstated; smaller companies, value stocks, and international markets have outperformed in 2026, not underperformed. Rate cuts later in 2026 or 2027 would re-rate stocks upward.

The Bear Case: Valuations at 2000 levels usually precede 40-50% corrections, regardless of earnings. Higher interest rates (Fed just hiked in September) reduce the present value of future earnings. Capex spending is circular—Meta and others are investing heavily to defend market positions, not to grow new ones. A slowdown in capex growth translates immediately to stock weakness. Broadcom's weak guidance in September spooked the market because it suggested capex cycle saturation.

The Nuanced View: This isn't a classic bubble of hype and no fundamentals. It's a valuation bubble where the price already assumes aggressive earnings growth and no recession. If growth disappoints, or if capex cycles slow faster than expected, 15-20% corrections are reasonable. But a 50% crash requires earnings estimates to be wildly wrong—not impossible, but not the base case.

Why India Investors Should Care

Indian investors are exposed to this through TCS, Infosys, and other IT exporters that serve US tech companies. If the US market corrects, capex budgets tighten, and consulting demand softens. Conversely, if AI capex accelerates, Indian IT stands to gain disproportionately (offshore development, testing, infrastructure support). The outcome hinges on the next 6 months of earnings reports.

For domestic Indian stocks, the risk is indirect: a US correction weakens emerging market inflows and the rupee, raising import costs and inflation. RBI tightening could follow.

The Practical Action Plan

For Concentrated Tech Positions: Take profits on positions that have doubled or tripled. Rebalance into names with lower P/E ratios and clearer near-term cash flow (financial services, utilities, consumer staples).

For IT Services Investors: Monitor capex guidance from US tech firms. If it stays elevated, hold. If it flattens, trim exposure.

For Defensive Investors: Build positions in gold, rupee-denominated bonds, and non-cyclical consumer stocks. The next 12 months will test the valuations thesis.

Bottom Line

The bubble is real, but it's priced more richly than the headlines suggest. Markets are "wary," not euphoric. The next correction, when it comes, will feel sharp but may not be a crash. Investors who panic-sell at the first 15% pullback will regret it. Those who rebalance carefully now will be better positioned either way.