A recent market analysis report highlighted an interesting phenomenon—the growth momentum of the US stock market is entirely monopolized by seven tech giants.

Let's look at some data. The S&P 500 earnings forecast for 2026 has gradually risen from a negative zone at the end of last year to about a 1% positive revision, which sounds decent. But behind this lies extreme divergence: the Magnificent 7(, or Mag7), have their 2026 EPS expectations climbing from near 0% to about 10%, while the remaining 493 stocks, excluding these seven, still face earnings revisions in the negative -2% zone. The gap between them has directly reached 12 percentage points.

This is no small discrepancy. It reflects the current "dual-speed" pattern of the US economy—AI prosperity is highly concentrated among a few tech giants, while the rest of the economy is still struggling.

**Within Mag7, also diverging**

Nvidia's performance is particularly eye-catching. Over the past 60 days, analysts have upgraded next year's EPS forecast by about 16%, with data center order backlog remaining strong. Alphabet's TPU-related revenue is expected to reach $13 billion in 2027, with unit forecasts sharply revised from an early 3 million units to 5 million(2027) and 7 million units(2028).

In comparison, Amazon and Meta, due to valuation reverting to historic lows( with expected P/E ratios of 30.7x and 21.9x respectively, have attracted more attention. Microsoft's earnings forecast has also been revised upward, but more moderately. Overall, Mag7 is expected to achieve nearly 23% profit growth in 2026, far exceeding the S&P 500's 12.3%.

**But here's the problem**

Valuations are already approaching the limit. The year-end expected forward P/E of 22x is similar to the current level, meaning that 12% of the return is almost entirely from earnings growth rather than valuation expansion. A 22x P/E significantly exceeds the long-term average)15-17x(. If earnings growth underperforms expectations, valuations have no downside buffer.

Even more concerning is that AI monetization is still in its early stages. The global AI adoption rate is only 16.3%, and while 78% of enterprises have adopted AI, most are still in pilot phases. If, by 2026, companies start questioning the ROI of AI investments, the $539 billion cloud computing expenditure forecast may be overly optimistic, and the 16% EPS upgrade for Nvidia has already priced in the continuation of this boom.

**Macroeconomic headwinds are also emerging**

Weak employment growth. In December 2025, only 50,000 new jobs were added, and in October and November, revisions lowered this by 76,000. The unemployment rate is expected to rise to 4.5% in 2026, with the labor force participation rate dropping to 53.8%, hitting a historic low since 1947. Per capita sales in department stores have plummeted by 70% from their peak.

The underlying logic is—AI prosperity is driven by capital expenditure from a few tech giants, while employment in other sectors is shrinking. The sustainability and inclusiveness of this "dual-speed economy" are truly questionable.

**Technical signals are also flashing red**

Only 64% of the S&P 500 components are above their 200-day moving average, 16 percentage points below the "solid bullish" threshold of 80%. The tech sector is even worse, with only 58% of its components above this line. Positions in certain commodities suggest they are at extreme statistical levels, indicating crowded trades and reversal risks.

**Impact on different sectors**

The short-term trend favors Mag7, especially Nvidia, Alphabet, Amazon, and Microsoft, but valuations are already overstretched, and risks of AI monetization slowdown are looming.

Data center REITs, utilities, copper and aluminum producers—those in the AI infrastructure supply chain—will benefit from the cloud computing expenditure forecast.

However, cyclical stocks within the S&P 493)—industrials, materials, energy(—still face earnings revisions of -2%, which do not support allocation unless a major rotation catalyst appears) such as an unexpected large rate cut by the Fed(.

The financial sector, although leading in market breadth, may see rising credit card default rates due to consumer financial pressures.

**Final recommendations**

The S&P 493 needs to wait until earnings revisions recover from -2% to at least 0% before considering rotation. The Mag7 story is not over, but warning signals are already flashing.