Data center capital expenditures will grow by more than 50% next year! JPMorgan: Earnings expectations for AI-related stocks are underestimated.

```

As the market worries about a decline in AI capital expenditures, JPMorgan has significantly raised its forecast for data center capital expenditure growth, believing that the market seriously underestimates the profit potential of AI-related stocks.

According to Zhuifeng Trading Desk, JPMorgan’s North America Equity Research team stated in its latest report that the arms race among large tech companies is not slowing down, but is actually accelerating.

According to the report, the growth rate of data center capital expenditures is expected to exceed 50% in 2026, a significant increase from the previous forecast of 30%. This means that next year there will be more than $150 billion in new incremental spending, basically covering the expected revenue of core AI chip manufacturers such as Nvidia, AMD, Broadcom, and Marvell.

More importantly, due to ongoing serious lags in AI computing power supply compared to demand, JPMorgan believes the actual growth rate could reach over 60%, bringing unexpected performance growth opportunities to AI infrastructure-related companies. The market’s current profit forecasts for core AI industry chain stocks remain too conservative.

Substantial Upward Revision to Capital Expenditure Forecast: Explosive Growth Expected in 2026

The latest analysis from JPMorgan’s IT hardware team shows that the 2025 data center spending growth rate has been significantly raised from the previously forecast 55% to about 65%. The driving force behind this adjustment is hyperscale cloud service providers continuing to ramp up AI infrastructure construction, and the fundamental reason is that available AI computing power supply still falls far short of market demand.

In absolute terms, incremental spending in 2025 compared to 2024 will exceed $115 billion, significantly higher than the 2024 year-on-year increment of $75–80 billion. It is worth noting that this $115 billion in incremental spending already basically covers the combined incremental AI-related GPU/ASIC and network device revenue expected this year by Nvidia, AMD, Broadcom, and Marvell (over $85 billion).

What’s even more noteworthy for investors is the outlook for 2026. JPMorgan now expects the growth rate of data center capital expenditures in 2026 to exceed 50%, a substantial rise from the previous 30% forecast. At this rate, incremental spending in 2026 will surpass $150 billion.

Historical data show that upward revisions to capital expenditure growth forecasts during a year are the norm. Both 2024 and 2025 show this trend: as time progresses and visibility improves, the market’s expectations for capital expenditure growth in future years have been consistently revised upward. JPMorgan believes that 2026 will repeat this pattern. As confidence in continued growth increases, the market's capital expenditure expectations for 2026/2027 will likely continue to climb.

In terms of the total capital expenditures by the four major U.S. hyperscale cloud service providers, the forecast for 2025 is about $363 billion, a year-on-year growth of roughly 65%; for 2026 it is expected to reach about $447 billion, continuing strong growth momentum.

Wall Street’s Blind Spot: Underestimated Chip Revenue

For core AI infrastructure suppliers such as Nvidia, Broadcom, AMD, and Marvell, this means consensus analyst forecasts have lagged behind reality. JPMorgan believes that the market’s current revenue projections for these companies for 2026 have not fully reflected the upcoming $150–175 billion in new capital expenditures.

“Given the strong (and urgent) demand we are seeing for AI computing, we would not be surprised if data center capital expenditures in 2026 ultimately reach a year-on-year growth range of over 60% ... This would bring AI revenue upside that is not yet included in Wall Street's forecasts.”

Simply put, when capital expenditures turn into orders, Wall Street will have to again be forced to raise profit forecasts for these chip giants.

Order Backlogs and Overlooked “Non-Traditional” Buyers

The report specifically mentions the market’s misreading of the backlog order values for Broadcom and Nvidia. For example, regarding Broadcom, JPMorgan believes that buyers have seriously misinterpreted management's comments about the "18-month $73 billion backlog", underestimating the speed with which it will convert into actual revenue. More critically, investors’ focus often remains limited to the four big U.S. cloud firms, while ignoring other rising forces.

Merely focusing on the spending of the top 4 or 5 U.S. hyperscale enterprises does not account for major spending happening outside traditional hyperscale enterprises, including the increasingly influential neoclouds (new cloud service providers) and sovereign AI projects as they scale up.

This means that, aside from the well-known tech giants, sovereign nations and emerging cloud platforms are becoming new pillars supporting AI chip demand, which will provide the industry with a longer and more intense growth runway than expected.

~~~~~~~~~~~~~~~~~~~~~~~~

The above excellent content comes from Zhuifeng Trading Desk.

For a more detailed interpretation, including real-time analysis and frontline research, please join [Zhuifeng Trading Desk · Annual Membership]

Risk Warning and DisclaimerThe market has risks, and investments should be made cautiously. This article does not constitute personal investment advice, nor does it take into account the special investment objectives, financial situations, or needs of individual users. Users should consider whether any opinions, views, or conclusions herein are suitable for their own circumstances. Any investment based on this is at your own risk. ```