Cloud Orders Surge 150% to $1.7 Trillion: JPMorgan Says AI Investment Is Paying Off, Tech Stock Valuations Present Investment Opportunity

For the past year, markets have questioned whether large-scale investments by major tech companies in artificial intelligence (AI) infrastructure would lead to 'high input, low return' risks. However, a recent report from JPMorgan indicates that surging cloud demand is turning AI capital expenditures into substantial orders and future revenues, suggesting that market skepticism over AI investment returns may be fading—and large-cap tech stocks could see valuation recovery.

JPMorgan analysis shows that hyperscaler cloud providers now have cloud business backlogs growing over 150% year-on-year, totaling approximately $1.7 trillion—significantly outpacing the roughly 80% growth in capital expenditures (CapEx) during the same period.

The report argues that order growth far exceeding CapEx suggests AI infrastructure investment is starting to yield higher potential revenue returns, countering earlier market concerns about overinvestment.

Mark Schilsky, JPMorgan analyst, stated that both cloud backlog and new annual recurring revenue (ARR) are consistently growing faster than CapEx, implying future revenues could adequately support current hundreds of billions of dollars in AI investments.

Second-quarter earnings season also revealed further signs of expanding demand. Amazon (AMZN-US) CEO Andy Jassy said during the earnings call that the company’s outlook for Amazon Web Services (AWS) long-term growth has become more optimistic.

Jassy noted that while Amazon previously believed AWS could grow into a multi-hundred-billion-dollar annual revenue business, it now expects at least double that scale, with potential to become a $1 trillion annual revenue business in the future.

He added that demand visibility through 2028 is already "stunning," and enterprise adoption of AI inference services remains in very early stages, indicating massive future growth potential.

Beyond Amazon, Microsoft (MSFT-US), Alphabet (GOOGL-US), and Meta Platforms (META-US) executives have recently conveyed similar signals, believing AI commercialization is still in an early expansion phase, enterprise market demand remains immature, and vast growth opportunities lie ahead.

Despite improving AI fundamentals, tech stock valuations continue to be suppressed.

JPMorgan notes that after July’s market correction, the S&P 500 Information Technology sector’s forward P/E ratio dropped to around 20x, nearing its lowest level in a year—falling into the 1st percentile of historical valuation ranges and below the 10-year average of approximately 23x.

Currently, large-cap tech stocks (excluding semiconductor firms) have forward P/Es more than two standard deviations below their 2018–2023 average.

If valuations rebound to just one standard deviation below the historical mean, related stocks could see about 30% upside; if they return to the long-term average, potential gains could reach approximately 56%.

Additionally, hyperscalers’ relative performance against the S&P 500 index has fallen to a three-year low. JPMorgan states that historically, such levels often precede strong mean-reversion rallies.

On fund allocation, Deutsche Bank (DB-US) data shows that despite improving earnings and forecasts for large tech firms, institutional investors maintain only modest overweight positions—far below holdings seen during previous strong earnings growth cycles.

Meanwhile, market capital flows this year have concentrated heavily in semiconductor stocks, leaving large-cap tech stocks (excluding semiconductors) significantly under-allocated.

JPMorgan believes that if market focus shifts from 'Is AI CapEx too high?' to 'AI investment returns are materializing,' the next leg of tech stock gains may be led by large tech platforms rather than continuing to be driven solely by AI chip stocks.

Technically, the Roundhill Magnificent Seven ETF (MAGS-US), which tracks the seven largest U.S. tech giants, has rebounded nearly 10% from recent lows, reclaiming its 200-day moving average and approaching the long-term uptrend line since April last year.

JPMorgan points out that the 200-day moving average is gradually flattening, indicating prolonged consolidation. Historically, the longer the consolidation, the more explosive the breakout tends to be—making future tech stock performance worth continued investor attention.