The clearest signal on whether artificial intelligence spending is starting to generate returns, as evidenced on Wall Street this week, is to examine cloud computing revenue and stock price performance.
Following the release of Microsoft's earnings, its shares surged more than 13% in a single day, gaining a cumulative 19% for the week, rising from $389 to over $460. Amazon.com climbed over 15% in the same period, moving from $231 to $271. Alphabet rose 7% this week, with its stock price increasing from $326 to $356, as the market reassesses the commercial value of its AI driven by accelerating cloud business. Oracle also saw a boost from cloud computing expectations, gaining over 10% in the same period.
However, the market is not indiscriminately rewarding all AI-related plays. Meta Platforms, Inc. fell 6% this week, dropping from $593 to $556, due to its continued heavy investment in AI infrastructure while lacking a direct monetisation channel like cloud computing. NVIDIA, the core supplier of AI computing power, only rose 2% for the week, with its stock trading in a range between $195 and $200 as investors await its August 26th earnings report to validate the next cycle of AI capital expenditure.
A clear divergence is forming in this market movement: investors are no longer just rewarding the scale of AI spending, but are instead favouring companies that can prove their AI investments are translating into revenue, profit, and cash flow.
Cloud Business Surge: The Path to AI Commercialisation Becomes Clearer
The second-quarter earnings reports from major tech companies sent a clear signal: AI investment is transitioning from a capital expenditure phase to a revenue realisation phase.
Revenue from Amazon.com's AWS, Microsoft's Azure, and Alphabet's cloud business all significantly exceeded market expectations in the second quarter. AWS revenue grew 37% year-over-year, Azure grew 43%, and the combined cloud revenue for the three companies increased 48% year-over-year.
Amid ongoing debate over AI business models, cloud computing has emerged as the most easily verifiable path to monetisation. The giants build data centres, purchase chips and computing equipment, and then provide these computing resources to enterprise clients through long-term contracts, converting infrastructure investment into cash flow returns.
AWS's second-quarter operating profit margin reached 39%. CEO Andy Jassy stated that the average payback period for computing equipment is less than three years, while most AI computing customer contracts extend beyond five years, indicating that current investments have a long earnings realisation cycle. "The resulting revenue, free cash flow, and return on invested capital are very compelling," Jassy said, outlining a long-term vision for AWS to potentially become a $1 trillion annual revenue business. For comparison, FactSet data shows analysts currently expect AWS to generate approximately $170 billion in revenue this year.
AI demand is becoming the core driver of cloud business growth. The current industry bottleneck has shifted from "is there demand?" to "is there enough computing power?" Amazon.com's April agreement with Anthropic, a cooperation deal worth over $100 billion over ten years, is a testament to the explosive demand for AI computing power.
Meta's Dilemma: Heavy Investment Lacking a Cloud Outlet
Compared to Microsoft, Amazon.com, and Alphabet, Meta Platforms, Inc. faces a more significant market test.
This social media giant is also heavily investing in AI infrastructure but lacks a mature cloud computing business. It cannot directly generate revenue by leasing computing power like AWS, Azure, and Google Cloud.
Since its second-quarter earnings report, Meta Platforms, Inc.'s stock price has fallen approximately 5%. The company previously raised its 2026 capital expenditure expectations. CEO Mark Zuckerberg indicated that Meta Platforms, Inc. is considering building its own cloud computing business but currently lags far behind the three major cloud providers.
Wall Street is forming an increasingly clear evaluation standard: the key to AI investment is not the scale of spending, but whether a company has a business model that can convert capital expenditure into sustainable cash flow. Cloud computing is currently the most mature answer, and companies lacking this commercialisation path are facing more rigorous scrutiny.
Earnings Quality Becomes the Core Test for Tech Stocks in the Next Phase
Despite strong earnings performance from tech giants, market reactions to their reports have been divergent.
Goldman Sachs data shows that within the TMT sector, companies that beat EPS estimates underperformed the S&P 500 by an average of approximately 192 basis points on the day following their earnings release. In contrast, similar non-TMT companies outperformed the S&P 500 by an average of about 75 basis points.
This indicates that the market's pricing logic for tech stocks is changing: investors are no longer focused solely on whether short-term results beat expectations but are placing greater importance on the sustainability and source of earnings growth.
The tracked year-over-year EPS growth rate for the S&P 500 in the second quarter reached 45%. Of this, approximately 19 percentage points came from "other income" related to equity investments totalling roughly $151 billion from Alphabet and Amazon.com, with Microsoft contributing another approximately $3 billion. Excluding these factors, core earnings growth still reached 26%, the fastest pace since 2021.
Currently, companies related to AI infrastructure have contributed roughly one-third of the S&P 500's second-quarter EPS growth. Analysts expect this proportion could exceed 50% by the second half of 2026 and 2027.
Nevertheless, cost pressure remains a key risk for the market. Over the past few quarters, the net profit margin for the median S&P 500 company has essentially stagnated, and profit margin expectations for most industries have been revised downwards.
Goldman Sachs maintains its year-end target of 8,000 points for the S&P 500 and projects S&P 500 EPS to reach $385 in 2027. For the AI-driven market narrative, the focus will shift from the pace of capital investment to the sustainability of earnings growth.
Comments