This week, Google's parent company Alphabet experienced its first quarterly negative free cash flow since its initial public offering, completely shaking the market. The massive expansion of AI computing power investment has overwhelmed one of the world's most profitable tech companies, and the pressure on future capital consumption will only increase further.
Next week, Microsoft, Meta, and Amazon will release their earnings reports in quick succession. As the performance of AI-related stocks gradually diverges, the market dynamics are reminiscent of the final stage of the tech and internet bubble era, with investors constantly worried that AI investments may not deliver substantial returns.
Intensifying Competition
Even though Google Cloud, which rents out AI computing power, saw revenue surge 82% year-over-year, hitting its fastest growth rate ever, Alphabet's second-quarter free cash flow result of negative $5.9 billion took the market by surprise. The company raised its full-year 2026 capital expenditure guidance by $15 billion and warned that capital spending will continue to increase in 2027, indicating that the cash drain from AI infrastructure will only worsen.
The shift to negative free cash flow is one of the clearest signals that AI is reshaping the business models of major tech companies. Previously, the market favored tech giants because of their ultra-high profit margins and steady cash flows, which were sufficient to independently fund new business initiatives. Now, major cloud companies are forced to raise funds through debt issuance or stock offerings, with total industry AI capital expenditure expected to exceed $700 billion in 2026. Their own operational cash flows are no longer enough to cover these investments. This sets the stage for a stringent review of next week's earnings reports from Microsoft, Meta, and Amazon.
Stock prices have collectively fallen, reflecting widespread investor concerns that other tech giants are likely to follow Google's lead in raising AI capital expenditure expectations, while the pace of AI business returns is far behind the rate of investment.
Aggregated forecasts from institutions show that Google and Amazon's full-year free cash flow will remain negative in 2026. Meta's full-year free cash flow is expected to shrink by 95.7%, leaving only $1.85 billion. For Microsoft, whose fiscal year ends next June, full-year free cash flow is projected at $25.39 billion, less than half of the $58.74 billion reported in the previous fiscal year.
"The risk of further upward revisions to capital expenditure remains elevated, especially for companies like Microsoft, which face ongoing shortages of computing capacity," said Charu Chanana, Chief Investment Strategist at Saxo Markets. "But investors will increasingly focus on how much more cash companies need to invest to maintain competitiveness, and whether AI-related revenue growth can outpace the combined costs of capital expenditure, equipment depreciation, and operating expenses."
Meanwhile, industry competition is set to intensify further. Meta is in talks to lease its own computing power to AI startup Anthropic. With the entry of AI cloud providers like CoreWeave, the competitive landscape is becoming more crowded. "As computing capacity keeps increasing and the cost of developing large models continues to decline, various cloud computing services are becoming more commoditized," said Lale Akoner, Global Strategist at eToro. "This will force all cloud providers to continuously increase their investments while accepting lower profit margins."
Will History Repeat Itself?
Some analysts are warning that the market may face a more dangerous wave of collective sell-offs.
The current market dynamics are reminiscent of the final stages of the internet bubble. This week, both Google and Tesla announced increases in AI capital expenditure, triggering a new round of sell-offs that wiped out over $200 billion in market value.
The expansion of AI capital expenditure benefits Asian chip makers like SK Hynix and Samsung Electronics, confirming this year's market trend: The tech giants spending heavily on AI see their stock prices pressured, while chip and hardware companies receiving AI orders continue to rally. The MAGS ETF, which tracks the "Magnificent Seven" tech stocks, has risen only about 1.5% year-to-date. In contrast, the Philadelphia Semiconductor Index (SOX) has surged over 70% this year, highlighting a stark divergence in performance.
Jason Hunter, a technical strategist at JPMorgan, has issued a new risk warning regarding the extreme divergence within the AI sector. He suggests that if major cloud companies (the supercomputing giants) fail to break through key technical resistance levels in the short term and face renewed selling pressure, while the semiconductor index remains constrained by a core resistance line, the current capital rotation within the AI sector could escalate into a more dangerous, full-scale tech-sector crash.
He notes that during the later stages of the internet bubble, the market saw an identical split in performance between the two major tech sectors. Internet companies providing end-user communication services saw their stock prices stagnate, while infrastructure and equipment stocks continued to rise strongly, creating a false impression of a robust market. This mirrors the current situation: The cloud giants (the "Magnificent Seven") at the forefront of the AI boom are experiencing weakening free cash flow and stagnant stock prices, while chip, computing infrastructure, and optical communication hardware companies continue to climb.
"There is a highly similar risk logic here. In a healthy bull market, the performance of backend infrastructure should lag behind that of leading front-end companies. But the current market trend is completely reversed, and this is something we must be highly vigilant about," Hunter cautioned.
Comments