The Q2 2026 earnings season has officially kicked off. This is not just another ordinary earnings cycle - it may be the most important and decisive earnings season since the rise of the artificial intelligence wave.
Over the past year and a half, the AI-driven technology stock frenzy has pushed the US stock market to new highs, with S & P 500 index Constituent Stock Quarter 1 profits increasing by 28.8% year-on-year. However, the tension between high valuations, huge capital expenditures, and profit sustainability has reached a critical point. Just this week, US technology stocks encountered a "Black Tuesday": the Nasdaq Composite Index fell 2.21%; the Philadelphia Semiconductor Index plummeted 7.87% in a single day, the largest drop since June 5th.Micron Technology fell more than 13%, ARM fell more than 10%, Mywell Technology fell more than 9%, Qualcomm fell more than 8%, ASML fell nearly 8%, Intel fell more than 6%, Chaowei Semiconductor fell more than 5%, and Broadcom fell more than 3%. The trigger for this sell-off was the news that SK Hynix slowed down the expansion of HBM4 production capacity and tilted resources towards general DRAM. A fuse ignited the valuation anxiety of the entire AI hardware sector.
At this critical moment, the Q2 financial report will answer a core question: Can the "spending story" of AI continue?
This article breaks down the five core points that must be paid attention to in this financial report season from five aspects: total performance, demand source, upstream realization, horizontal differentiation, and risk boundary.
First, let's look at the overall situation. As of mid-June 2026, the overall profit of the S & P 500 index in Q2 is expected to increase by 21.2% year-on-year, with revenue increasing by 10.7%. This expectation has been significantly revised up from 18% in early April. FactSet's forecast is more optimistic, believing that Q2 profit growth can reach 22.0%, marking the second consecutive quarter of S & P 500 achieving profit growth of over 20%.
Eleven of the 16 Z industry categories are expected to achieve positive profit growth. However, there are key structural features hidden beneath the total: the technology sector is expected to have a high profit growth rate of 42% in Q2. If the technology sector is excluded, the profit growth rate of the rest of the S & P 500 will plummet from 21.2% to 11.3%. The Z data further shows that from a full-year perspective, the overall profit of the S & P 500 is expected to grow by 21.1% in 2026, but the growth rate will drop to 12.6% after excluding the technology sector.
It is worth noting that John Butters, a senior payoff profile analyst at FactSet, pointed out that without Nvidia and Micron Technology, the expected profit growth rate of the S & P 500 index in the second quarter would have dropped from 22% to 14.9%. Only the two companies contributed about 7 percentage points of growth to the entire index - a concentration level not seen since the dot-com bubble.
JPMorgan Chase raised its year-end target for the S & P 500 index from 7600 to 7800 on June 25th, and raised its earnings per share forecast for 2026 to $350 (a year-on-year increase of 29%), predicting earnings per share of $390 in 2027. JPMorgan Chase's global head of market strategy admitted, "Looking back, our initial judgment on the S & P 500's earnings prospects was too conservative. As of now, the consensus earnings forecasts for 2026 and 2027 have been raised by about 10%, respectively."This kind of market-wide collective increase in profits is very rare and has only occurred in the recovery phase after major economic shocks or recessions.
However, JPMorgan also warned: 'For two consecutive quarters, corporate financial reports have significantly exceeded expectations, raising the market expectation threshold for Q2 performance. It is significantly more difficult for companies to significantly exceed market expectations in both profitability and capital expenditures dimensions.'
The core conclusion of this section: The total number is impressive, but extremely concentrated. Investors holding S & P 500 index funds appear to have diversified allocation, but their actual returns are highly concentrated in a few AI leaders. Once the AI sector corrects, the index's resistance to decline will be severely overestimated.
After understanding the aggregate characteristics of "highly concentrated growth", the next question that must be answered is: where does the driving force behind concentration come from?
The answer is the four major cloud providers - Microsoft, Google, Amazon, and Meta. They are burning money to build AI infrastructure at an unprecedented rate, forming the main source of demand for AI chips.
Goldman Sachs predicts that the capital expenditures of the four hyperscale Cloud as a Service providers alone will reach about $725 billion in 2026. Between 2025 and 2030, this number will accumulate to $5.30 trillion, forming an unprecedented capital expenditure supercycle.
Microsoft's Q2 fiscal year 2026 (ending December 31, 2025) revenue was $81.30 billion, a year-on-year increase of 17%; GAAP Net Profit was $38.50 billion, a year-on-year increase of 60%. Capital expenditures reached $37.50 billion, a year-on-year increase of 66%, setting a historical record.
The more critical data is that the commercial residual performance obligation (RPO) increased by 110% year-on-year, reaching $625 billion - equivalent to about 2.5 years of contract revenue that has been locked in advance. Cloud business revenue 51.50 billion US dollars, a year-on-year increase of 26%; Azure and other Cloud as a Service revenue increased by 39% year-on-year.
Google's total revenue in Q1 2026 was $109.90 billion, a year-on-year increase of 22%. Google Cloud revenue exceeded $20 billion for the first time, reaching $20.02 billion, a year-on-year increase of 63%. CEO Sundar Pichai clearly stated that "enterprise-level artificial intelligence solutions became the main growth driver of cloud business in Quarter 1 for the first time." Cloud backlog orders nearly doubled from the previous quarter, exceeding $460 billion.
The company raised its full-year capital expenditure guidance to $180 billion to $190 billion and management said capital spending would "increase substantially" in 2027.
Amazon's Q1 2026 AWS revenue was $37.60 billion, up 28% YoY, the fastest growth rate in the past 15 quarters. The annualized revenue run rate was $150 billion. However, Q1 capital expenditures were as high as $43.20 billion, and free cash flow plummeted to $1.20 billion in the past 12 months.
Meta's Q1 revenue 55.024 billion USD, up 33% YoY. The company raised its 2026 capital expenditure guidance to 125 billion - 145 billion USD, nearly double its 2025 expenditure level.
The core conclusion of this section: The capital expenditure of cloud vendors is not "burning money out of thin air". Microsoft's $625 billion and Google's $460 billion contract backlogs are real money invested by customers with prepayments and long-term contracts - AI demand is real and has been locked in revenue for the next 2-3 years. But the pressure on free cash flow is also real: Amazon's free cash flow in the past 12 months is only $1.20 billion. The money is indeed being spent, but whether it can continue to translate into healthy cash flow still needs to be verified.
Where does the huge capital expenditure of cloud vendors ultimately flow? The answer is the AI chip industry chain. This is the link with the largest stock price increase and highest valuation in this round of AI market, and it is also the most direct evidence of financial report verification of "whether the money is worth it". The three key companies have all submitted their answers.
NVIDIA 2026 fiscal year Q2 (as of April 26) actual revenue 46.70 billion dollars, an increase of 56% year-on-year, exceeding market expectations of 46.058 billion dollars; Data center revenue 41.10 billion dollars; GAAP Net Profit 26.422 billion dollars, an increase of 59% year-on-year.
The key is guidance. The company's outlook for third-quarter revenue of $54 billion (fluctuating 2%) exceeded market expectations, but the growth rate was significantly narrower than in previous quarters. When "exceeding expectations" becomes the norm, "meeting expectations" may be interpreted as negative.
Broadcom's Q2 revenue for fiscal year 2026 was $22.187 billion, a year-on-year increase of 48%. AI semiconductor revenue reached $10.80 billion, a year-on-year increase of 143%, and exceeded $10 billion for the first time in a single quarter. AI semiconductor orders for the quarter exceeded $30 billion, far exceeding shipments of $10.80 billion. The company reiterated that AI semiconductor revenue for fiscal year 2027 will exceed $100 billion.
However, the market's response to its guidance was not enthusiastic - Broadcom fell more than 14% after the market closed. The reason is that the market expects not only "good", but also "the best". When a stock has already been included in extremely high expectations, "good" is no longer enough.
Micron has released its Q3 financial report for fiscal year 2026 after the close of trading on June 24th Eastern Time.
Indicator | Actual data | Market expectations | Gap |
Revenue | 41.46 billion dollars | 35.84 billion dollars | Exceeded expectations by 15.7%. |
Net Profit (GAAP) | 28.24 billion dollars | — | Increased nearly 14 times year-on-year |
Adjusted earnings per share | $25.11 | $20.49-20 | Exceeded expectations by 20% + |
Gross profit margin | 84.6% | — | — |
Revenue grew 346% YoY and 73.7% QoQ, setting a new record for the fifth consecutive quarter. What's even more exciting is the forward guidance: Q4 revenue is expected to reach $49 billion-51 billion, with a median range of $50 billion, which is 15.6% higher than analysts' expectations; gross profit margin is expected to be about 86%. The company has signed 16 long-term agreements covering binding purchases for three to five years. After the financial report was released, the US CD rose by about 16%.
The core conclusion of this section: Three financial reports jointly prove that the end point demand for AI chips is still extremely strong. Micron's $41.46 billion quarterly revenue, Broadcom's $30 billion AI order backlog, and Nvidia's $46.70 billion revenue all point to the same fact: the capital expenditures of cloud vendors are actually being translated into revenue for chip companies.
However, the slope of growth rate is changing. NVIDIA's Q3 guidance growth rate has significantly slowed down, and Broadcom's "good but not good enough" has encountered market sell-off. AI chip companies are switching from "accelerating growth rate" to "maintaining high growth rate but no longer accelerating". Can the growth rate be maintained in the next quarter on such a high base? This is a problem that holding AI hardware stocks must face.
When AI chips and cloud vendors occupy all the attention, another easily overlooked theme of the Q2 earnings season is the sharp differentiation between industries .
The financial sector has shown strong profit contribution in Q1 - financial stocks contributed $3 to the S & P 500's earnings per share of $9 in the quarter. The market expects the financial sector to grow by about 116% in Q2, accounting for about 25% of the overall profit growth of the S & P 500.
However, there is also a sharp differentiation within financial stocks: as of June 12, 2026, Citigroup has risen by 20.86% this year, Goldman Sachs has risen by 22.08%, Morgan Stanley has risen by 21.88%, while Wells Fargo has fallen by 9.20% during the same period - the yield gap in the same sector exceeds 30 percentage points.
The improvement in earnings expectations for the energy sector is directly related to geopolitics - the Iran situation has pushed up energy prices, boosting the earnings prospects of energy companies. The chemical industry in the basic materials sector has also benefited from this. Dow Chemical and LyondellBasell's Q2 earnings per share expectations have more than doubled in the past month.
The core conclusion of this section: FactSet data shows that without NVIDIA and Micron, the expected profit growth rate of the S & P 500 will drop from 22% to 14.9%. The overall growth rate of more than 490 constituent stocks outside of technology stocks is limited, but sectors such as finance, energy, and basic materials are showing structural opportunities under macro environment and geopolitical factors. Retail investors should examine whether their holdings are overly concentrated in the AI track and miss value gaps in other industries.
After understanding the total amount, demand, upstream realization, and industry differentiation, the final question is: what could disrupt the current fragile balance?
The head of Delta One, the derivatives trading department of Goldman Sachs, issued a clear warning: the AI market is a rubber band pulled to the limit - in the past few weeks, the market has almost ignored all negative signals in AI capital expenditure transactions. At the same time, a structural divergence is widening: large-scale cloud providers continue to raise the ante expenditure commitments, but their stock prices continue to underperform the market.
Goldman Sachs warns that the condition for triggering the "breakpoint" is for a major spender to conclude that "spending slightly less money is more beneficial to shareholder returns". Once this conclusion is established, the valuation system will face revaluation pressure. Goldman Sachs further pointed out that the main risk of AI trading is no longer just a "valuation bubble" - the forward Price-To-Earnings Ratio is not significantly out of control, because profit expectations are revised up synchronously. What really needs to be tested is whether the current strong profits can continue to be maintained after the capital expenditure cycle peaks.
The "Black Tuesday" of this Tuesday has sounded the alarm - the Philadelphia Semiconductor Index plummeted 7.87% in a single day. The trigger for the sell-off was not the lower-than-expected performance, but the adjustment of SK Hynix's pace of slowing down the expansion of HBM4 production capacity - the market is extremely sensitive to any signal that may suggest a "change in the pace of AI investment".
Cloud vendors cut capital expenditure guidance : Goldman Sachs defines it as the core trigger for "breakpoint"
AI chip companies guide growth to continue to narrow : will confirm the judgment of "growth peak"
Free cash flow continues to deteriorate and backlog growth slows down : indicating that burning efficiency is declining
The core conclusion of this section is that the demand is real and the performance is strong, but the market's tolerance is narrowing. When any major cloud vendor or chip company releases a signal of "slowing growth", the entire valuation system may face drastic restructuring.
The reason why the Q2 financial report season in 2026 is "the most important since the rise of AI" is that it will follow the complete chain of total amount → demand → upstream → differentiation → risk, verifying a core proposition: can the huge investment in AI infrastructure continue to translate into revenue and profit growth?
Micron's $41.46 billion revenue gives one answer. Broadcom's $30 billion AI order gives the second answer. Microsoft's $625 billion contract backlog gives the third answer.
But signals of a peak, a 7.87 per cent one-day plunge in the Philadelphia semiconductor index and the risk of a "breaking point" warned by Goldman Sachs constitute a complex and fragile balance.
The data is beautiful, the story is touching, but investment ultimately returns to fundamentals. The Q2 financial report season is the moment to test everything along this logical chain.
Disclaimer: This report is for research reference only and does not constitute any investment advice. Crypto asset prices fluctuate greatly, geopolitical events and macroeconomic changes may have a significant impact on the market. Investors should make independent judgments based on their own risk tolerance. Any platform products or trading pairs mentioned in the report are only presented as objective data and do not constitute buying or selling advice.