Key takeaways
– Alibaba reported June-quarter revenue of RMB268.95 billion, up 9% year over year.
– AI Cloud and Compute Services revenue increased 45% to RMB48.44 billion.
– Net income attributable to ordinary shareholders fell about 76% to RMB10.54 billion; adjusted measures fell less sharply, showing that one number does not explain the full quarter.
– Capital expenditure rose 75% to RMB67.68 billion as Alibaba expanded AI infrastructure and bought more costly chips.
– This is evidence of strong reported demand and heavy investment, not proof that every AI project or cloud provider will earn attractive returns.
Alibaba reported a sharp acceleration in its AI cloud business for the quarter ended Jun 29, 2026, but the cost of building that capacity weighed heavily on profit. The results offer a useful snapshot of the current AI infrastructure race: demand can grow quickly while chips, data centers, model development, and competitive investment consume cash even faster.
The company said total quarterly revenue reached RMB268.95 billion, an increase of 9% from a year earlier. Revenue from AI Cloud and Compute Services rose 45% to RMB48.44 billion. At the same time, net income attributable to ordinary shareholders declined roughly 76% to RMB10.54 billion, while capital expenditure increased 75% to RMB67.68 billion.
Alibaba published the results on Aug 19, 2026. Reuters, the Associated Press, The Wall Street Journal, and the South China Morning Post independently reported the main figures. This article separates the company’s disclosed results from interpretation and does not offer investment advice.
The verified numbers
The headline figures point in different directions, so they should be read together.
| Metric | June quarter 2026 | Year-over-year change | What it indicates |
|---|---|---|---|
| Total revenue | RMB268.95 billion | +9% | The group continued to grow overall |
| AI Cloud and Compute Services revenue | RMB48.44 billion | +45% | Strong reported demand for cloud and AI services |
| Net income attributable to ordinary shareholders | RMB10.54 billion | About -76% | Accounting profit fell sharply |
| Non-GAAP net income | RMB20.72 billion | -38% | Core adjusted profitability also weakened, but less than reported profit |
| Capital expenditure | RMB67.68 billion | +75% | Infrastructure investment accelerated substantially |
The distinction between reported and adjusted profit matters. Reported net income can be affected by investment gains and losses, impairments, share-based compensation, and other items. Non-GAAP measures exclude selected costs or gains and are defined by the company. Neither measure should be viewed alone.
Why profit fell while revenue grew
Alibaba is spending to expand computing capacity before all of that capacity produces mature revenue. AI infrastructure requires accelerators, networking, power, facilities, storage, engineering, and ongoing model work. Procurement costs can arrive earlier than the customer revenue those assets may eventually support.
The company’s capital expenditure reached RMB67.68 billion in the quarter. Reuters reported that chip costs were a major contributor. Alibaba has committed to a multi-year AI and cloud investment program and said it has already deployed about half of the planned RMB380 billion through 2029.
Profit was also affected by factors beyond AI infrastructure. The official results and independent coverage point to lower investment gains and other accounting items. That is why “AI spending caused a 76% profit fall” is too simple. AI investment was a major pressure, but the difference between reported and adjusted declines shows that several factors were involved.
What the 45% cloud growth shows
AI Cloud and Compute Services revenue of RMB48.44 billion was one of the strongest figures in the release. It suggests that customers are paying for a growing combination of computing, model services, and related cloud products. Alibaba also said AI-related product revenue achieved triple-digit growth for a twelfth consecutive quarter.
Those are company-reported figures, not an independent audit of customer outcomes. Revenue growth shows demand for services; it does not show whether customers are receiving positive returns, whether usage is recurring, or how margins compare with other cloud businesses.
The result also cannot be generalized automatically to the whole market. Alibaba operates a broad ecosystem across cloud, commerce, logistics, and consumer services. Its ability to bundle infrastructure, models, and distribution differs from that of a smaller provider.
Alibaba’s AI infrastructure strategy
Alibaba is pursuing several connected layers:
– Cloud infrastructure that rents computing capacity to customers.
– Qwen models and model-as-a-service products.
– In-house T-Head chips intended to reduce reliance on commercial accelerators in selected workloads.
– AI features across commerce and other Alibaba products.
– A large capital program designed to add capacity ahead of expected demand.
Management said it expects the investment economics of AI computing to improve over time. Reuters reported that Chief Executive Eddie Wu discussed a possible two-and-a-half to three-year break-even period for AI infrastructure investment. That is a management expectation, not a guaranteed outcome.
Replacing some commercial chips with Alibaba-designed hardware could improve control over cost and supply. It also creates execution risk: chip performance, software compatibility, manufacturing availability, and customer workload requirements all matter. Proprietary hardware is useful only when the surrounding software and operations work reliably.
The DCM scorecard: demand, cost, monetization
Quarterly AI results are easier to interpret with three separate lenses.
. Demand
Ask whether customers are increasing paid usage, not merely joining pilots. Alibaba’s 45% segment growth is a positive reported demand signal. Better evidence over time would include recurring consumption, customer concentration, retention, and workload diversity.
. Cost
Track the complete cost of providing the service: chips, energy, networking, depreciation, facilities, software, and support. Alibaba’s 75% capital-expenditure rise shows that revenue growth currently requires substantial upfront resources.
. Monetization
Examine whether prices and utilization eventually support durable margins. Fast growth with low utilization or aggressive discounts may not translate into attractive economics. Segment profit, free cash flow, and management’s disclosed utilization trends become important here.
| Signal | Current evidence | Still needed |
|---|---|---|
| Demand | AI cloud revenue +45%; repeated AI-product growth claims | Retention, customer mix, workload durability |
| Cost | Capex +75%; higher chip procurement | Depreciation, energy, capacity utilization |
| Monetization | Management expects improving economics | Sustained segment margins and cash generation |
The DCM scorecard prevents a common error: treating rapid revenue growth as the same thing as proven profitability.
What the results mean for business technology teams
For buyers of AI services, the quarter reinforces three practical points.
First, model choice is becoming an infrastructure decision. Teams should evaluate availability, location, privacy, portability, and total cost instead of selecting a provider only from a model benchmark.
Second, prices may change as providers balance utilization and expensive new capacity. Businesses should measure cost per completed task, not only token price. A cheaper model that requires repeated calls or extensive review may cost more overall.
Third, customers should preserve exit options. Standard interfaces, exportable logs, and direct-provider fallbacks reduce dependence on one cloud or gateway. Alibaba’s expansion may increase choice, but buyers still need their own resilience plans.
Risks and open questions
The results leave several questions unanswered:
– How much of the 45% growth came from existing customers versus new customers?
– What portion of AI infrastructure is operating at efficient utilization?
– How will depreciation affect future profit as new assets enter service?
– Can proprietary chips handle enough production workloads to improve margins?
– How much price competition will emerge among Chinese and global cloud providers?
– Will AI services produce durable cash flow before the next large hardware cycle?
Geopolitical restrictions and semiconductor supply also remain material. The article does not assume that reported growth will continue at the same pace.
Frequently asked questions
How fast did Alibaba’s AI cloud business grow?
Alibaba reported that AI Cloud and Compute Services revenue rose 45% year over year to RMB48.44 billion in the June 2026 quarter.
Why did Alibaba’s profit fall?
Heavy AI infrastructure spending was a major pressure, while lower investment gains and other accounting items also affected reported profit. Adjusted profit fell less sharply than reported net income.
How much did Alibaba spend on capital equipment?
Capital expenditure was RMB67.68 billion for the quarter, up 75% year over year.
Is the 45% growth figure independently audited?
It is a company-reported segment figure contained in Alibaba’s financial results. Independent outlets confirmed that the company disclosed it, but that is different from independently auditing customer outcomes.
Does this article recommend Alibaba shares?
No. It explains operating results and business implications and is not investment advice.
