Alibaba Google AI Race Exposes a 75% Profit Drop and a Widening Cloud Gap

Alibaba reported a 75% profit decline after sharply increasing AI infrastructure spending, turning the Alibaba Google cloud contest into a test of financial endurance. Net income fell to 10.5 billion yuan for the April-to-June quarter. It stood at 43.1 billion yuan one year earlier.

The decline did not come from collapsing demand. Alibaba’s quarterly revenue rose 9% to almost 269 billion yuan, while cloud and computing revenue increased 45% to 48.4 billion yuan. Capital expenditures, which include AI infrastructure, climbed 75% to 67.7 billion yuan.

That combination creates the central conflict. Alibaba is finding substantial demand for AI capacity, but it must build that capacity before the resulting revenue fully appears. Google is making the same broad infrastructure bet from a much stronger earnings position.

The comparison matters because Google Cloud reported 82% year-over-year growth during the same calendar quarter. It also produced 8.8 billion dollars in operating income, up from 2.8 billion dollars one year earlier. Alibaba’s cloud operation is growing, but Google already shows how AI infrastructure can support both acceleration and substantial operating profit.

This is not simply another quarter of expensive AI development. Alibaba has reached the stage where customer demand, chip procurement, capacity constraints, and financial returns must start moving together. The next few quarters will show whether the spending creates a profitable cloud cycle or leaves shareholders funding a long capacity race.

Alibaba’s June Quarter Put AI Spending Ahead of Profit

Alibaba’s earnings reveal a company purchasing future cloud capacity with current profit.

The company reported its June-quarter results on August 20, 2026. Revenue reached almost 269 billion yuan, representing 9% growth from the prior-year period. However, net income dropped from 43.1 billion yuan to 10.5 billion yuan.

The linked headline describes a 76% decline, while the reported figures imply a reduction of roughly 75%. The distinction does not change the underlying story. Alibaba absorbed a major profit contraction while expanding infrastructure for AI workloads.

Capital expenditures reached 67.7 billion yuan, about 75% above the prior-year level. That spending covered infrastructure needed for cloud computing and AI services, according to the company’s results and subsequent reporting.

The increase reflects more than routine data-center expansion. Alibaba cited changing procurement cycles, additional central processing capacity, higher component costs, and anticipated demand for AI agents. An AI agent is software that can plan and execute multiple tasks with limited human direction.

Alibaba is therefore installing capacity before every unit has a committed customer workload attached to it. This approach reduces the risk of turning away demand, but it transfers utilization risk onto Alibaba’s balance sheet.

The revenue side provides evidence that the investment is addressing a real market. Cloud and computing revenue rose 45% to 48.4 billion yuan. AI-related product revenue also continued growing rapidly, according to Alibaba.

Those figures explain why management has not responded to lower profit by slowing construction. CEO Eddie Wu said increased supply should support faster AI and cloud revenue growth during coming quarters. He also tied that growth to improving profitability.

That forecast remains a company expectation rather than an established result. The quarter demonstrates rising demand and rising capacity. It does not yet demonstrate that the returns on new infrastructure will exceed the cost of supplying it.

The quarterly results should also be read alongside Alibaba’s other spending commitments. The company previously said it would invest at least 380 billion yuan in cloud and AI infrastructure over three years.

That commitment now looks less like a distant plan and more like an active financial constraint. One quarter of spending equaled more than six times the quarter’s net income. Maintaining that pace requires Alibaba’s other businesses to keep producing cash while cloud demand develops.

The profit decline also cannot be reduced to AI spending alone. Alibaba is simultaneously investing in commerce, consumer experiences, and other technology initiatives. Those programs compete for the same management attention and financial resources.

Still, AI infrastructure is now large enough to shape the group’s consolidated results. The latest earnings coverage recorded both sides of the quarter: faster cloud growth and a steep profit decline.

That is the change readers should remember. Alibaba is no longer discussing AI primarily through model launches or benchmark results. It is exposing the cost of turning those models into an industrial cloud service.

Why the Alibaba Google Comparison Matters Now

The Alibaba Google comparison shows that AI demand alone does not determine whether infrastructure spending looks successful.

Alphabet reported its second-quarter results on July 22, less than one month before Alibaba’s release. Google Cloud revenue rose 82% year over year to 24.8 billion dollars. Cloud operating income increased to 8.8 billion dollars from 2.8 billion dollars.

Alphabet’s earnings release attributed that growth to demand for AI infrastructure and AI solutions. Google also reported broad adoption of its Gemini products across large enterprises.

The most important contrast is not simply Google’s faster cloud growth. Google Cloud converted that growth into much higher operating income while Alphabet’s advertising businesses continued generating substantial cash.

Alibaba lacks an equivalent global advertising engine. Its commerce operations remain large, but they face intense domestic competition and require their own spending. That makes each yuan directed toward AI infrastructure more visible in consolidated profit.

The companies also operate in different markets. Google sells cloud and AI services across a broad international customer base. Alibaba holds a major position in China and serves international customers, but geopolitical and regulatory conditions limit a direct market-share comparison.

Chip access creates another difference. Google designs Tensor Processing Units, or TPUs, which are specialized processors for training and running AI systems. Its 2026 filing says Google Cloud began recognizing revenue from sales of TPU systems during the second quarter.

Alibaba develops proprietary chips and orchestration software, but its supply environment is more constrained. Export controls have limited Chinese access to some advanced processors, increasing the importance of domestic chips and efficient cluster management.

These differences make Google a useful reference point without making it a perfect twin. Both companies operate models, developer platforms, cloud infrastructure, and enterprise software. Both must spend before customers fully consume the resulting capacity.

However, Google entered the latest investment cycle with a larger international cloud business and a profitable advertising base. Alibaba entered while defending commerce activity and expanding consumer-facing services.

The Alibaba Google race is therefore best understood as two versions of the same AI infrastructure wager. Google is scaling a profitable cloud platform supported by another highly profitable business. Alibaba is attempting to improve its business mix while financing the transition.

That distinction changes how investors interpret capital spending. Google can report high infrastructure costs alongside rising cloud operating income. Alibaba’s spending currently appears beside a 75% decline in group profit.

The pressure falls most directly on Alibaba management. It must show that cloud revenue can continue accelerating after new capacity comes online. It must also demonstrate improving margins without abandoning strategically important commerce investments.

Customers also have a stake in the outcome. More capacity can reduce waiting times for computing resources and support larger AI deployments. Yet customers need predictable availability and service quality, not only headline investment totals.

For developers, the relevant question concerns the durability of the platform. Model access, inference capacity, deployment tools, and pricing stability depend on the provider’s ability to keep funding infrastructure. A temporary capacity surge offers less value than a sustainable service.

Google’s performance raises the standard for what sustainable growth looks like. Its cloud unit is not merely expanding revenue. It is producing operating income while absorbing higher technical infrastructure usage costs.

Alibaba does not need to copy Google’s business structure. It does need to prove that its own structure can fund a comparable cycle of investment, utilization, and recurring revenue.

The Real Tradeoff Is Capacity Today Versus Returns Tomorrow

Alibaba is accepting lower near-term earnings because unused or unavailable computing capacity could cost it future AI customers.

AI infrastructure requires commitments long before revenue becomes certain. Providers must secure chips, networking equipment, power, data-center space, storage, and technical staff. Delaying those commitments can leave a cloud provider unable to fulfill customer demand.

Alibaba says it expects wider adoption of AI agents to increase computing requirements. Those systems can perform repeated model calls, retrieve information, generate software, and interact with business applications. A single user task can therefore consume more computing capacity than a conventional search or software request.

That demand pattern gives management a reason to build early. If customer workloads grow as predicted, newly installed capacity can produce recurring cloud revenue. Higher utilization can then spread fixed infrastructure costs across more usage.

The opposite outcome is expensive. If adoption develops more slowly, Alibaba may hold underused assets while depreciation and operating costs continue. Hardware can also lose economic value quickly as newer processors become available.

This is the core tradeoff behind the profit decline. Alibaba can protect present earnings by limiting expansion, but doing so risks surrendering future workloads. It can protect future capacity by spending heavily, but doing so exposes current investors to uncertain returns.

The company’s cloud growth offers meaningful support for the second choice. Revenue increased 45%, representing a substantial acceleration from earlier periods. AI-related products have also maintained strong growth over multiple quarters.

Alibaba’s March-quarter results provide an important baseline. Cloud Intelligence Group revenue grew 38% during that period, while external cloud revenue grew 40%. AI-related product revenue reached 8.97 billion yuan and accounted for 30% of external cloud revenue.

The March results showed that AI was already becoming a material portion of cloud activity before the June acceleration. This reduces the risk that the latest growth reflects a single new contract or isolated launch.

However, revenue growth alone cannot settle the investment case. Alibaba must eventually disclose enough segment detail for readers to connect AI revenue, utilization, depreciation, and profit. Group-level net income can move because of investment valuations and other nonoperating items.

Operating measures deserve particular attention for that reason. They provide a clearer view of whether commerce and cloud activities are earning more after direct expenses. Free cash flow also shows how much cash remains after infrastructure commitments.

Google illustrates how this progression can work. Its cloud division once generated operating losses while the company built scale. It now reports significant operating income, with infrastructure and platform services driving growth.

That historical pattern supports Alibaba’s strategy, but it does not guarantee the same result. Google benefits from global distribution, proprietary infrastructure, existing enterprise relationships, and a large software portfolio.

Alibaba has different advantages. It can integrate cloud services with a substantial commerce network, a major domestic customer base, and the Qwen model family. Its position in China can also generate workloads that overseas providers cannot easily serve.

Yet those advantages must translate into paid, recurring consumption. Downloads, benchmark attention, and developer experimentation do not produce the same economics as sustained enterprise inference.

Inference is the process of running a trained model to produce an answer or action. Its economics depend on hardware efficiency, customer usage, service pricing, and the provider’s ability to keep systems busy.

The June quarter indicates that Alibaba is building for inference at scale. It does not tell readers how quickly new clusters will reach productive utilization. That missing connection separates a credible growth strategy from a completed financial success.

Google Shows the Profit Benchmark Alibaba Must Reach

Google has moved beyond proving AI demand, while Alibaba is still proving that demand can outweigh the cost of serving it.

Alphabet’s quarterly filing offers a sharper benchmark than revenue growth alone. Google Cloud operating income rose by about 6 billion dollars year over year during the June quarter. Revenue increased by approximately 11.1 billion dollars.

The regulatory filing says infrastructure and platform services drove the cloud increase. It also notes higher technical infrastructure usage costs and employee compensation expenses.

In other words, Google faced rising costs while expanding profit. That outcome represents the mature version of the AI cloud argument: infrastructure spending increases capacity, customer usage fills that capacity, and operating leverage eventually appears.

Alibaba currently occupies an earlier and more difficult point in that sequence. Its cloud growth is accelerating, but group profit has dropped sharply. Management must show that new AI capacity improves cloud economics before other investments create additional pressure.

The comparison also highlights the role of funding sources. Alphabet’s Search business increased revenue and operating income during the quarter. Advertising effectively provides a large internal financing engine for data centers, chips, models, and product distribution.

Alibaba’s commerce engine remains central, but its domestic market is competitive. Investments in delivery, merchant services, consumer acquisition, and user experience can reduce the cash available for AI.

This creates a capital-allocation conflict. A commerce slowdown can pressure Alibaba to spend more on consumer incentives precisely when cloud infrastructure also demands more money. Management cannot assume both programs will produce immediate returns.

The Alibaba Google comparison therefore exposes a deeper difference than model performance. Google can distribute AI across Search, Workspace, Cloud, Android, and YouTube. Each product provides customer relationships, data, or revenue that can support the broader investment.

Alibaba can distribute Qwen and AI services through Cloud, Taobao, Tmall, DingTalk, and other operations. However, it must prove that this distribution creates profitable usage instead of raising costs across several businesses.

Google’s position is not risk-free. Alphabet also increased capital expenditures substantially, and rising depreciation will affect future margins. Its filing warns that investments, product changes, competition, regulation, and customer behavior can alter results.

The market can also punish spending even when cloud demand appears strong. Investors want evidence that providers can turn costly capacity into lasting cash flow. Rapid revenue growth does not eliminate concerns about infrastructure intensity.

That qualification matters because Alibaba does not need to match Google’s exact margin immediately. It needs to establish a direction. Cloud revenue growth should remain strong, and segment profitability should improve as additional capacity reaches paying users.

A slowdown in cloud growth would challenge the strategy. So would continued group profit compression without clearer evidence of cloud operating leverage. Either result would suggest Alibaba built ahead of durable demand.

Conversely, sustained cloud acceleration with improving profit would strengthen management’s case. It would show that the June spending surge represented a deliberate capacity expansion rather than uncontrolled cost growth.

Google has already made that transition visible in its reported cloud results. Alibaba’s next task is to produce a similarly clear connection between infrastructure, customer usage, and profit.

What the Earnings Still Do Not Prove

One strong cloud quarter cannot establish that Alibaba’s AI investment will earn an acceptable long-term return.

The first uncertainty concerns accounting. Net income can change because of investment gains, impairments, taxes, and other items outside daily operations. A headline decline does not measure infrastructure economics by itself.

Readers should therefore avoid treating the 75% fall as a precise measure of AI losses. Alibaba attributed higher spending to several programs, including infrastructure and broader technology initiatives. Commerce investments also affected profitability.

The second uncertainty concerns capacity utilization. Alibaba disclosed much higher capital spending and stronger cloud revenue, but those figures do not reveal how much new computing capacity is already serving paid workloads.

That timing matters. Infrastructure installed near the quarter’s end may contribute costs before generating meaningful revenue. A later increase in utilization could improve the economics without requiring another comparable spending jump.

The reverse is also possible. Persistent chip shortages or customer growth could require continued investment at the same pace. Revenue might then rise while free cash flow remains constrained.

The third uncertainty concerns the composition of AI demand. Enterprise contracts, developer experimentation, consumer applications, and internal Alibaba workloads carry different margins. They also produce different levels of recurring revenue.

Management says AI and cloud growth should accelerate as supply expands. That statement is plausible given the company’s reported demand, but it has not been independently verified. Future results must confirm it.

Competition adds another source of uncertainty. Chinese customers can evaluate services from Alibaba, Tencent, Baidu, and other providers. Price reductions or aggressive capacity expansion could weaken returns even if total demand grows.

Google and other global providers exert indirect pressure as well. Their model releases and infrastructure improvements shape customer expectations for performance, reliability, and development tools. Alibaba must remain competitive while operating under a different chip supply environment.

Regulation can also influence the available market. Data rules, export controls, model requirements, and cross-border restrictions can limit where providers deploy infrastructure or serve customers.

These constraints can favor domestic providers inside China, but they can complicate international expansion. Alibaba may gain protected demand in one market while facing greater difficulty reaching another.

The fourth uncertainty concerns model differentiation. Qwen has attracted developers through an expanding family of models, but model quality can converge quickly. Cloud providers need durable advantages in deployment, cost, security, support, and application integration.

A popular model can create initial adoption. It does not automatically create a defensible cloud margin. Developers can move workloads when compatible alternatives offer better economics or availability.

Customers should separate three questions when evaluating Alibaba’s progress. Is Qwen capable enough for the intended workload? Can Alibaba supply reliable computing capacity? Can it provide that capacity on economically sustainable terms?

The June results answer the second question only partially. Spending shows an effort to expand supply, while cloud growth shows increasing use. Reliability, utilization, and returns remain less visible.

The quarter also does not prove that Alibaba is losing the AI race. A temporary profit decline can accompany a rational investment cycle. Google itself spent years building cloud infrastructure before the unit became a major profit contributor.

Still, that precedent should not become an excuse for indefinite spending. Alibaba must provide measurable progress. Otherwise, the comparison with Google shifts from a long-term opportunity to evidence of a structural disadvantage.

Three Signals Will Decide Whether Alibaba’s Bet Works

Cloud growth, segment profit, and capital intensity will determine whether Alibaba’s AI spending created value or only capacity.

The first signal is the next reported cloud growth rate. Alibaba’s June-quarter cloud and compute revenue increased 45%, following strong growth in the March period. Maintaining or improving that pace would support management’s claim that supply constraints were limiting revenue.

A sharp deceleration would weaken the argument. It would suggest that the latest capacity expansion arrived as demand began normalizing. Readers should also watch whether AI-related products continue taking a larger share of external cloud revenue.

The second signal is cloud profitability. Alibaba needs to show that additional usage is producing operating leverage, meaning revenue grows faster than the operating costs required to serve it.

An improving cloud margin would strengthen the Google comparison. It would indicate that Alibaba is progressing along the same broad path from infrastructure investment toward recurring operating income.

Flat or declining profitability would require closer examination. The cause could be new depreciation, low utilization, price competition, expensive chips, or the cost of launching new services.

The third signal is capital spending relative to operating cash flow and cloud revenue. One large quarter can reflect procurement timing. Several quarters at the same intensity would indicate a more demanding financial commitment.

If capital expenditures moderate while cloud growth remains high, Alibaba will have evidence that earlier purchases created usable capacity. If spending stays elevated and growth slows, the investment thesis will face greater pressure.

Google provides a live control case for these signals. Its cloud revenue and operating income are rising together, although Alphabet also faces questions about the scale of its capital program. The company’s results show that high spending can accompany improving economics, but only when customer consumption catches up.

The next Alibaba Google earnings comparison should therefore focus less on which company announces the most capable model. It should examine which provider converts installed computing capacity into durable, profitable usage.

For enterprise buyers, the immediate lesson is not to select a provider from a single earnings release. Buyers should assess availability, model performance, data requirements, regional coverage, and the provider’s ability to support production workloads.

Developers should also watch how capacity investment changes practical access. Faster inference, higher usage limits, and more reliable deployment would show that Alibaba’s spending is reaching customers. Announcements without operational improvements would offer weaker evidence.

Knowledge workers will encounter the results through products built on these platforms. Better infrastructure can improve response speed, context handling, and agent reliability. It can also encourage more organizations to deploy AI across routine workflows.

Alibaba has chosen to absorb a steep earnings decline while building for that demand. Google has shown that similar infrastructure can become highly profitable when scale, utilization, and distribution align.

The decisive question now is measurable: can Alibaba keep cloud growth near its latest pace while restoring profit and reducing capital intensity? Watch those three signals in the next results, because they will reveal whether the spending surge purchased a stronger business or merely a larger bill.