Chinese tech giant Alibaba is making a significant gamble in the global AI competition.

The financial results for the April-June quarter of 2026 appeared strong at first glance. Revenue reached approximately 268.9 billion yuan, marking an increase of about 9% compared to the same period last year. Growth in the cloud business, supported by AI demand, is also accelerating.

However, the numbers that investors focused on were not the sales figures.

Net profit fell to about 10.4 billion yuan, a decrease of about 75% from the same period last year. Additionally, capital investment, primarily in AI infrastructure, expanded rapidly, resulting in a significant negative free cash flow.

Despite the company's growth, profits and cash are declining.

This seemingly contradictory financial result symbolizes the current situation Alibaba finds itself in.

The company is no longer just competing as one of China's largest e-commerce companies. It is in the midst of a massive competition alongside Amazon, Microsoft, Google, and domestic rivals like Tencent and Baidu, over who will control the computing infrastructure of the AI era.

And the cost of participating in this competition is higher than imagined.

Unusual financial results: 9% increase in revenue, 75% decrease in net profit

In this financial report, Alibaba's revenue reached 268.953 billion yuan.

This represents an approximately 9% increase compared to the same period last year, which is not a weak number considering the company's size.

However, net profit was about 10.444 billion yuan, a significant decrease from approximately 43.1 billion yuan in the same period last year.

Several factors contributed to the profit decline, but the market particularly focused on the massive capital investment centered around AI and cloud.

In the April-June quarter of 2026 alone, capital investment reached about 67.678 billion yuan, a 75% increase from the same period last year, equivalent to about 10 billion dollars.

This means that such a large amount of funds was invested in just three months.

Operating AI services requires large-scale data centers.

These centers are filled with GPUs, CPUs, storage, networking equipment, power facilities, and cooling systems. As the number of users increases, computing resources must be scaled up proportionally.

The expansion of generative AI demand presents a significant opportunity for software companies, but it also means a transformation into a "huge equipment industry."

Alibaba's financial results illustrate this reality in numbers.

AI Cloud grows by 45%

However, it is premature to simply view this investment as "wasteful."

The most important factor is the growth rate of the cloud business.

Alibaba's AI Cloud and Compute Services revenue increased by 45% compared to the same period last year, expanding to approximately 48.4 billion yuan.

According to the company, this growth rate is the highest in 22 quarters.

Additionally, sales of AI-related products have continued triple-digit growth for 12 consecutive quarters compared to the same period last year.

This means that Alibaba's massive investment is not just based on "expectations for the AI boom," but there is actual customer demand.

When companies adopt generative AI, they need not only the model itself but also the cloud environment to run it.

As large-scale inference processing, data storage, learning processes, and AI agent execution begin to operate on the cloud, the consumption of computing resources increases rapidly.

Alibaba is targeting this massive "AI computing demand."

380 billion yuan AI investment, half already deployed

Alibaba has announced plans to invest a total of 380 billion yuan in AI and cloud infrastructure over three years.

This is a massive investment that reaches several trillion yen when converted to Japanese yen.

By the end of June 2026, it is reported that approximately 190 billion yuan has already been invested.

This means that about half of the plan has been executed.

Importantly, the company has shown little sign of easing the investment accelerator at this point.

The management anticipates that the current shortage of AI computing capacity could continue for several years and aims to strengthen infrastructure as much as possible while demand exceeds supply.

This overlaps with the era when Amazon expanded AWS.

In the early stages, the cloud business bears a heavy investment burden for data centers. However, once a certain scale is exceeded, there is potential for profit margins to improve due to increased utilization rates.

Alibaba is attempting to replicate this structure with AI cloud.

Management's calculation of "recouping in three years"

A particularly noteworthy aspect of the earnings call was the management's specific thoughts on the payback period for AI-related capital investments.

Based on the current average gross profit margin of AI products, they believe AI-related capital investments can generally be recouped in about three years.

Furthermore, if profit margins improve and the use of in-house developed chips increases, the payback period could potentially be shortened to two to two and a half years.

This is a very important point.

The market is not afraid of the "amount of investment in AI" itself.

The real issue is

how much profit that investment will generate in the future

.

Even if 10 billion dollars are invested, there is no problem if the facilities can generate 20 billion or 30 billion dollars in profit.

Conversely, if demand forecasts are incorrect and data center utilization rates stagnate, massive investments will weigh on the company's earnings for an extended period.

In that sense, the important numbers for Alibaba going forward will not only be the "performance of AI models" but also the utilization rate of AI infrastructure, the profit margin of the cloud business, and the payback period of capital investments.

In-house AI chips as the key to improving profit margins

Alibaba has another important card.

Its in-house developed semiconductors.

In AI data centers, semiconductors like GPUs account for a significant portion of capital investment costs.

Continuing to purchase high-performance commercial AI semiconductors from external companies makes it difficult to increase profit margins even if cloud service sales grow.

Therefore, Alibaba is promoting the expansion of the use of its own designed chips.

Through T-Head, the company is developing CPUs, AI semiconductors, and network-related chips, and is expanding the use of its latest generation AI processors in Alibaba Cloud.

If the performance of in-house chips improves sufficiently and the proportion of expensive externally procured semiconductors can be reduced, the cost structure of AI cloud could significantly improve.

This is not just about cost reduction.

If the "full-stack strategy" of integrating cloud, AI models, and semiconductors can be realized, it could become a significant advantage over competitors.

AI model "Qwen" as the gateway to the cloud

Understanding Alibaba's AI strategy requires recognizing the presence of the large language model "Qwen."

The Qwen series actively adopts an open model strategy and is used by developers worldwide.

Even if the model is spread for free or in an open form, there may be cases where significant revenue cannot be obtained from the model alone.

However, Alibaba's real aim lies beyond that.

As more companies and developers use Qwen, the likelihood increases that they will use Alibaba Cloud as the computing environment to run the model.

For cloud companies in the AI era, models also serve as gateways to guide customers to the cloud.

By expanding open models, providing the cloud that can operate those models most efficiently, and even combining in-house semiconductors,

if this is completed, Alibaba will have multiple revenue sources in the AI market simultaneously.

Concerns remain for the core e-commerce business

In contrast to the growth of AI cloud, investors are wary of the e-commerce business.

In China, price and service competition with JD.com and PDD Holdings continues, and significant competition costs are incurred in quick commerce, including immediate delivery.

Funds are not only being used for the AI business, but investments are also needed to maintain market share in the existing e-commerce market.

This is where Alibaba's difficulty lies.

Funds must be invested in the growing AI business.

However, the e-commerce business, which has generated significant profits, cannot be neglected.

Investment is required in both the "future AI" and the "current e-commerce," resulting in short-term profits and cash flow being easily pressured.

Selling non-core assets to focus funds on AI

Alibaba is also advancing the reorganization of its business portfolio.

One example is the sale of the gaming business Lingxi Games.

The direction of divesting non-core businesses and concentrating funds and management resources on key areas such as AI, cloud, and e-commerce is becoming clear.

This is not just about raising funds; it can also be seen as an indication of "what kind of company Alibaba will become in the future."

In the past, Alibaba expanded its business into a wide range of areas, focusing on e-commerce, finance, logistics, video, gaming, and local services.

However, entering the AI era, it is trying to narrow down the center of corporate value again.

At the center are AI and cloud.

Post-earnings, stock price drops over 8%

In response to this long-term strategy, the short-term evaluation by the stock market was harsh.

On August 21, Alibaba's stock was heavily sold in the U.S. market, closing down about 8.6%.

This indicates that the market focused more on the sharp decline in profits and the deterioration of cash flow due to massive investments than on the growth of AI cloud.

The biggest change currently occurring with investments in AI companies is that the phase of "AI equals stock price increase" is coming to an end.

Investors are gradually starting to focus on

"How much has AI-related revenue increased?"

"How much capital investment is needed?"

"How long will it take to recoup that investment?"

"How much can the profit margin of the AI business increase?"

.

Alibaba's recent financial results symbolize this turning point.

Opinions on social media are divided

Interestingly, despite the significant drop in stock price, the evaluation by individual investors on social media is not entirely pessimistic.

In the Alibaba investor community on Reddit, opinions on the recent financial results were quite divided.

The bullish faction is most focused on the 45% growth in the cloud.

Responses such as "The content of the earnings call is more important than the stock price drop" and "If AI cloud growth accelerates and profit margins improve, it's positive in the long term" were observed.

There is also significant attention on the management's statement that the payback period for AI capital investments is about three years.

Some individual investors perceive the explanation that increased use of in-house chips could further improve profit margins as "the current cash flow deterioration is an investment to create future cloud profits."

On the other hand, the concerns of the cautious faction are clear.

Reactions such as "The amount of capital investment is frightening," "I expected stronger numbers from the e-commerce business," and "Can AI growth compensate for the weakness of other businesses?" were observed.

Indeed, there are posts about reducing holdings and voices waiting to buy until the stock price drops further.

In other words, even in the discussions on social media,

the growth of AI cloud is appreciated. However, the issue is how much to pay for that growth

, and opinions are divided on this point.

What is even more interesting is Stocktwits.

Even during the period when the stock price fell significantly, the sentiment indicator for BABA showed "Extremely Bullish."

It can be seen that a certain number of investors view the short-term stock price decline not as a result of the "AI growth story collapsing," but as an opportunity to buy a long-term growth stock cheaply.

Of course, posts and sentiment indicators on social media do not represent the entire market.

However, judging from the reactions this time, the market's points of discussion regarding Alibaba have become quite clear.

The issue is not whether there is demand for AI.

The demand already exists.