Alibaba Group Holding Limited

Good day, ladies and gentlemen. Thank you for standing by. Welcome to Alibaba Group's June quarter 2025 results conference call. At this time, all participants are on listen-only mode. After management's prepared remarks, there will be a Q&A session. I would now like to turn the call over to Lydia Liu, Head of Investor Relations of Alibaba Group. Please go ahead.

Good day, everyone. Welcome to Alibaba Group's June quarter 2025 earnings conference call. With us today are Joe Tsai, Chairman, Eddie Wu, Chief Executive Officer, Toby Xu, Chief Financial Officer, Jiang Fan, Chief Executive Officer of Alibaba eCommerce Business Group. This call is also being webcast from the IR section of our corporate website. A replay of the call will be available on our website later today. Now, I will quickly cover the safe harbor. Today's discussions may contain forward-looking statements, particularly statements about our business and the financial results that are subject to risks and uncertainties which could cause actual results to differ materially from those contained in the forward-looking statements. Please refer to the safe harbor statements that appear in our press release and investor presentation provided today. Please note that certain financial measures that we use on this call are expressed on a non-GAAP basis. Our GAAP results and reconciliations of GAAP to non-GAAP measures can be found in our earnings press release. And now I will turn the call over to Eddie.

Hello, everybody, and welcome to this quarter's earnings call. This quarter, we delivered solid growth. Excluding revenue from Sunart and InTime, our total revenue on a like-for-like basis grew 10% year-over-year. Revenue growth of our core businesses remained strong. Customer management revenue from our China e-commerce business rose 10% year-over-year. Cloud intelligence group revenue growth accelerated to 26% year-over-year with AI-related product revenue maintaining triple-digit growth for the eighth consecutive quarter. Revenue from AIDC grew by 19% year-over-year. In AI plus cloud, the accelerated development of AI applications and increasing AI product adoption by customers drove a 26% year-over-year revenue increase from external customers. During the quarter, AI-related revenue accounted for over 20% of revenue from external customers, as AI demand continued to grow rapidly. We're also seeing AI applications driving great growth momentum of traditional products, including compute and storage. SAP and Alibaba entered a strategic partnership focused on cloud and AI. As SAP's global cloud computing partner, Alibaba Cloud will support SAP customers to run and manage their core software systems on Alibaba's platform. Leveraging our Q1 models, SAP will also provide AI transformation services for its enterprise customers. This partnership signifies the recognition of our cloud infrastructure and AI capabilities by the global leading enterprises in the SAP ecosystem. We've continued to advance the capabilities of our AI foundation model since July Alibaba has released upgraded QN3, including a non-thinking model, reasoning model and AI coding model, which are recognized as global top performers in their respective categories. Notably, our QN3 coder model has rapidly increased QN's user adoption in overseas markets. We also open-sourced several models, such as the video generation model WAN 2.2 and the text-to-image model QnImage. By continuously upgrading our open-source models, we're empowering our customers to develop their own AI applications. Meanwhile, Alibaba's own AI-native applications continue to advance. AMAP has undertaken a comprehensive AI transformation with the launch of AMAP 2025, the world's first AI-native location-based application. The upgrade brings spatial intelligence into dynamic real-world scenarios, and AMAP is well-positioned to become a new gateway for future lifestyle services. ThinkTalk has also completed its latest AI upgrade, creating the world's first agent-driven work feeds to explore next-generation workplace application paradigms. On our Taobao platform, we see immense AI-powered opportunities emerging, such as AI Search and AI Advertising Platform. In consumption, We undertook a strategic combination of Taobao and Tmall Group, Ulema and Fliggy into Alibaba China e-commerce group. This organizational change creates a comprehensive consumption platform and upgrades our consumer experience. We have consolidated supply chains, user bases and membership benefits across our businesses and launched a tiered loyalty program that connects Ulema, Fliggy and AMAP. The newly integrated benefits enhance our members' experience across a full spectrum of consumption scenarios. Since May, our investments in QuickCommerce have rapidly surpassed key milestones and created synergies. In August, monthly active consumers on our QuickCommerce business are approaching 300 million, contributing to a 25% increase in monthly active consumers on the Taobao app. Daily order volume of our China e-commerce group continued to achieve new records. Looking ahead, Alibaba Group has two historic opportunities to build a technology platform centered on AI plus cloud and to create a comprehensive shopping and daily life services consumption platform. We will invest at scale to capture the opportunities. This also marks a new entrepreneurial chapter for the company after 26 years. In line with this, in February, we announced an investment of 380 billion RMB over the next three years to build our cloud and AI infrastructure. In July, we announced plans to invest 50 billion RMB in consumption. The transformative impact of AI on all industries combined with a deep integration of AI and cloud will present the most significant opportunity in the technology sector over the next decade. For Alibaba, we have the world's fourth largest and Asia's leading cloud infrastructure, along with full-stack technology capabilities spanning AI computing power, AI cloud platforms, AI models, an open-source ecosystem, and AI applications. This quarter, our CapEx investment in AI and cloud infrastructure reached 38.6 billion RMB. Over the past four quarters, we have cumulatively invested over 100 billion RMB in AI infrastructure and AI product R&D. Our investments in AI have begun to yield tangible results. This is evidenced by Alibaba's Alibaba Cloud's return to rapid growth driven by AI demand and our AI enhanced experiences across consumer and enterprise facing scenarios. So we're seeing an increasingly clear path for AI to drive Alibaba's robust growth. We're also well positioned in China, the world's largest e-commerce market and the most promising service consumption market. China has a well-developed e-commerce infrastructure, high population density and strong demand for service consumption, providing a solid foundation for the integration of our quick commerce business and the Taobao app. We believe this convergence will fulfill consumer needs for a one-stop consumption experience and meet merchants' desire to serve consumers across multiple scenarios. It will enhance commerce efficiency and pave the way for an all-in-one AI assistant for consumption. Alibaba's strategic positioning in QuickCommerce has ambitions beyond competing in a single category. We aim to meet the one-stop consumption needs of our 1 billion consumers and shape business models of a comprehensive consumption platform in the AI era. In consumption, our long-term goal is to create a comprehensive consumption platform catering to our 1 billion consumers' full spectrum of shopping and daily life needs. We aim to offer the best experience to the largest consumer base with the highest purchase frequency, ultimately leading in an RMB 30 trillion addressable market. Over the next three years, Alibaba will embark on a new journey with an entrepreneurial mindset to drive robust business growth through sustained investments centered on the strategic areas of consumption and AI plus cloud. We're confident that these investments in the core business will sharpen our competitive edge and fuel long-term growth. Thank you.

Thank you, Eddie. As Eddie said, we are embarking on a new chapter of entrepreneurship by investing into strategic pillars of consumption in the AI plus cloud. These represent the two biggest long-term opportunities we are systematically pursuing. To reflect this sharpened focus, we have also adjusted our financial reporting accordingly. Starting from this quarter, we undertook a strategic combination of Taobao and Tmall Group, Erlema, and Fliki into Alibaba China e-commerce group. transforming our value proposition into a comprehensive consumption platform. This is not simply an organizational change. It's a major strategic investment aimed at redefining the consumer experience and unlock long-term value across our ecosystem. This quarter marked meaningful progress on this front as we deepen investment in quick commerce. an increasingly essential use case for capturing new demand and shaping future consumer experience. Our quick commerce business achieved the key milestones while contributing to the 25% year-over-year growth in the Taobao APP monthly active consumers in the first three weeks of August. In tandem, we are building the AI plus cloud infrastructure to support the next wave of technological transformation, positioning Alibaba as a key enabler of enterprise AI adoption across industries. Our cloud business delivered accelerated growth as segment revenue and revenue from external customers both grew 26%, driven by surging AI demand and increased customer adoption of public cloud services to support AI workloads. At the same time, we remain focused on improving operating efficiency and profitability. In the quarter, AIDC delivered solid progress, approaching break even while sustaining strong growth momentum. Now let's look at the financial results on a consolidated basis. Total revenue was RMB 247.7 billion Excluding revenue from Sunnah and InTime, revenue on a like-for-like basis would have grown by 10% year-over-year. The adjusted EBITDA decreased 14%, primarily due to our strategic focus on scaling quick commerce to capture new consumption patterns and drive future monetization opportunities, partly offset by margin improvements across several businesses, including AIDC and other units that made continued progress in operating efficiency. Our gap net income increased 76%, primarily due to the mark-to-market changes from our equity investments and the gain arising from the disposal of local consumer service business of Trendio. Operating cash flow was RMB 20.7 billion, Free cash flow was an outflow of RMB 18.8 billion. This was mainly attributed to our accelerated pace on expanding AI plus cloud infrastructure as CapEx ramped up to approximately RMB 39 billion and investment in Taobao Instant Commerce. Backed by nearly US dollar 50 billion in net cash, a healthy and low-leveraged balance sheet In our strong access to capital markets, we have ample flexibility to support long-term strategic investments while maintaining financial resilience. This quarter, we bought back approximately 7 million ADSs for a total of US$815 million in our share repurchase program. We remain committed to shareholders' return through a mix of share buybacks, dividends, and investment for growth. And we will continue to adjust the pace and form of returns based on market conditions and strategic priorities. Now, let's look at the segment results, starting with Alibaba China e-commerce group. Revenue from Alibaba China e-commerce group was RMB 140.1 billion, an increase of 10%. Custom management revenue of our e-commerce business increased by 10%, primarily driven by the improvement of take rate. We had a successful June 18th shopping festival, which delivered strong consumer growth on the Taobao APP. As we implemented user-friendly promotion mechanisms and increased the support for merchants that provide high-quality products, and the customer services. The number of ADA VIP members, our high spending consumers group, continue to increase by double digits year over year, surpassing 53 million. Revenue from our quick commerce business increased 12%, mainly due to order growth as a result of rollout of Taobao Instant Commerce at the end of April. Since its launch, we have seen encouraging business progress, reflecting strong user adoption and growing order momentum. In the meantime, we have expanded our product offerings and front warehouse coverage for non-food categories as part of our efforts to improve user experience and enhance operating efficiency. We executed our plan to generate the synergies between quick commerce and the rest of Alibaba's ecosystem by leveraging supply chains, users, and membership benefits across our businesses. In August, Taobao APP launched a tiered loyalty program that connects Alibaba Group's China e-commerce, quick commerce, and travel platforms. The adjusted epitaph from Alibaba China e-commerce group decreased by 21%. Excluding the investments in our quick commerce business, our Alibaba China e-commerce group, Epitar, has grown year over year. Revenue from AIDC grew 19%, primarily driven by strong performance in cross-border businesses. AIDC's adjusted Epitar loss narrowed significantly, approaching break-even. As we continue to improve our operating efficiency, the UE of choice and the Trendios international business improved significantly on a sequential basis. Looking ahead, we're committed to enhancing operating and investment efficiency. As a result, our profitability will continue to improve. Cloud segment revenue grew by 20%. 6%, primarily driven by public cloud revenue growth. AI revenue continued its triple-digit growth. As AI demand continues to grow rapidly, we are also seeing increasing demand of compute, storage, and other public cloud services to support AI adoption. The adjusted EBITDA margin remained relatively stable year-over-year at 8.8%. We will continue to invest in customer growth and technological innovation, including AI products and services, to increase cloud adoption for AI and maintain our market leadership. As previously mentioned, we have updated our segment reporting to better reflect our focus. We simplified the financial reporting structure by reclassifying Chai Niao, AMAP, and Hu Jing DME into All Others. All Others' segment revenue decreased by 28%, primarily due to the disposal of Sunna and Intime. All Others' adjusted EBITDA was a loss of RMB 1.4 billion, primarily due to the increased investment in technology businesses, partly offset by the improved results of businesses, including Freshable. The all other segment comprises a set of innovative initiatives, including several strategic AI-driven technology infrastructure and businesses. While we continue to drive efficiency improvements across business lines, we are also investing in AI opportunities to maintain our competitive edge and to drive future growth. In closing, this quarter marked meaningful progress in our strategic investment on two pillars that will power Alibaba's next phase of growth, our comprehensive consumption platform and AI plus cloud infrastructure. In commerce, the integration of multiple businesses under Alibaba China e-commerce group is driving stronger synergy across supply chains, user networks, and membership programs, enabling us to better serve evolving consumer needs and capture long-term growth potentials. In cloud, revenue growth accelerated on the back of robust AI-driven demand. As we expand infrastructure capacity, we are helping more customers deploy and scale their AI workloads. We are investing with clarity and conviction on the two historical opportunities ahead, AI and domestic consumption. Our commitment of RMB 380 billion technology investment reflects our long-term ambition to build infrastructure essential for AI proliferation, while our focused expansion in quick commerce is designed to unlock new demand and long-term consumption potential in China. With strong balance sheet, operating cash flow, and business momentum, we are well positioned to support these investments, drive sustainable growth, and strengthen core capabilities that would define Alibaba's future. Thank you. That's the end of our prepared remarks. We can open up for Q&A.

Hi, everyone. For today's call, you are welcome to ask questions in Chinese or English. A third-party translator will provide consecutive interpretation for the Q&A session. Please note that the translation is for convenience purpose only. In the case of any discrepancy, our management statement in the original language will preview. If you are unable to hear the Chinese translation, Bilingual transcripts of this call will be available on our website within one week after the end of the meeting. 大家好,今天的电话会欢迎您用中文或英文提问。 我们会有第三方工作人员提供实时的交替传译。 翻译目的是方便大家理解。 如有任何疑义,请以我们管理层原始语言所做的陈述为准。 如无法听到中文翻译,本次电话会议的双语记录 Operator, please start Q&A session. Thank you.

Thank you. If you wish to ask a question, please press star then one on your phone and wait for your name to be announced. If you wish to cancel your request, please press star two. And to give more people the opportunity to ask questions, please keep yourself to no more than one question at a time. Thank you. Your first question comes from Alicia Yap at Citigroup. Please go ahead.

Hi, thank you. Good evening, management. Thanks for taking my questions. Congrats on your solid cloud revenue growth. I have a question related to your recent step-up investment in the quick commerce and also the food delivery business. So can management share with us what is your vision for the quick commerce growth opportunity in China And what is your investment plan for the quick commerce? How long will the heavy investment last? And how will the investment bring the long-term value for overall Taobao and your China e-commerce platform? Can management share a bit of the latest progress of the Sun Gold, which is the quick commerce business? What are the synergy you have realized so far? and how should we expect the investment to impact our GMB and also the CMR growth in the coming quarters? Thank you.

Good evening, Director. Thank you for accepting my question.

I also congratulate you on the rapid growth in the cloud revenue. 那我想提的问题是关于其实零售 淘宝闪购这方面最近投资加大 那么是不是可以请管理层分享一下 对于在中国的其实零售增长机会 你们有什么愿景 对其实零售有什么投资计划 当前这种大额的投入会维持多久 Thank you for your question. First, I would like to review some of the stages of our retail business.

At the same time, I would like to share some of our expectations for the flash shopping business. Since Taobao flash shopping was launched four months ago, first of all, I think we have achieved success in terms of users, commercial operations, logistics construction, and market marketing. Especially from the beginning of July, our order size, users, commercial supply, and supply size are far exceeding expectations. We have already dominated the industry in terms of the order volume of delivery to home orders. I would like to share a few specific figures with you. First, our daily order volume reached 120 million orders. In August, our weekly daily order volume reached 80 million orders. Second, from the user side, the total number of buyers of monthly transactions reached 300 million. Compared with April, it increased by 200%. From the business supply side, over the past period of time, with the rapid growth of business size, a large number of new users and new business owners joined Taobao to buy, especially the high-quality supply reached the leading level of the industry. In terms of operating scale, Taobao's retail shopping has reached 2 million in total. This is three times more than in April. This means that we have created more than 1 million new jobs. In the last quarter, I also mentioned that our retail shopping's first-stage goal is first of all the size and quality of the users. After the past few months of development, our first-stage goal has exceeded expectations.

Thanks for your question. Let me begin by reviewing some of the progress we've made so far in our instant commerce business and then share our expectations for instant commerce going forward. Since we launched Taobao Instant Commerce four months ago, we believe that we've been highly successful in engaging users and merchants, building logistics capabilities and marketing. From July onwards in particular, the growth of order volume, user scale, merchant supplies, and delivery capacity have all exceeded expectations. In fact, if you just look at the food delivery to home category, we are now already the market leader in terms of orders. Let me share some figures. First, our peak daily order volume reached 120 million, and weekly average daily orders reached 80 million in August. Second, user scale. QuickCommerce monthly active consumers, MAC, reached 300 million in August, representing 200% growth compared to before April. Third, merchant supply. The rapid growth in business scale has also attracted new merchants to join Taobao Instant Commerce. High quality merchant supply in particular has reached industry leading levels. Fourth, fulfillment capacity. Daily active riders have exceeded 2 million. That is a three times increase from April. And that also means that we've created over a million new jobs. As I stated last quarter, The first stage goals for our quick commerce business were to scale up user growth and build consumer mindshare. With these developments over the past few months, we have already achieved our first stage goals beyond our own expectations.

接下来我在讲一下闪购对电商业务的拉动。 首先从用户视角,淘宝闪购显著带动了手淘整体用户规模跟活跃度。 The share price of Taobao increased by 20% in August. The share price increased by 20% in August. The share price increased by 20% in August. The share price increased by 20% in August. The share price increased by 20% in August. Next, let me talk about the synergies between QuickCommerce and our e-commerce business.

First, QuickCommerce is a significant driver of overall user scale and engagement. In August, QuickCommerce drove 20% growth in the Taobao Apps DAUs. The high frequency pattern of QuickCommerce has also contributed to a significant increase in average purchase days per user. With the increase in user engagement, we are seeing a clear trend of QuickCommerce driving incremental income for our e-commerce business. Specifically, first, The growth in traffic drives advertising and CMR growth. Second, as a result of heightened user engagement, incremental new user acquisition, and customer re-engagement, we can reduce sales and marketing expenses. We expect this trend to continue and to expand in our ongoing operations and are confident that it will drive significant incremental income on the e-commerce side.

Next, I would like to share with you our views on the efficiency and benefits of flash purchases. First, I would like to talk about the efficiency. I don't think we can leave out the scale of efficiency. In the past, our order size was one-third of the same line. In many cities, our previous share price was even less than 20%. Under such a huge market share price gap, efficiency is meaningless. At present, flash purchases have reached the top of the scale. We will quickly improve our efficiency. We are doing very well in the delivery industry, especially in terms of efficiency. We are trying to reduce this gap. In terms of efficiency, we have a huge room for improvement. In terms of short-term and long-term, our loss reduction in the short term comes mainly from several aspects, including the optimization of user structure. In the past four months, we have invested a lot of market fees and brought a lot of new users. But new users need higher investment in the acquisition stage. Our new users flow is very good. As the proportion of old customers increases rapidly, the platform's superiority will be optimized with the structure of users. Second, the optimization of order structure. In the next stage, we will continue to increase the proportion of high-value orders. Consumers' high-value regular orders and retail orders are all increasing. As the overall AOV of the platform increases, it will also bring overall superiority. Third, the optimization of contract efficiency and cost. In the early stage, In the process of increasing the order size by 4 times, our primary task is to ensure user experience. Due to short-term short-term supply shortage, in July and August, we made additional large-scale investments in supply. As the order size of the platform stabilizes, our logistics costs will significantly drop, leading to further optimization of platform advantages.

Next, let me share my views on the operating efficiency and unit economics, UE, of quick commerce, because I know these are aspects to which you pay special attention. First, when talking about operating efficiency, you cannot disregard scale. In the past, our scale was just one-third of our peers, and in many cities and provinces, our market share was even lower than 20%. With such a huge gap in market share, it's meaningless to talk about efficiency. Now, however, our quick commerce business is leading in scale and we can quickly improve our operating efficiency. Our peers have achieved excellent performance in the food delivery industry, especially in terms of efficiency, and we are now actively working to narrow that gap. Our scope to improve efficiency is substantial. Let's look at it in the short term and then in the long term. In the short term, our losses will narrow primarily driven by the following. Number one is optimization of customer mix. We have scaled up marketing expenses for user growth nationwide over the past four months and acquired a large number of new customers. New user acquisition requires substantial upfront investment, but we excel at user retention. and as the proportion of repeat customers increases, UE will improve. Number two is optimization of order mix. In the next stage, we aim to increase the proportion of high-value orders, including high-value meal orders, and increase the proportion of non-food category orders. UE will therefore improve as a result of higher average order value, AOV. And the third is optimization of fulfillment efficiency and costs. In the initial phase, as order volume ramped up by four times, the number one priority for us as a platform was to ensure the user experience. In July and August, therefore, we made an additional large investment to address the severe short-term lack of delivery capacity. Going forward, as order scale stabilizes, our logistics costs will decline significantly, contributing to further UE improvement.

短期通過物流跟補貼效率的提升, 還有用戶訂單結構的優化, 我們預計在保持消費者當前優惠投入的情況下, UE虧損可以縮減一半。 長期看,隨著我們訂單密度的顯著變大, 我們的物流成本對比4月之前依然有很大的提升空間。 In addition, there is a lot of room for improvement in our detailed operation and trust business. I think scale is the first factor that determines efficiency. With the new scale and market share, we are confident that we will reach the leading level of the industry in terms of long-term efficiency. At the same time, we will not look at the profit and loss of takeout alone, but consider the comprehensive profit of e-commerce. We believe that we can keep the price competitiveness under the premise of long-term price competitiveness, and buy back the overall positive economic profit of the platform.

So in the short term, we expect that while continuing to maintain investment in consumer benefit through logistics and subsidy efficiency improvements and order structure optimization, our UE losses can be reduced by half. In the long term, as order density increases, there is considerable scope to optimize our logistics costs compared to four months ago. In addition, there is also considerable scope to improve our targeted engagement of offline merchants. In my view, scale is the primary factor in achieving efficiency. With our newly achieved scale and market share, we are confident that we can achieve industry-leading efficiency in the long term. At the same time, we will not look at the standalone profitability of quick commerce delivery. If combined with the incremental benefits to our e-commerce business, we believe QuickCommerce will bring sustained positive economic value to the overall platform while maintaining price competitiveness.

Then I will talk about the development of non-commercial retail products. First of all, we divide the non-commercial retail retail into two parts. The first part is the in-store native mode. The second part is the in-store combination. Under the in-store raw material model, on the one hand, we rely on strong commodity supply and supply chain to develop the lightning warehouse model. In the past few months, the lightning warehouse supply has developed rapidly. Our lightning warehouse has more than 50,000 orders, and the order volume has increased by more than 360%. Of these, 25% of the lightning warehouse supply comes from the Alicentai supply chain. Secondly, we rely on Hema to greatly develop the front-end warehouse contractability of fresh products. After Hema was involved in Taobao's flash purchase, Next, let me talk about development of the non-food category.

In QuickCommerce, we divide non-food deliveries into two parts. One is the original hyperlocal QuickCommerce e-commerce model, and the other is e-commerce, a hybrid model of QuickCommerce plus e-commerce. On the hyperlocal side, we have leveraged our abundant supplies and strong supply chain to develop a lightning warehouse model Over the past few months, supply in these warehouses has rapidly expanded. We now have over 50,000 Lightning warehouses with order growth of over 360% year on year. 25% of the supply in these Lightning warehouses comes from supply chains within the Alibaba ecosystem. Secondly, we have fulfillment capacity from front warehouses that have been developed by Fresh Ipoh for its fresh groceries category. Following Fresh Ipoh's supply connection into Taobao Instant Commerce, order volume has already exceeded 2 million, up by 70% year-on-year. In terms of the hybrid model, Tmall Supermarket is upgrading from a traditional B2C fulfillment model to a quick commerce model. This maintains its price competitiveness while achieving much faster shipping speeds.

At the same time, we will actively introduce Tianmao brand offline stores to enter Taobao flash shopping to realize Tianmao brand offline integrated management. We expect that in the future, there will be millions of offline stores to enter Taobao flash shopping. The combination of Tianmao and Taobao flash shopping will bring new business growth to the brand, and also bring a new shopping experience to the consumer. In general, We expect that in the next three years, retail and real-time retail will bring 10 million new deals to the platform. At the same time, we also believe that the retail industry, from a single family to multi-platform participation, will give the rest of the consumers more choice. Long-term, it will definitely be beneficial to the industry. In this process, the platform's investment in Zhenning Baiying has created more than one million jobs, and while promoting the transformation of the industry, it has also promoted consumption and economy. Thank you.

We are also actively onboarding Tmall Brands offline stores into Taobao Instant Commerce, enabling unified online offline O2O operations for Tmall Brands. We expect up to 1 million branded offline stores to join Taobao Instant Commerce over time. The integration of Tmall and Taobao Instant Commerce will open up new growth drivers for brands and offer consumers a new shopping experience. Overall, we expect Taobao Instant Commerce and QuickCommerce to add 1 trillion RMB in annualized incremental GMV to the platform within the next three years. We also believe the food delivery markets shift from a single dominant player to multi-platform competition gives merchants and consumers more choice, which benefits the industry long-term. Throughout this transition, We as a platform are committing real financial resources and investing to create over 1 million direct jobs, drive industry transformation, and stimulate consumption and the broader economy. Thank you.

Next question, please.

Thank you. Your next question comes from Thomas Chong at Jefferies. Please go ahead.

Hi, good evening. Thanks management for taking my question and congratulations on a very solid set of results. My question is about Alibaba Cloud. We have seen Cloud business is doing very well and accelerating to 26% year-on-year this quarter. My first question is how should we think about this acceleration? Should we expect acceleration continues for coming quarters as well as our expectation for FY26. Because when I look into our acceleration and look into overseas peers, how should we think about the pace of monetization versus the U.S.? And on the other hand, we also see our crowd margin reaching 8.8%. How should we think about the margin outlook into the future? And on that front, given our acceleration is very impressive, can we talk about how different industry sectors actually perform in this quarter? I remember last quarter we talked about multiple sectors, including traditional sectors, also embrace the cloud opportunities. Are we really seeing something different or making a lot more progress? for this quarter. And on that, CapEx, how should we think about our CapEx outlook given we have seen our CapEx is very likely during this quarter? Thank you.

感谢管理层接受我的问题, 也祝贺你们本季度取得的扎实的业绩。 The question I would like to ask is about Aliexpress. We see that Aliexpress's business is doing very well. The increase in this quarter has reached 26%. My first question is to understand how we should think about its increase and acceleration. In the next few seasons, or the entire fiscal year of 2026, will we continue to maintain this increase? So considering this increase and comparing it to overseas, for example, compared to the United States, how should we think about our rate of change? On the other hand, I see that the profit rate of our cloud Now it's up to 8.8%. So how should we think about the future profit and loss outlook? Because on the one hand, our growth is very fast. Uh, uh, uh, uh, uh, uh, uh, uh, uh, uh, uh, uh, uh, uh, So these are the questions, thank you.

Okay, thank you for your question. There are still quite a lot of questions. Let me answer the first question first, about our real-time expectations. From what we can see now, our Aria Cloud customers are using AI products, or within their company, When we were developing AI products, we saw some very clear trends. From this trend point of view, we feel that there is a lot of demand. This is mainly reflected in two aspects. We saw a lot of companies, because the AI model is getting stronger and stronger, they are developing a lot of new applications or creating a lot of new user cases. On the other hand, we also saw that a lot of traditional application functions to quickly use the ability of big models to replace traditional CPU calculations. So when we look at this trend, we see that the needs of customers are still quite abundant. On the other hand, we also see that the growth of the demand for training and reasoning. In recent seasons, we have seen, in addition to the rapid growth of the demand for training and reasoning, we have also seen some new industry observations in the field of training. In addition to the company with some basic big models in the industry, we have also seen many new opportunities. For example, like our car factory, for example, like our education-related enterprises, or some multi-media application-related companies, they now have some needs to train them with their own exclusive data to train exclusive models. And these needs for AI, for the use of our AI's overall infrastructure, will actually be a a relatively good promotion. On the other hand, we also see some new opportunities in training, that is, many companies, based on the open source model that we agreed on, actually have a very strong demand for post-training. For example, we just mentioned these education companies, or some related to the medical and medical industry, and a lot of development tools, such as these platform companies. In fact, we found that there are some very good opportunities here. Many of their exclusive models will use their own data and their own scenarios to do post-training on our integrated model. On this basis, it will naturally use our Aliyun cloud computing platform. At the same time, we will gradually develop some commercialized services for our open source model post-training. This is also a new commercialization opportunity we have seen in the open source model. So we expect that in combination with the rapid growth of reasoning needs, we expect that in the next few seasons, the overall growth rate of Aliyun will still be in a process of continuous improvement.

Okay, thank you very much for those questions. And in noting them down, I saw there were quite a few questions, but let me start with the first one regarding the outlook for our growth rate. So we certainly see a very clear trend among our customers in terms of utilizing AI products. and they're developing AI products, and this represents a very strong demand. First of all, as AI model capabilities continue to strengthen, more and more new AI applications are being developed and applied into more and more new use cases. At the same time, many vendors that had original use cases where they were running workloads on traditional CPUs are now beginning to leverage large models and using AI to run those original functions. So for all these reasons, demand is very strong and continues to grow robustly. In recent quarters, we've seen fast growth in inference workloads as well as continued growth in training and in inferencing. Over the past few quarters, apart from that rapid growth in inferencing, we've also observed some new industry trends on the training side. Apart from the large fundamental models that continue to be upgraded in the market, there is also a lot of training that is taking place and creating new opportunities. For example, automobile vendors, for example, companies in the education sector, or companies in the multimedia application space. They all have training requirements. They're leveraging their own proprietary data to train proprietary models to meet their demands. And that, in turn, is driving higher utilization of our overall AI infrastructure. So this is a very strong boost to infrastructure use. At the same time, we also see new opportunities around training whereby companies are leveraging our open source models and they have a lot of training demands in doing so. For example, we mentioned education companies as well as companies in the healthcare sector and companies that are creating development tools, platform companies. So we see a lot of excellent opportunities with these proprietary models where they're leveraging their own data to meet their own use cases by performing post-training or fine-tuning of our QN models. And on that basis, they're making use of our Alibaba Cloud computing platform. And at the same time, on that basis, we can also begin gradually to develop commercialized services, post-training services for our open source models.

Okay, the commercialization rhythm of the overseas market and our net profit, I think, can be summarized as one problem. I think the answer to this problem is actually based on our judgment. What we see now is our own judgment on the Chinese cloud computing market. We think that because of the change in the trend of AI for the cloud computing market, we think that the market concentration will be much higher than the traditional cloud computing market concentration. Because of the appearance of AI technology, there will be many more technology stations that developers need to develop applications. Therefore, we will choose with all-round technical product combinations, especially on each single product, with leading capabilities. And Aliyun has an advantage on traditional computing storage, database, and big database products. At the same time, we have an advantage on all-round leading capabilities on new AI technology stations, AI algorithm, and AI Infer, AI model capabilities. So we have a more friendly open-source environment for developers. Therefore, At this stage, our goal is to use the opportunity of the AI age to continue to expand the market share of Alibaba in the Chinese market or beyond the average market share. So in this strategic target situation, our overall priority is to acquire more users, acquire more users' application scenarios, rather than focusing on increasing our profit margin now.

Your other question had to do with gross margin and benchmarking against what we see overseas. So perhaps I can share with you my judgment overall as to the Chinese cloud market. And what I see as the trend is for there to be a higher level of market concentration in China as compared to what we see in overseas markets because developers here require full-stack comprehensive capabilities for what they're developing, and we have that full-stack with complete capabilities across traditional cloud computing, compute storage, AI, compute our AI models, and of course we have a very developer-friendly open ecosystem as well. So our goal is to maintain a growth rate above the average rate of market growth to at the same time increase our market share. And in terms of our strategic objectives, the priority is on growing the number of users, expanding into new use cases and not in the short term on increasing gross margin.

Okay. AI方面的Kapex支出我觉得我们还是会坚持原来三年投资3800亿这样的一个计划但是每个季度可能根据供应链的不同可能会有一些波动同时呢根据根据全球的AI芯片的供应和政策的变化我们其实也有后备不同的和别的合作伙伴共同去做供应链的不同方式的一个储备所以我们相信面对不同的 So turning now to the question about our CAPEX capital expenditures on AI.

We will continue to implement our three-year plan to invest $380 billion in cloud and AI. However, based on the supply chain situation in different quarters, there may be quarter-to-quarter fluctuation. At the same time, based on changes in policies around AI chips and supply, We have backup plans in place to work with various different partners and to be able to respond to different situations in respect of supply chains. So I'm confident that no matter what changes may crop up in the industry, we will continue as planned and as expected to move forward with that plan capex investment of $380 billion.

Next question, please.

Thank you. Your next question comes from Kenneth Fong at UBS. Please go ahead.

Hi. Good evening, management, and thanks for taking my question. Given the cross-sellings that we have seen already achieved in our food delivery part, do we plan to ramp up the in-store part of local service, i.e. Dao Dian Ye Wu, because we noticed that LMA Kao Bei in-store coupon have been stepping up promotion in certain regions. So how should we think about any further investment or expand in this in-store part of business in the coming months on top of the food delivery? Thank you.

Good evening, Mr. Guan. Thank you for accepting my question. As we have seen in the delivery section, the cross-exploitation and sales results have been achieved. Will we plan to expand the delivery and consumption content in local life services? We have also noticed that the consumer advice 我们每天的这个闪购业务的频道的这个用户已经是非常大的一个规模了。

about 1.5 billion customers every day. We also see that some of our customers also have the need to order themselves, including ordering and purchasing. So we will look at it from the point of view of satisfying the needs of our customers, especially with our offline business. With our delivery business, we will also consider providing more diversified services to our customers. So we are also doing some testing and exploration in some cities.

Yes, thank you.

Let me take this question. So the first thing I would say is that this is very much connected to the scale that we have achieved in QuickCommerce that I just shared with you. We have a massive scale now of users on our QuickCommerce channel every day. Basically, we're talking about 150 million active users on a daily basis. And within those users, there are some who go to the store themselves to do self-pickup, and that also includes group purchase using coupons. So from the perspective of satisfying the needs of our users, and in particular those who go offline in-store, there's a lot of synergy there, and so we will also consider providing more diverse services to these users. And in fact, we're already doing some testing and piloting in certain selected cities.

Next question, please.

Thank you. Your next question comes from Joyce Ju at Bank of America. Please go ahead.

Thank you for giving me the opportunity to ask this question. My first question is about the three-year commitment of the officials mentioned by the management to invest in AI and internal demand. These two major historical opportunities. First, because we know that in terms of AI, we have a 3.8 billion investment plan for three years. So in terms of internal demand, how do we consider the rhythm of this investment? It covers part of it, except for what we just mentioned, which is timely retail and takeout. which will also include, for example, the investment in the supply chain or consumer side, and then what will be its rhythm. Then my second question is about the growth of our core CMR. This quarter, we have maintained a very strong 10-point annual growth, but after September this year, the problem of the technology brought by the increase in basic service fee last year may gradually begin. 我们接下去去怎么理解这个全战推渗透所驱动的比如说未来几个季度的这个CM的这个持续的这个take rate的增长。 Thank you management for taking my question.

My first question has to do with the pace of investments going forward. You mentioned your commitment to the three-year plan to invest $380 billion in cloud and AI. I'm wondering if you could also tell us about your investment plans going forward on the commerce side, on the consumption side. What will the pace of those investments look like apart from investing in quick commerce, what other investments are you contemplating, for example, in supply chains or in users, and what will the pace of those investments look like? And then secondly, I'd like to ask about CMR. So we saw a strong growth in CMR this quarter, up 10% year on year, but we know that heading into September, the positive impact of the software service fee that you started implementing last year will diminish so looking forward how much positive impact will qct penetration have on cmr growth and beyond that the incremental traffic and gmv being driven by quick commerce how much positive impact will that have on cmr thank you

Okay, thank you for your question. Let me answer this question. First of all, regarding our investment in the consumption of such a historical opportunity, in fact, we are not just starting to invest now. In this regard, we have been maintaining our investment, whether it is in the supply chain or in the user side. Here, through our different businesses, including Taobao Tianmao, we used to make a lot of investment in the user side. For example, in the supply chain, whether we are in Taobao or Tianmao, we are in Hema, etc. These are some of the investments on the supply chain. In recent years, we have mentioned such a large investment of 500 billion yuan, and we have seen such a significant investment from the long-term market to the long-term market consumption. In the case of such a large historical opportunity, we have increased such a 500 billion yuan investment. From the perspective of investment, we will control the rhythm of our investment according to the specific market situation, At the same time, based on the development of the entire market to determine our investment investment, the final one is a big one. Such some investment ultimately makes us the investment of the past plus this investment ultimately makes us have a very strong ability in the consumer market. It also allows us to do such a integration this time, and to grab such a historical opportunity together. So this is an investment in the consumer side of our overall consideration.

Thank you. Let me take this question. So given the historic opportunity to invest in the consumer market today, we're certainly grasping that, but our investments in the consumption market are not just starting now. We've always been making investments in the user side, certainly the Taobao and Tmall group, had been investing in users on the supply chain side, not just the TABO and Tmall Group, but also FreshHippo and other of our businesses were investing in the supply chains as well. So when we talk about this large investment that we're now making in the consumption market of 50 billion for developing quick commerce, this is incremental. This is a new investment on top of those investments that we were making, we will pace the investment cadence and rhythm in line with the market circumstances and market developments to ensure that it makes sense.

你的第二个问题是关于这个CMR,我想这个最关键还是这个take rate的这个影响,那这个季度我刚才说了,这个季度的CMR的增长最主要还是因为在take rate上面的十分大的改善。 There are two main reasons for the improvement of take rate. The first is the service fee of our 1,600 users. The second is the increase in the penetration rate of the AI-powered full-time推ed commercialized products. For these two points, in the following few seasons, we will still see a better positive impact on take rate. At the same time, we see a very positive impact on retail shopping, whether it is in terms of users or in terms of buying bottles. In these aspects, it will also increase the positive impact on us, including on CMR. So from the current situation, we expect that in the next few seasons, Thank you.

So your second question had to do with CMR, and I think the key point there was the impact of take rate. In this quarter, as we said, CMR growth was positive and robust, and this was mainly because of a large increase in take rate. And there were basically two major reasons for that. One was the 0.6% software service fee, and the other is QZT, which is itself powered by AI, continuing to achieve deeper penetration. And I think that in the coming quarters we can expect to see also positive impacts on CMR from these aspects. At the same time, there will be positive impact on CMR from quick commerce as well, where we're driving user growth as well as higher user frequency, which will result in higher take rate. So we would expect that in the coming couple of quarters, you can expect to see relatively rapid growth in CMR as you are now seeing.

Next question, please.

Thank you. The next question comes from Yuanliang from Citix. Please go ahead.

Hello, Mr. Guan. Thank you for answering my question. We have been paying attention to the progress of the company model. In the Q&A block, we actually saw that Q&A 3 is from the era of training models to the era of training agents. My question is, Thanks.

Good evening, management. So we've been following very closely the developments of your models and looking at the Q1 3 block. what we see is that it marks a transition from an era where the focus is on training to one where it's all about agents, an agent-centered era. So my question is, What additional capabilities and resources or investments do we need in this transition period? And can you tell us a bit about some of your latest developments around agent products and agent applications?

Okay, let me answer your question. We can now see that the glasses of the entire model are made from the original It's just a simple chat box, and now the glasses reach the entire Igentic glasses. In the Igentic model of this era, we now see a few very obvious trends. We need the model to be able to deeply understand the upper and lower window of the upper and lower window. Only the upper and lower window of the upper and lower window can solve a more complex task or a longer thinking chain. At the same time, we also see that our model needs to use various different tools to be able to integrate various different internal systems of enterprises. This is also the ability of the models to learn. On the other hand, due to the outbreak of AI Aging, we also see that the use of Aging or the infrastructure behind Aging is actually a new opportunity for cloud computing companies. For example, when we see a lot of agents that need to solve a task, they need to run a lot of virtual machines, or open a lot of browsers, or at the same time open a lot of mobile virtual machines. In fact, Aliyun also has a new product. We call it Agent Bay, which is specifically able to provide an underground sandbox environment for agents. I think the whole model goes into In this age of agent drive, I think it's a very good opportunity for cloud computing manufacturers and for Alibaba. Because Alibaba carries a lot of companies from different industries. At the same time, we have a very good environment for agent operation. At the same time, our model is also evolving and adapting to the model used by agents.

Thank you. Let me take that question. So indeed, we do see the overall evolution path when it comes to models going from simple chatbots toward agents. And there are several trends that we can see in the era of agents. One certainly is that models require larger context windows to be able to undertake more complex tasks with longer chains of thought. They need to be able to utilize multiple different tools or chains of tools. And the agents need to be able to access different systems, call on different internal systems within the enterprise. So these developments also create new opportunities for us as an infrastructure provider. Many of these agents will need virtual servers or they'll need to have many browser windows open simultaneously or they'll need many mobile virtual machines and sandbox environments. So for a cloud vendor like Alibaba Cloud, these are all very positive developments. We're supporting numerous clients in numerous different sectors, and we can provide all of these with a new product called Agent Bay, which provides a sandbox environment specifically for agents. So this is very much an agent-driven era in AI and cloud development, and for us, this is an excellent opportunity.

Okay. I would like to add two more points. We are thinking about this issue. Firstly, I would like to add one more thing about models. Especially in the age of AI driven by Agents, the ability to code models is very important. Because we think that in the future, most of the scenes can be combined with a model that is good at coding, with a lot of tools, a lot of business systems can actually solve a lot of these tasks within the company or some more complex tasks for consumers. On the other hand, I would like to mention that in terms of aging products, we still have a lot of products that are combined with the entire big ecosystem of Ali Group. For example, we now have a lot of companies and e-commerce companies that need some customer-oriented products. In addition to providing basic algorithms and basic facilities, Aliyun is also cooperating with Taobao, Dingding, Gaode, and Alipay to provide more tools that can automate tasks at the business level. Through these tools, AI Agents developed by our companies can solve problems better.

Let me add to that by sharing some more of our thinking about models as they relate to agent-driven AI. I think that coding capabilities will become very important. because models with strong coding capabilities can be connected to lots of different tools and lots of different enterprise internal systems. And in that way, they'll be able to solve for many tasks within the enterprise or even to solve for some more sophisticated tasks on the consumer side. Another point I should add with respect to agentic products is that we have quite a few products that can tie into that. They're excellent within the Alibaba ecosystem. For example, a lot of enterprises, especially e-commerce enterprises, need a client service product Within Alibaba Cloud, apart from providing compute and infrastructure, we're also able to synergize with Taobao, with DingTalk, with AMAP, and even with Alipay in order to be able to provide more kinds of automated solutions at the business level. So through these tools, these enterprises can develop AI agents that are better able to solve for these various tasks within the enterprise.

Next question, please.

Thank you. Your next question comes from Alex. Please go ahead.

Go ahead. Thank you, Guanli, for accepting my question. 我想再问一个关于这个外卖和闪购的问题 那其实这次也不是我们第一次这个主动进攻这个市场 在18年我们把俄罗马收购进来以后 其实在之后的两三年 还是花了很多的这个精力和资源去尝试这个进一步的加大我们在外卖这个行业里面的市场份额 But in the last two or three years, from the results, our strategic intention has not been realized. I also asked Deep Seek why Ali's mother-in-law still can't beat the US. Deep Seek's answer was that her son can't beat the poor entrepreneur. Regardless of what Deep Seek said, we must have gone through a lot of review and thinking. In this round of large-scale attack on takeout, thank you

Thanks, management, and good evening. I have a question coming back to quick commerce or instant commerce, because it's not the first time that Alibaba has tried to enter this market. From the acquisition of Ulema back in 2018, lots of time, energy, resources were spent on trying to grow our market share in that space, but it seems that over a period of two to three years, that original strategic intent really wasn't realized. So I asked DeepSeek why it is that Ulema under Alibaba still couldn't beat Meituan and be bigger than Meituan. And what DeepSeek told me was that the son of the rich merchant can't beat the entrepreneurial wolf who was raised by a poor family. So anyway, it doesn't matter whether DeepSeek is right or wrong, but I'm sure that we've done a lot of reflection and thinking going into this new round in the battle. So I'm sure that you will be taking some different strategies or approaches to maximize success. But if you could tell us simply what's different this time and what are the different tactics that we're taking?

Yes, I think you can ask DeepThink again, but I can answer it first. I think that in fact, we have been buying and investing in Euroma for many years. First of all, I think Euroma has made a lot of progress in recent years. This progress is not reflected in the market share. Of course, because the market share is actually related to our investment, also related to our strategy, and also related to our flow of investment in various aspects. However, in the past year, I think the infrastructure has been greatly improved. Otherwise, we wouldn't be able to make such a rapid progress in such a short period of time. We wouldn't have been able to do 1.2 billion orders in four months. And we wouldn't have been able to deliver such a good user experience. So I think these are the capabilities of Ermo in the past years.

Thanks. Well, I guess you could ask Deep Sika again, but I'll take a stab at answering that. It's true that we invested, we acquired Ulema and have been investing in it for many years. But I think I would begin by saying that Ulema has actually made a lot of progress over these years, and that may not be reflected in market share, because market share is is related to our investment as well as our strategy, as well as traffic. But I would say that over these years, Ulema has made tremendous improvement in terms of infrastructure development and capabilities development. And if that weren't the case, then it simply wouldn't have been possible for Taobao Instant Commerce to have achieved such rapid development in such a short period of time. We couldn't start from zero and build this business from scratch and in just two to four months be able to service 120 million orders and not just 120 million orders but to do so with an excellent user experience. I think that all of that credit goes to Ulema and the experience and capabilities that we built up following that acquisition.

Yes, and this business needs to have enough business owners, a sufficient number of operators, and a sufficient number of C-end users. Only when these three are available can our investment be efficient. If these three, one or two of them, are very efficient, If so, then our investment efficiency will be very low. So today, coincidentally, this is a combination of Elon Musk and Taobao. We can see that we have a large number of users like Taobao, and they are all very active users. Then we have a business like this based on Elon Musk. Then at the same time, we also have a whole supply system like this, which is a logistics system. So we can make such an investment on this basis. Then the second is that our investment This kind of logic is not only based on the delivery business alone, but also on the comprehensive income of the entire e-commerce industry in the short term, medium term, and long term. Our thinking logic is also different from before. Of course, we have to do this business well. We still have a lot of work to do.

To get this business right, you need to have enough merchants, you need to have enough fulfillment capacity or delivery capacity, and you need to have enough consumers, users on the consumer end. And if you're lacking in any one of those things, then your investment efficiency is going to be poor. So in terms of where we are today following the integration of Ulema into the Taobao app, we have a vast number of users on Taobao and these are highly active users. We also have the merchant base that has been built up on Ulema and at the same time we also have the fulfillment capacity, the logistic system that we have built up. So that's the foundation on which we're making this investment. And then the second thing I would say relates to our investment logic, the way we think about this, because we're not simply looking at the quick commerce business on a standalone basis. We're really looking at what it can do, the overall incremental positive benefit it can provide for our overall e-commerce business in the short term, in the medium term, and in the long term. So the logic and the way we're thinking about this is different from the past. And of course, for us to get this business right and succeed in all those goals, there's still a lot of work that we need to do.

Arvinder, let's take the last question.

Thank you. Your next question comes from Gary Yu at Morgan Stanley. Please go ahead.

Hi. Thank you, management, for the opportunity. I have a question regarding return on invested capital. Given that we are now spending close to $50 billion in probably a couple of months on quick commerce, how should we look at the rate of return of investor capital from these investments? Because I would imagine if the same amount were to be spent on something on AI, maybe with much bigger TAM and much bigger, faster growth in cloud, we may be able to see a much better return on investment. So how should we think about how we allocate capital internally between retail and AI investment going forward? Thank you.

Thank you, Mr. Guan, for giving me the opportunity to ask a question. My question is about the return on investment. It seems that we are In a few months, we invested about 5 billion yuan in real-time retail. How should we look at the return rate of these investments? Because I think if we invest the same amount of money, for example, in AI, because the cloud increases faster, and the potential market size This so-called TAM may also be bigger. So internally, how do we distribute our investment between retail and AI? What are your thoughts on this?

Okay, Gary, let me answer this question. Thank you for this question. I think first of all, I think whether it's Eddie or me today, we have all mentioned that we have now seen the two biggest opportunities in history. Thank you.

Well, as Eddie and I have both mentioned already on the call, we are faced with two huge historic opportunities today. The first of these is AI, which we've been discussing, but the second is consumption. And the opportunity is to transform consumption on that journey from e-commerce to local, hyperlocal e-commerce to instant commerce, And we absolutely need to firmly grasp both of those opportunities. They're both crucial.

对于在这两个领域的投入来看,当然这个金额是十分巨大的。 从我们自己的角度也是历史性的这么样一个大的投入。 但是就像我刚才在我准备的这个里面讲到的, Whether it is from the perspective of ability or from the perspective of resources, whether it is from the perspective of our cash reserves or our cash flow, or from the perspective of our entire asset balance sheet and our entire ability, it allows us to have enough resources to make an investment in these two aspects using what Eddie called a safe investment. For us, the most critical direction of this investment is how to balance a relationship of short-term and long-term returns. Using an investment in AI, I think everyone can clearly see that because of such an investment, it brings a continuous increase in the whole increase of our cloud. Eddie just mentioned that not only this quarter, but we expect that in the coming few quarters, it will also further increase. Here, of course, coming back from the point of view of the entire EBITDA margin, because of the whole investment of an increase, our current first priority is still this investment. So from the point of view of profitability, we may see relatively little, but it does not mean that we do not pay attention to profitability. So in terms of AI, for the investment in retail, as I just said, from the point of view of investment now, although we Taiwan. 我们需要很好的平衡好这一点

So in terms of our investment in both of those areas, of course, we're talking about very substantial, very large investments. And from our own point of view, these are historic investments that we're making. As I said in my prepared remarks, be it from the perspective of our capabilities or our resources, I want to talk about resources that could include our cash reserves as well as our cash flows. as well as resources across our entire balance sheet, we have sufficient resources to be able to make these very substantial investments, as Eddie said, investments at scale in both of these huge opportunities. So the key for us really is how to balance short-term versus long-term returns and where the priority should lie. And we can look at that in the context of our AI investments that we've been making. And I think you will also have seen very clearly already that these investments have already driven increase in the growth rate of our cloud business. As was just said earlier, that's not just for this quarter, but we expect the following few quarters to also see increase in the growth rate in cloud. So, of course, we will also look at the business from the point of view of EBITDA margin, but our first priority at this point is on making these investments and not making the profit rate the higher of those two priorities. That doesn't mean that we're not looking at and we don't care about profit rates, of course. So in terms of AI investments and the investments in quick commerce, as Jiang Fan just said, we're not yet making any money from those investments in instant commerce, but we can already see very clearly that with the integration of instant commerce into the Taobao app, that's driving increased traffic, it's driving increased frequency, and as a result of that, that's also driving increased advertising as well on the platform. So with the progression from e-commerce to instant commerce, All of these investments are going to drive good returns going forward, and I'm confident that we have sufficient resources to be able to do this and to remain focused on what is truly critical. I think it's important to note that we should not simply focus on short-term returns and by doing so lose out on the opportunity to make those long-term returns. So this does require the proper balance.

Thank you everyone for joining us today. We appreciate your time and we look forward to speaking with you again soon.

Thank you. That concludes our conference for today. You may now disconnect your lines. Thank you.