Alibaba profit triples as AI spending weighs on cash flow | Magica
Alibaba profit triples, but AI spending keeps free cash flow negative
Editorial Team
••📖8 min read
Alibaba’s September-quarter profit more than tripled as investment gains and a commerce recovery amplified stronger cloud earnings. Cloud revenue rose 53%, but AI application losses widened and infrastructure spending left free cash flow negative. A large share placement bolstered liquidity.
Profit attributable to ordinary shareholders rose 207% to 64.535 billion yuan, boosted by investment gains and stronger operations.
Cloud and compute revenue grew 53%; commerce delivered the largest improvement in segment earnings.
Free cash outflow reached 34.472 billion yuan, worse than a year earlier but smaller than in the June quarter.
Alibaba Group, the Hangzhou-based commerce and cloud computing company behind Taobao and Tmall, more than tripled quarterly profit while spending heavily to expand its AI business. The rebound shows stronger cloud earnings, but investment gains and a recovery in commerce make it an unreliable shorthand for returns on AI.
Alibaba’s company-reported group net income before attribution to ordinary shareholders, alongside non-GAAP net income excluding investment gains and losses and other adjustments. Source: Alibaba Group.
Investment gains amplify the profit rebound
Alibaba reported profit attributable to ordinary shareholders of 64.535 billion yuan for the three months ended September 30, 2026, up from 20.990 billion yuan a year earlier. Revenue rose 10% to 271.519 billion yuan, according to its November 19 results announcement.
The stakes extend beyond one quarter. Alibaba has pledged more than US$53 billion over three years to AI hardware and software. A report on the results said the company raised about US$10.2 billion through a Hong Kong share sale in August, after its chief executive said spending would exceed the original targets. The expansion therefore involves both a larger investment ambition and fresh shareholder capital.
The headline profit exceeded the 23.72 billion yuan FactSet analyst consensus cited in a separate earnings report. Revenue was only modestly above the 270.5 billion yuan average analyst estimate in the first report.
Anthropic raised $81.3 billion in a record IPO and opened 34% above its offer price before retreating. Rapid revenue growth and adjusted operating profits sit alongside vast computing commitments and limited investor control.
Group net income, before attribution to ordinary shareholders, rose 213% to 64.420 billion yuan. Interest and investment income, net, reached 73.305 billion yuan, against 20.092 billion yuan a year earlier. Alibaba attributed that increase chiefly to higher mark-to-market gains on equity investments, partly offset by higher investment impairments.
Operations also improved: operating income rose to 29.116 billion yuan from 5.365 billion yuan. Alibaba’s non-GAAP net income, which excludes investment gains and losses, non-cash share compensation and other adjustments, rose 92% to 19.909 billion yuan. The six-month comparison is less strong: non-GAAP net income for the period ended September fell 7% to 40.624 billion yuan.
Cloud growth reaches outside customers, while applications lose more
Alibaba’s AI Cloud and Compute Services segment combines the former Cloud Intelligence Group with its chip-design subsidiary, T-Head. It supplies computing infrastructure and services to outside customers and other Alibaba businesses. This reporting structure began in the June quarter; Alibaba recast the prior-year figures to match.
Quarterly revenue reached 60.973 billion yuan, up 53% from 39.914 billion yuan. Its adjusted EBITA margin increased to 14.5%, from approximately 10.0% calculated using the prior-year segment figures. This company-defined earnings measure excludes interest, taxes, investment income, amortization and certain other items.
Growth was stronger outside Alibaba than within it. The segment revenue breakdown shows about 47 billion yuan from external customers and 14 billion yuan from internal customers, rounded to whole billions. External revenue rose 64%, compared with 23% internally.
Recurring external revenue, excluding compute equipment solutions, rose 67%. Alibaba includes pay-as-you-go services as well as monthly and annual subscriptions in that measure, principally public cloud services and license and software renewals. It should not be read as fixed subscription commitments alone.
AI-related product revenue within the cloud and compute segment was 17.918 billion yuan, about 29% of its total, calculated from the disclosed figures. Alibaba said this was the thirteenth consecutive quarter of triple-digit annual growth. The broader segment’s 53% growth rate thus covers more than AI products.
The separate AI Labs and Applications segment brings together model development, consumer AI services and QwenWork, Alibaba’s enterprise AI agent platform. Its Qwen family of large language and multimodal models powers services across the company. The consumer Qwen app provides assistance across shopping, travel, work and other tasks, drawing on Alibaba’s services and third-party partnerships.
That business grew without approaching profitability. Revenue rose 51% to 4.435 billion yuan, but its adjusted EBITA loss more than tripled. Alibaba attributed the deterioration to greater investment in AI capabilities and higher inference costs for Qwen—the computing cost of running models to answer users’ requests.
The applications segment’s loss exceeded the cloud segment’s adjusted EBITA profit. Internal transactions and group reconciliation items mean these figures are not a consolidated AI profit calculation. They nevertheless show that stronger computing-service earnings coexist with growing costs elsewhere in the AI business. The announcement does not provide per-request costs or prices that would establish how Qwen’s usage translates into sustainable application margins.
Alibaba’s company-reported adjusted EBITA for selected segments, with prior-year figures recast. Other businesses and reconciliation items are excluded; these figures are not consolidated AI profit. Source: Alibaba Group.
Commerce contributes most of the operating repair
Commerce delivered a much larger increase in adjusted EBITA than cloud. The like-for-like segment figures in the results show where the improvement came from:
Adjusted EBITA, billion yuan
September quarter 2025
September quarter 2026
Change
Alibaba E-commerce Group
11.336
38.851
+27.515
AI Cloud and Compute Services
4.000
8.838
+4.838
AI Labs and Applications
−3.435
−10.503
−7.068
Changes are calculated from Alibaba’s recast figures, before group-level reconciliation items.
The e-commerce group now combines Alibaba’s domestic and international commerce operations with Freshippo and some commerce businesses within Cainiao. Alibaba attributed the profit improvement chiefly to better performance at Taobao Instant Commerce, its rapid-delivery business, and other commerce operations. It said quick commerce improved unit economics while maintaining market share, helped by higher average order values and a greater focus on higher-value food and non-food orders. These are the company’s operating and market-share claims.
The earnings recovery was much stronger than sales growth. Total commerce segment revenue rose 2%; China e-commerce revenue fell 5%, while China Quick Commerce revenue rose 22%, primarily driven by Freshippo. International e-commerce revenue fell 8%. Alibaba attributed that decline chiefly to lower AliExpress revenue affected by EU customs reform, partly offset by growth at Trendyol.
Accounting presentation also matters. Customer management revenue fell 5%; Alibaba said it would have grown 1% without the revenue-reducing effect of a new business development program. Relevant subsidies previously classified as sales and marketing expenses are now recorded against revenue. Sales and marketing expenses fell to 41.341 billion yuan from 66.496 billion yuan, reflecting both that change and what Alibaba described as more efficient investment in instant commerce.
Alibaba’s company-reported cash measures. Capital expenditure is shown as positive spending and free cash outflow below zero. Non-GAAP free cash flow includes adjustments, so the bars are not an exact reconciliation. Source: Alibaba Group.
The next tests are cash generation and T-Head’s financing
Quarterly capital expenditure increased 60% from a year earlier to 50.477 billion yuan. Alibaba cited additional computing capacity, higher chip-component prices, anticipated demand for CPU capacity from AI agents and procurement timing. These are company explanations for spending; anticipated demand is not the same as realized sales.
Operating cash inflow rose 55% to 15.690 billion yuan. Free cash flow was still an outflow of 34.472 billion yuan, compared with 21.840 billion yuan a year earlier. Alibaba said lower instant-commerce losses and increased cloud customer prepayments partly offset higher infrastructure spending. It did improve sequentially: subtracting the September-quarter outflow from the six-month total of 79.142 billion yuan gives a June-quarter outflow of 44.670 billion yuan.
Free cash flow is Alibaba’s non-GAAP liquidity measure. It deducts relevant capital purchases from operating cash flow and adjusts for merchant buyer-protection deposits. It is distinct from both net income and operating cash flow.
The financing cushion remains large. Unrestricted cash and other liquid investments stood at 487.929 billion yuan at September 30, down from 520.824 billion yuan at March 31. Alibaba said proceeds from an HK$80 billion placement of new shares partly offset six-month free cash outflow and dividends. Its six-month financing cash inflows also included 14.802 billion yuan of net bank borrowing proceeds.
Computing supply remains a constraint. The report on chip access described new U.S. restrictions aimed at closing an indirect route to advanced chips through overseas computing capacity. It did not quantify the effect on Alibaba’s available capacity or revenue.
T-Head is Alibaba’s in-house chip operation. The same report relayed its chief executive’s statement that more than 560,000 Zhenwu chips had been shipped. That cumulative shipment figure does not establish how the next generation will perform. Alibaba’s earnings announcement schedules commercial release of the Zhenwu V900, a successor to the M890 intended for both model training and inference, for the first quarter of 2027.
The chip business could also gain a separate financing route. In its November 10 spin-off filing, Alibaba said wholly owned T-Head submitted a confidential Hong Kong listing application on September 11, after the exchange allowed the proposed spin-off to proceed. Alibaba expects T-Head to remain a subsidiary. It argues that a listing would let the business access equity and debt markets independently and make more efficient use of group financial resources.
The terms remain unsettled. The offering’s size, structure and resulting reduction in Alibaba’s ownership had not been finalized, and further regulatory steps and company decisions were still required. Existing Alibaba shareholders will not have an assured entitlement to T-Head shares under an exchange waiver. The US$85 billion potential valuation cited in the earnings report is a Nomura estimate, not a completed transaction or capital raised.
A completed offering would clarify how much outside capital T-Head can attract and how much ownership Alibaba retains. Commercial delivery of the V900 and subsequent cash-flow results will test whether the expanding AI business can support its infrastructure bill. Alibaba says it is targeting positive free cash flow, but its announcement gives no deadline.
The Kimi developer reportedly abandoned planned Nvidia B300 capacity for 2027 after new US restrictions. The deal exposes financing risks for overseas AI clouds, but neither a model delay nor a $150 million cash loss is established.
Kong is expanding from API management into hosting AI agents and applications. Volcano bundles backend services with a LavaFlow repair preview, but human review, plan restrictions and deployment controls narrow its autonomous-software pitch.
Anthropic's pre-debut price indications reached $105, about 26.5% above its offer price, after an $81.3 billion IPO. Strong demand for the Claude maker confronts enormous computing commitments and a governance structure that limits ordinary shareholders' influence.
Google has booked New Glenn for a dedicated Project Suncatcher launch in early 2029. Laboratory tests support parts of its orbital AI design, but deployment, launch reliability and commercial economics remain unproven.
Canva is rolling out free controls for Free and Pro accounts to disable supported generative AI tools and reduce AI-made assets in search. The settings extend individual choice while leaving AI enabled by default and preserving the company's investment in generation.
Alibaba targets 30%–45% cloud revenue growth next fiscal year and may reduce its long-term capacity plan. Domestic chips underpin its unchanged 2030 revenue ambition, but production, funding and customer allocation remain critical tests.
Singapore’s central bank plans a common protocol for financial institutions to assess customer AI agents, alongside a sandbox for treasury and payments. Its digital settlement network has completed a S$100 million bill issuance, but wider adoption, cost savings and agent accountability remain unresolved.
Jeff Bezos says Prometheus tools are in private trials, including two Blue Origin projects. Its claims of faster design now face the harder test of manufacturing gains, as rivals automate existing engineering workflows.
The IEA projects data-center electricity demand reaching 1,000 terawatt-hours annually by 2030, but connection queues, equipment shortages and local opposition could limit how much of the investment boom gets built.
European individuals secured about $1.1 billion of Anthropic shares, roughly 1.4% of the public offering’s value. The UK offer combines broad participation with rationing, while the AI lab’s governance and compute commitments define what investors are buying.
Federal prosecutors accuse Igor D’iu of acting for Russia to surveil employees of a Silicon Valley defense technology company identified in reports as Palantir. An FBI affidavit describes screen photographs, employee dossiers and $45,000 in reported payments, but leaves the handlers’ identities and the success of attempted electronic collection unresolved.
President Donald Trump rejected international restrictions on advanced AI at APEC. His administration has already outlined voluntary cybersecurity cooperation, alongside deregulation and support for American AI exports and infrastructure.
London-based LATO has raised $4.7 million to expand AI agents that interview customers and experts for investment research. Its early traction claims come as rivals offer similar research workflows and Salesforce absorbs Listen Labs; comparative cost and quality remain unproven.
Snap’s free Easy Lens CLI connects AI coding assistants to Snapchat’s AR creation and publishing workflow. It builds on earlier prompt-based tools as TikTok offers a competing approach.
Nvidia’s first earnings report since Jensen Huang’s departure showed revenue more than doubling, but shares fell 2.9%. Interim CEO Colette Kress faces a permanent leadership search, lower forecast margins and expanding infrastructure commitments, including conditional guarantees.
Needham’s $150 target stands about 81% above Anthropic’s $83 IPO price, versus Morningstar’s $92 fair-value estimate. Rapid enterprise growth and fresh financing support the business, but pricing power, compute commitments and shareholder control complicate the investment case.
A cyberattack disrupted access to European Commission payslips, leave management and medical reimbursements. The HR platform is back online, while the health insurance system remains unavailable and the extent of any staff data theft is unknown.
Waymo suspended Bay Area service during the November 18 storm after reports of robotaxis obstructing flooded streets, raising questions about how it withdraws its fleet when road conditions deteriorate.
Lenovo reported record quarterly revenue of $30.2 billion as its infrastructure business swung into profit. A $1.91 billion noncash warrant charge drove a reported loss, while expanding inventories and overdue receivables raised the stakes for cash collection.