Alibaba Exceeds AI Spending Plans Amid Quarterly Margin Pressure

The Earnings Announcement: A Mixed Bag for Alibaba

Alibaba Group Holding Limited, the e-commerce behemoth and one of China's leading technology companies, released its fiscal fourth quarter and full-year 2026 results on May 13, 2026. The numbers painted a picture of aggressive expansion in high-growth areas clashing with short-term profitability challenges. Total revenue for the quarter ended March 31, 2026, climbed 3 percent year-over-year to 243.38 billion yuan (approximately 35.28 billion U.S. dollars), falling slightly short of analyst expectations of 247.22 billion yuan. On a like-for-like basis, excluding the impacts from divestitures like Sun Art and Intime, revenue growth accelerated to 11 percent, signaling underlying strength in core operations.

However, profitability took a significant hit. Adjusted earnings before interest, taxes, and amortization (EBITA), a key non-GAAP profitability metric used by tech firms to assess operational performance excluding one-time items, plummeted 84 percent to 5.1 billion yuan. Adjusted net income dwindled to just 86 million yuan, nearly wiping out profits and marking a stark contrast to the prior year's robust figures. This downturn was largely attributed to escalated capital expenditures (capex) on artificial intelligence (AI) infrastructure and investments in quick commerce, Alibaba's ultra-fast delivery service aiming to fulfill orders in under 60 minutes.

For the full fiscal year 2026, revenue grew 3 percent to 1,023.67 billion yuan, again with like-for-like growth at 11 percent. Full-year adjusted EBITA fell 56 percent to 76.4 billion yuan, underscoring the toll of strategic spending. Free cash flow swung to a negative 46.6 billion yuan outflow from a positive 73.9 billion yuan inflow the previous year, primarily due to capex totaling 126 billion yuan focused on cloud and AI buildout.

AI and Cloud: The Growth Engines Accelerating

Amid the profit pressures, Alibaba's Cloud Intelligence Group shone brightly. Quarterly revenue surged 38 percent to 41.63 billion yuan, with external customer revenue up 40 percent. Notably, AI-related products now constitute 30 percent of external cloud revenue, achieving triple-digit growth for the 11th consecutive quarter. This momentum reflects surging demand in China for AI computing power, driven by enterprises adopting large language models (LLMs) and generative AI tools.

Alibaba's homegrown Qwen series of AI models has emerged as a cornerstone of this success. Qwen, short for 'Qianwen' or 'thousand questions,' represents Alibaba's suite of open-source and proprietary LLMs that have topped global benchmarks in reasoning, coding, and multimodal capabilities like video generation. Recent advancements include Qwen 2.5-Max, trained on over 20 trillion tokens, positioning it among the world's most advanced models. The company has integrated Qwen deeply into its ecosystem, launching enterprise AI agents for office productivity and coding, as well as consumer-facing apps like the Qwen Shopping Assistant on Taobao.

Full-year cloud revenue rose 34 percent to 158.13 billion yuan, with AI commercialization scaling rapidly. CEO Eddie Wu emphasized a 'full-stack AI' approach, encompassing infrastructure, models, and applications, to capture China's booming AI market projected to exceed 200 billion yuan by 2026 according to government estimates.

Capex Surge: Exceeding AI Investment Plans

Alibaba explicitly stated it will surpass its previously announced 380 billion yuan three-year investment plan in AI, originally set in early 2025. This commitment, equivalent to about 56 billion U.S. dollars, underscores the company's determination to lead China's AI infrastructure race. Quarterly net cash used in investing activities reached 26.9 billion yuan, predominantly for cloud servers, GPUs, and data centers optimized for AI workloads.

The process of AI infrastructure buildout involves several steps: first, procuring high-performance chips like Nvidia's H100 or domestic alternatives amid U.S. export restrictions; second, constructing hyperscale data centers with liquid cooling for efficiency; third, developing software stacks for model training and inference; and finally, offering these as cloud services via platforms like Model Studio, whose customer base grew eightfold year-over-year. This capex intensity mirrors global trends, with U.S. hyperscalers like Amazon and Microsoft planning over 700 billion dollars in 2026 AI-related spending, but Alibaba's scale in China positions it uniquely.

Wu noted that demand for AI infrastructure 'exceeds expectations,' prompting the overspend. The strategy prioritizes market share over immediate returns, with AI-related revenue expected to exceed 50 percent of cloud growth drivers within the next year.

Margin Pressures: Quick Commerce and Competitive Investments

Beyond AI, margins faced headwinds from quick commerce expansion. Revenue from this segment jumped 57 percent to nearly 20 billion yuan, but subsidies and logistics buildout eroded profitability. Adjusted EBITA for the Taobao and Tmall Group (Alibaba's core China e-commerce) declined 40 percent to 24 billion yuan, despite customer management revenue up 1 percent (8 percent like-for-like).

China's e-commerce market, valued at over 15 trillion yuan annually, is fiercely competitive. Rivals like PDD Holdings (Pinduoduo) and ByteDance's Douyin aggressively discount, forcing Alibaba to match with promotions and faster delivery. Wu affirmed quick commerce unit economics will turn positive by fiscal 2027 end, with improving average order values and customer retention.

Overall adjusted EBITA margin compressed to 2 percent in the quarter from 14 percent last year, and 7 percent full-year from 17 percent. Toby Xu, Alibaba's CFO, highlighted offsets from better customer management but acknowledged investment drag.

Photo by Andrey Zvyagintsev on Unsplash

Qwen's Rise: Powering Alibaba's AI Ambitions

At the heart of Alibaba's AI push is the Qwen family, which has amassed over 1 billion downloads globally by early 2026. Qwen 3.5 introduces agentic AI—autonomous systems that plan, execute, and learn from tasks—targeting enterprise use cases like supply chain optimization and personalized marketing.

In China, where data sovereignty and cost matter, Qwen's open-source variants allow customization, fostering a developer ecosystem. Alibaba Cloud's Model-as-a-Service (MaaS) platform democratizes access, with integrations across Taobao (for search and recommendations), Alipay (fraud detection), and Fliggy (travel planning). This vertical integration differentiates Alibaba from pure cloud providers, embedding AI into daily consumer touchpoints serving over 1 billion users.

Recent launches include Wukong, an AI agent for merchants to automate listings and pricing, boosting seller efficiency amid economic headwinds.

China's Tech Landscape: Alibaba vs. Peers

Alibaba isn't alone in its AI fervor. Tencent, reporting similar pressures, doubled AI capex in 2026, with cloud growth lagging at mid-teens. Both firms lost market value post-earnings—Alibaba down 6.9 percent year-to-date in Hong Kong, Tencent 23 percent—as investors demand AI monetization clarity.

Baidu leads in autonomous driving AI, while ByteDance invests in content generation. Government support via the 'Made in China 2025' initiative and national computing centers bolsters domestic players against U.S. sanctions. Alibaba's edge lies in its e-commerce moat, where AI enhances gross merchandise value (GMV) exceeding 8 trillion yuan annually.

• Tencent: Gaming and WeChat mini-apps strong, but AI revenue nascent.
• PDD: Low-price strategy squeezes all margins.
• ByteDance: TikTok global, but domestic focus on e-commerce.

Market Reaction: Shares Dip, Analysts Cautious Yet Optimistic

U.S.-listed Alibaba shares (BABA) fell 2.3 percent in premarket trading post-earnings, reflecting disappointment over the profit miss. Hong Kong shares mirrored the slide. Analysts maintain a consensus 'Buy' rating, with targets around 170-180 U.S. dollars, citing AI as a multi-year catalyst despite near-term pain. Yahoo Finance analyst consensus highlights cloud's potential to offset e-commerce slowdowns.

Risks include U.S.-China trade tensions impacting chip access and regulatory scrutiny on monopolies.

Strategic Priorities: Growth First, Margins Later

Wu's mantra: 'Growth faster than market average to gain share; margin is secondary.' Cloud gross margins expected to rise in coming quarters as utilization improves. Long-term, Alibaba eyes 100 billion U.S. dollars in annual AI-plus-cloud external revenue within five years. Separating AI into a dedicated unit under Wu aims for profitability.

This echoes global Big Tech: Amazon's AWS turned profitable after years of capex; Alibaba bets on similar trajectory in China's 500 billion yuan cloud market by 2026.

Photo by Rubén Bagüés on Unsplash

Broader Implications for China's AI Drive

Alibaba's moves align with Beijing's push for AI self-reliance. Initiatives like the National Data Bureau and subsidies for domestic GPUs accelerate adoption. Challenges persist: talent shortages, energy constraints for data centers, and ethical AI governance. Success could propel China to global AI leadership, with economic multipliers from productivity gains estimated at 1-2 percent GDP boost annually.

For stakeholders—merchants, consumers, investors—Alibaba's pivot promises smarter services but demands patience.

Future Outlook: Navigating Risks Toward Dominance

Looking ahead, FY2027 guidance emphasizes sustained AI/cloud investment amid e-commerce stabilization. Risks: macroeconomic slowdown in China (property crisis, youth unemployment), intensified competition, geopolitical frictions. Opportunities: agentic AI revolutionizing industries, export of Qwen models via Belt and Road.

Alibaba's cash reserves of over 400 billion yuan provide a buffer, positioning it to outspend rivals and capture the AI dividend. As Wu stated, 'We are seeing a clear path for AI to drive robust growth.' Investors and observers will watch if execution matches ambition.