Alibaba's AI business delivered a "growth with profitability" performance last fiscal quarter: Alibaba Cloud's external commercial revenue surged 45% year-over-year, hitting a 22-quarter high in growth momentum, while the profit margin for AI cloud and computing services simultaneously rose to 12%. The dual-track advancement of large-scale commercialization for self-developed chips and accelerated data center delivery underpins this result, signaling that AI investment is transitioning from an expansion phase into a profit realization stage.
According to the financial report, for the company's first fiscal quarter ending June 2026, Alibaba Cloud's AI-related product revenue reached RMB 12.376 billion, marking the 12th consecutive quarter of triple-digit year-over-year growth. Profitability metrics show that EBITDA for the AI cloud and computing services segment surged 133% year-over-year, with adjusted EBITDA margin climbing to 12%.
During the earnings call, Alibaba management further disclosed that the annualized revenue run rate (ARR) for AI-related products surpassed RMB 49.5 billion, accounting for 35% of Alibaba Cloud's external commercial revenue.
The key driver behind the margin improvement is that AI-related products boast significantly higher gross margins than the average cloud product. This indicates that high growth in AI revenue is now converting into profit gains, rather than just "growth without profit." Alibaba Group CEO Eddie Wu stated that AI commercialization has advanced from the "inflection point breakthrough" seen in the fourth fiscal quarter to a stage of "accelerated growth and rising margins" in the first fiscal quarter.
The foundation of profit realization lies in the leap in efficiency from self-developed chips and infrastructure. T-Head's self-developed chip, Zhenwu M890, has now covered over 650 external clients, with supernode instances launched on Alibaba Cloud capable of running inference for large language models exceeding 2 trillion parameters. Large-scale AI data center delivery cycles have been compressed to 100 days, and management projects AI computing assets can achieve payback in approximately three years.
Revenue and profit rise in tandem
Alibaba Cloud's external commercial revenue grew 45% year-over-year this quarter, the fastest pace in 22 quarters. AI-related product revenue reached RMB 12.376 billion, sustaining triple-digit year-over-year growth for the 12th consecutive quarter, serving as the core engine accelerating cloud business growth.
Even more noteworthy is the simultaneous improvement in earnings quality. EBITDA for the AI cloud and computing segment soared 133% year-over-year, with adjusted EBITDA margin rising to 12%. The ARR for AI-related products surpassed RMB 49.5 billion, now representing 35% of Alibaba Cloud's external commercial revenue, with gross margins significantly above the average for cloud products.
Wu noted that an increasing number of customers are adopting the full-stack AI portfolio covering AI agents, models, cloud infrastructure, and self-developed chips. Combined with business scale expansion and improved pricing power from tight computing supply, AI is evolving from a revenue growth engine for the cloud business into a key driver of profit improvement. Whether the profit inflection point is firmly established requires validation from subsequent quarterly data, but the substantial single-quarter earnings improvement has already signaled its emergence.
Self-developed chips achieve commercial scale
The underlying support for profit realization is the commercialization of T-Head's self-developed chips. T-Head has built a full-stack proprietary system covering GPUs, CPUs, and network chips. The latest-generation AI processor, Zhenwu M890, has been commercialized through Alibaba Cloud across more than 650 external clients in over 20 industries, including autonomous driving, internet, and finance.
The commercialization process continues to accelerate. Supernode instances based on the Zhenwu M890 have recently launched on Alibaba Cloud and are now being sold at scale, with Wu stating that volume will continue to ramp up in the second half of the year.
During the earnings call, Wu revealed that over 500,000 units of the previous-generation T-Head chip have been produced and shipped to date. The latest-generation chip was deployed on Alibaba Cloud's AI platform in August, offered in Supernode form. He noted, "I believe we are among the few companies capable of deploying such self-developed domestic chips at this scale."
According to Alibaba Cloud, the "Zhen 5 M890" supernode based on the new-generation T-Head chip can run inference for large language models exceeding 2 trillion parameters. Kimi K3 and Qwen 3.8 Max are already leveraging this instance for service delivery.
For Alibaba Cloud, the full-stack coverage of GPUs, CPUs, and network chips with self-developed technology, coupled with commercial-scale deployment and external sales, means T-Head's computing power is shifting from internal support to volume sales, becoming a new growth variable for the cloud business.
Data center delivery accelerates
Efficiency gains on the computing supply side are equally significant. Alibaba Cloud has compressed the delivery cycle for large-scale AI data centers to 100 days and projects that production capacity efficiency for self-developed modular data centers will more than double in 2026, positioning the company to meet stronger AI computing demand.
Behind this efficiency lies front-loaded capital expenditure. As of the end of the first fiscal quarter, Alibaba had invested approximately RMB 190 billion of the RMB 380 billion three-year investment plan announced in February 2025.
Alibaba CFO Toby Xu stated that AI commercialization is built on infrastructure such as computing centers, emphasizing that "capex must come first to secure subsequent business growth."
According to his estimates, AI computing assets can recover costs in about three years, and with gross margins for AI-related products continuing to improve, the payback period could shorten further.
Going forward, market attention will focus on the volume ramp-up of Zhenwu M890 supernodes and whether the AI cloud and computing segment's profit margins can sustain their upward trajectory in upcoming quarters.
Comments