Alibaba shares rose after the company unveiled a new AI chip and a plan to massively expand data center capacity, with investors sending the stock up about 3% in Hong Kong trading and intraday gains reaching as high as 5% following the announcements at its annual Apsara Conference in Hangzhou.
Alibaba’s Zhenwu V900 and chip roadmap
Alibaba presented the Zhenwu V900 as its next-generation AI accelerator, saying the processor delivers three times the performance of its predecessor, the Zhenwu M890, which was released in May. The Zhenwu V900 is expected to enter mass production and become commercially available in the first quarter of 2027. Alibaba said its existing Zhenwu processors are already used by more than 650 customers across automotive, finance, energy and manufacturing sectors.
Data center expansion and the infrastructure race
CEO Eddie Wu announced that Alibaba Cloud plans to expand global data center capacity to more than 20 gigawatts by 2032, signaling the scale of computing power and electricity the company expects future AI systems to require. Alibaba said customer demand for AI remains exceptionally strong, but warned that shortages across the AI data center supply chain could constrain the speed of expansion. The planned buildout places Alibaba deeper into a global infrastructure spending race that has drawn billions of dollars from major technology firms.
The company’s announcement came alongside other large-scale infrastructure plans: Nvidia earlier outlined partnerships in Australia to support up to 2 gigawatts of AI infrastructure by 2027, and Meta announced in July a 1-gigawatt data center project in Alberta, Canada, expected to cost approximately $9 billion.
Qwen models and ambitions for trillion-parameter training
Alibaba said it is expanding its Qwen family of AI models: next-generation Qwen 4 is currently being trained, with Qwen 4.5 and Qwen 5 on the roadmap. Wu said the company ultimately plans to train a model containing between 5 trillion and 10 trillion parameters. He likened the current phase of AI development to the early years of electrification, stating: “AI coding is simply the light bulb of the machine intelligence era,” and predicting a much longer growth runway for machine intelligence.
The announcements arrive as Chinese technology companies work to strengthen domestic semiconductor capabilities amid U.S. restrictions on exports of some of Nvidia’s most advanced AI processors to China. Huawei last week unveiled new AI infrastructure it said could scale to as many as one million processors.
