RSS feed for Chips & Compute →
Alibaba has decided that its AI buildout needs more outside capital, not just the cash generated by its core businesses. On August 23, Nikkei Asia reported Alibaba’s plan to sell HK$80 billion, or US$10.2 billion, of equity beyond the U.S. market. Alibaba said it would invest all proceeds in its full-stack AI capabilities.
Nvidia has publicly rejected a report that it was preparing a new artificial-intelligence chip for Chinese customers by the end of the year. IT Home reported the company’s denial on August 22, saying Nvidia had no LPU sales in China and no China-specific LPU product in its roadmap.
Northern China’s next AI infrastructure test is not simply whether it can add more servers. It is whether low-carbon computing capacity can be turned into an exportable service.
China’s AI-chip competition is often described through processors, memory, and export restrictions. Packaging is less visible, yet it determines whether increasingly dense components can move heat away quickly enough to operate reliably. Pandaily reports that Huatian Technology had developed full-process thermal-design and simulation capability for fan-out packaging.
China’s AI industry is confronting a less visible constraint than model training: the machinery needed to serve models at scale after they have been built. South China Morning Post reports that companies are redesigning software to manage a sharp rise in inference demand while their access to high-end Nvidia processors remains limited.
Shanghai Lightelligence Co. Ltd. filed for an initial public offering on the Hong Kong Stock Exchange in March 2026 as it pivots from purely photonic computing to optical interconnect hardware for AI data center infrastructure, according to Pandaily’s report on the filing and strategy shift.
Brazil announced approximately 2.3 billion reais, or about 444.2 million dollars, in investments on August 20, 2026 to bolster its artificial intelligence ecosystem, according to a Reuters report.
Xiaomi has confirmed that a next-generation self-developed chip is nearing release. The update came from founder and CEO Lei Jun, who described the project as approaching that stage. Beyond this status, Xiaomi did not provide additional information about the chip.
OceanBase is reported to be pursuing an A round that targets about 2 to 3 billion yuan while positioning itself as an AI data platform. That yuan range is cited by Pandaily and appears in coverage that situates the company within the data infrastructure segment that supports AI workflows.
A Nanjing University team led by professors Miao Feng and Liang Shijun, working with collaborators from the National University of Singapore, has developed an intelligent visual sensor called the Guangyu Chip, according to a report in Science and Technology Daily.
Tencent Cloud said on August 18 that it will establish its first cloud region in Malaysia, with the planned site in Johor and designed for up to three availability zones.
A report by SCMP says Bank of America Global Research surveyed 98 Asia participants between August 7 and 13 who together oversaw US$272 billion in assets.
Chinese artificial intelligence firms have reportedly accessed advanced Nvidia compute by connecting remotely to data centers located in Thailand, Malaysia, and Japan.
Hesai Group, used its latest quarterly results to signal a broader ambition. The company is now positioning itself as a supplier of hardware, data, and software for what the industry increasingly calls physical AI, the category that covers autonomous machines operating in the real world, from self-driving cars to humanoid robots.
Chinese public cloud provider UCloud has begun offering servers powered by Hygon’s Tianxi series AI accelerators through its public cloud platform, giving external customers access to a domestically developed AI chip line through the public cloud.
Chinese custom silicon and semiconductor IP provider VeriSilicon reported first-half 2026 revenue of 1.864 billion yuan, up 91.37% from the same period a year earlier. The result underscores how quickly demand for artificial intelligence compute is reshaping the order books of design service companies that sit between fabless chip developers and foundries.
Ingenic Semiconductor has begun a Hong Kong global offering that places a Chinese fabless chip designer into the city’s active technology-fundraising market. The company is not an AI-only semiconductor business. Its portfolio spans memory, computing, and analog products.
Huawei has opened a new piece of software infrastructure for its Ascend AI hardware: AscendNPU IR, an intermediate representation designed for compiler work. The project is not a new chip, model, or consumer service.
A new research preprint from Peking University, StepFun, and Beijing University of Posts and Telecommunications focuses on an unglamorous but consequential part of large-model infrastructure: what happens to tensors after a model begins working.
Hong Kong-based Antimatter is trying to turn the rise of Chinese open-weight AI models into a cloud-infrastructure business. Its proposition is not simply to offer another model catalog. The company describes itself as a “neo-cloud” provider built for firms that want computing infrastructure designed around open-weight systems rather than the dominant US frontier-model platforms.
