November 8's AI headlines run in two directions. Model developers and global capital kept arguing over compute paths and where moats actually sit, while on a parallel track the robots and the industry deployments quietly kept moving — quietly on paper, loudly on the ground. Ten items below.
MiniMax founder on China-US AI paths. A rare interview, and a blunt one. MiniMax's founder said the US is stacking compute through "brute-force investment" while China is catching up on engineering efficiency. The gap between those two paths, in his telling, deserves close watch. (Phoenix Tech)
Zhang Yaqin and Yao Shunyu debate model moats. Zhang Yaqin pressed on where the moat of large models really lies. Yao Shunyu's answer put data and scenario accumulation ahead of raw model capability. (Sohu)
Topsstar unveils bipedal robot. The company released a new bipedal robot, extending a factory-floor lineup aimed at material handling and machine tending. More legs, more work. (Sing Tao Global)
26th China VR Conference opens in Guangzhou. Industry and academia gathered under one roof, and VR-AI fusion applications sat high on the agenda. (Guangzhou Daily)
Frost & Sullivan publishes compute-chip report. The 2026 report on global novel compute chips is out, and it names reconfigurable dataflow architecture as a potential path forward. (Frost & Sullivan)
Lingchuan Tech targets video-generation compute. The company is building a domestic compute base for multimodal models and entering the generative-video compute market. (CNR)
Hitachi uses its own factory as physical-AI first customer. Hitachi China's chief representative said at CIIE that physical AI starts with humanoid material handling inside Hitachi's own plants — then comes the customer rollout. Eat your own cooking first. (21st Century Business Herald)
Physical AI still climbing the early slope. An analysis in The Paper laid out three bottlenecks behind the physical-AI boom. The industry, it concluded, is still in early ramp-up. (The Paper)
Greater Bay Area hosts AI jobs fair. A GBA youth talent fair ran a dedicated AI session. Demand for AI roles ran strong. (Zhongshan Net)
Sichuan tours flagship national AI application scenes. The through-line of the tour, AI is moving from single-point assistance to full-process collaboration, acting as a "research partner" in major science scenarios. (Sichuan News)
From compute-path debates to bipedal robots on factory floors, the industry's question is shifting from "how strong is the model" to "how fast does it land."