IoT News for November 28, 2025

IoT News for November 28, 2025IoT News for November 28, 2025Quark AI Glasses Released with Alibaba’s QianwenAlibaba’s Quark AI glasses have been launched, integrating Alibaba’s Qianwen. In terms of hardware, they feature a dual-chip design with Qualcomm’s AR1 flagship chip and the Hengxuan BES2800 chip, paired with a dual-system of Android and RTOS, supporting dual display and optical distance adjustment technology; equipped with a dual battery system and a design that includes five microphones and one bone conduction array. The Quark AI glasses are available in two series, S1 and G1, with a total of six products. The S1 series offers three products in different frame styles and colors, with a minimum price of 3799 yuan. The G1 series, which focuses more on fashion and portability, offers two frame style options and includes sunglasses, starting at a minimum price of 1899 yuan. (Ke Chuang Ban Daily)IoT News for November 28, 2025

China’s AI Open Source Model Downloads Exceed Those of the US for the First TimeA research report jointly released by MIT and Hugging Face shows that in the past year, Chinese models accounted for 17% of the total global downloads of AI open source models, surpassing the 15.8% from companies like Google and Meta in the US. Researchers indicate that this is the first time China has led in this field.

Giant Network AI Lab Releases New Multimodal Generation Results, Related Technologies to be Open SourcedGiant Network AI Lab, in collaboration with Tsinghua University’s SATLab and Northwestern Polytechnical University, has launched three multimodal generation technology results in the audio and video field, with related research results to be gradually open-sourced on platforms like GitHub and HuggingFace.

IoT News for November 28, 2025

Alibaba’s Qianwen Wins Best Paper Award at NeurIPS 2025The prestigious AI conference NeurIPS 2025 has officially announced its awards. This year, the conference received 5524 papers, with an overall acceptance rate of 24.52%. Only four papers globally received the Best Paper Award. The Alibaba Tongyi Qianwen team won the award for their latest research on attention mechanisms. This paper reveals for the first time the impact of attention gating on the performance and training of large models, and the research has been applied to the Qwen3-Next model.

Wang Xingxing Discusses Why Yushu’s Robots Don’t Work: Current Technology is Not Quite ReadyAt the 2025 High-Level Talent Forum – Talent Regional Coordination Development and Yangtze River Delta Talent Integration Forum, Yushu Technology founder Wang Xingxing stated, “Some may wonder why Yushu’s robots don’t work. In fact, we have spent a lot of time making robots work. Our ultimate goal is for robots to truly work and enter thousands of households to help solve everyday problems, but current AI and robotics technology is not quite ready. Our goal is to enable robots to perform various complex actions, and once they can do that, we can combine those actions, which in a sense means they can work.” Wang Xingxing said that in the future, large-scale robots that can truly work will definitely be able to perform various actions.

JD.com Showcases Over 30 Robot Brands at the World Intelligent Manufacturing ConferenceOn November 27, the 2025 World Intelligent Manufacturing Conference kicked off at the Nanjing International Expo Center. JD.com, in collaboration with Yushu Technology and over 30 top global robot brands, showcased cutting-edge achievements in four major areas: retail supply chain, AI productivity, intelligent logistics, and industrial digital equipment, covering various scenarios such as companionship, entertainment, elderly care, and family education. JD.com’s self-operated robot rental service was also officially launched at this exhibition, allowing consumers to directly rent humanoid robots, quadruped robotic dogs, and other products through the platform.

Chinese Academy of Sciences Establishes Industrial AI Research InstituteThe Industrial AI Research Institute of the Chinese Academy of Sciences has been established in Nanjing, Jiangsu. This institute focuses on major national strategic and industrial innovation development needs, undertaking the full-chain innovation tasks of industrial AI from top-level design to system implementation, building a collaborative innovation system of “technology leadership – platform integration – empowerment loop” to promote high-quality development of manufacturing empowered by AI, and striving to create national strategic scientific and technological strength in the field of intelligent manufacturing. (CCTV News)

Li Xiang Claims the Most Valuable Robot is the CarAt the Li Auto Q3 2025 financial report meeting, Li Auto CEO Li Xiang stated, “In the next decade, the most valuable product of embodied intelligence will definitely be cars with automatic and proactive capabilities.” He believes that the ultimate form of a car should be “an embodied intelligent product in the physical world,” that is, a car robot. To achieve this goal, it is necessary to endow cars with complete capabilities of “perception-decision-action”: using sensors as “eyes and ears” to capture environmental information, using AI models as “brain neurons” to process data, and using high-performance hardware as “body organs” to support operations. (Sina Finance)

South Korean Government Forms AI Task Force with Samsung, Hyundai, etc., to Plan Deployment of 260,000 NVIDIA GPUsAfter NVIDIA announced last month that it would supply 260,000 GPUs to South Korea, the South Korean government, along with Samsung Electronics, Hyundai Motor Group, and other companies, established a joint AI task force this Thursday to plan the deployment of these 260,000 GPUs and promote the establishment of AI infrastructure across South Korea. According to the plan, the South Korean government will deploy 50,000 NVIDIA GPUs, while Samsung Electronics, Hyundai Motor Company, and SK Telecom will each deploy 50,000 units, and Naver will deploy 60,000 units.

IoT News for November 28, 2025

AI Computing Power Service Provider Wuwen Xinqiong Completes 500 Million Yuan Financing with Zhuhai State Capital InvolvementAI computing power service provider Wuwen Xinqiong announced that it has recently completed nearly 500 million yuan in A+ round financing, led by Zhuhai Technology Group and Foton Capital (Yuan Chuang Future Fund), with follow-on investments from Huiyuan Capital, Shangqi Capital, and Honghui Fund, as well as additional investments from existing shareholders such as Lenovo Capital, Junlian Capital, Xuhui Kechuang Investment, and Shenwan Hongyuan. Zhuhai Technology Group was established in May this year, integrating two local state-owned enterprises, Zhuhai Huafa and Gree, to promote the development of the technology industry in Zhuhai. Zhuhai Huafa Group Co., Ltd. and Zhuhai Gree Group Co., Ltd. hold 60% and 40% of the shares, respectively. (Caixin)

Investor Lishi Capital Reportedly Raises $289 Million to Invest in More AI StartupsInvestor Lishi Capital has reportedly raised approximately $289 million, seeking to invest in more companies in the Chinese artificial intelligence (AI) sector. According to insiders, this Chinese venture capital fund has surpassed its initial fundraising target of $265 million and plans to use the funds to support more tech startups. Lishi Capital, operated by former Sequoia China partner Cao Xi and former Boyu Capital partner Tim Wang, is one of the few institutions this year to successfully attract US dollar capital.

IoT News for November 28, 2025

Always Missing the Three-Minute News Push? Don’t Worry! Join Feng’s News Sharing Group, where you will receive carefully curated three-minute news updates from Feng’s News House every day, keeping you informed anytime, anywhere! Long press the QR code in the image above to join the group and share an information feast with thousands of news enthusiasts!

Recently, WeChat has modified the rules for subscription account pushes. To ensure you receive Feng’s News House updates promptly, please click the button below toLike+View!

Leave a Comment