Tracking Hong Kong Stock Concepts | Qianwen App's public beta week downloads exceeded 10 million. AI applications usher in the "iPhone moment" (with concept stocks)

AI assistant Qianwen App under Alibaba has been publicly tested for a week. As of November 24th, the download volume has exceeded 10 million times, surpassing ChatGPT, Sora, and DeepSeek, becoming the fastest-growing AI application so far. It is reported that since its launch on November 17th, Qianwen App has been highly sought after and entered the top three of the Apple App Store's free overall rankings within just two days. On November 17th, Alibaba announced its full force entry into the AI to C market, and the public beta version of Qianwen App was launched. Alibaba plans to integrate various life scenarios such as maps, take-out, ticket booking, office work, learning, shopping, and health into Qianwen App, enabling Qianwen to have more powerful task capabilities. Alibaba stated that the Qianwen App released this time is a preliminary version, using the most advanced models to create a "personal AI assistant that can chat and get things done." In addition to being smart enough in chatting, "being able to get things done" will be an important focus of Qianwen App. The strategic goal of Qianwen App is to create the future entrance to AI life. Industry insiders believe that the explosive public testing of Qianwen App is just the beginning. With the continuous evolution of open-source models, rapid iteration of Agent capabilities, and accelerated ecological integration, Qianwen App is expected to become a national-level application in the AI era, undertaking the mission of enabling China's AI to win top global competition. In addition, Ant Group recently launched the multimodal universal AI assistant "Lingguang," which has exceeded 2 million downloads within 6 days. Lingguang reached 1 million downloads within the first four days, surpassing the growth rate of many globally significant products. According to data from mobile app analysis agency Appfigures, ChatGPT had 606,000 downloads in its first week, while Sora2 achieved a million downloads in 5 days. Lingguang compressed this process to 4 days, setting a new industry record. As of now, domestic AI applications including Dou Bao, Qianwen, and Lingguang are ranked at the top of the App Store rankings, ranking first, fourth, and sixth in the China region's free rankings, respectively. Analysis believes that to some extent, the outbreak of Qianwen and Lingguang may mark the long-awaited "iPhone moment" of AI applications. With the promotion of industry giants, Chinese artificial intelligence is undergoing a crucial transition from "savoring chat with Siasun Robot&Automation" to "solving real-world problems with Agents." Huatai believes that new AI products and models will bring about a more vigorous demand for AI infrastructure. With Alibaba increasing its investment in C-end AI applications, the scale of data center energy consumption is expected to continue growing, and its AI infrastructure partners are expected to benefit in the long term. Orient stated that Alibaba is intensifying its AI application layout through the Qianwen App, and it is optimistic about strengthening Alibaba's full-stack AI advantages after the perfect layout of AI applications. Qianwen App can empower other life services within the group bidirectionally, achieve traffic sharing and income acceleration, and show potential for multi-directional growth within Alibaba's business. Related concept stocks: Alibaba-SW(09988): On November 17th, Alibaba officially announced the "Qianwen" project, entering the AI to C market with full force. Analyst Zheng Hongda of Western believes that in the future, Alibaba may establish a C-end profit loop through subscription, traffic entry, and other business models to reach and convert end users more widely with model+scenario-driven AI services. BIDU-SW(09888): On November 13th, at the 2025 Baidu World Conference, Baidu officially released the native multimodal large model Wenxin 5.0. Wenxin 5.0 has comprehensively upgraded basic capabilities, showing outstanding performance in multimodal understanding, instruction following, creative writing, factual, intelligent planning, and tool applications, and has strong understanding, logic, memory, and persuasiveness. In more than 40 authoritative benchmark comprehensive evaluations, its language and multimodal understanding abilities are on par with models like Gemini-2.5-Pro and GPT-5-High, and its image and video generation capabilities are equivalent to those of vertical domain specialized models, reaching a leading global level, verifying the capabilities and potential of native multimodal large models. SENSETIME-W(00020): On November 11th, SenseTime released and open-sourced the SenseNova-SI series models. According to introduction, SenseNova-SI surpasses the performance of international closed-source models such as GPT-5 and Gemini 2.5 Pro in spatial understanding and reasoning tasks in multiple evaluations.