鱼羊 发自 拉斯维加斯 量子位 | 公众号 QbitAI 豆包手机刚在2025年底掀起关于AI硬件新形态的讨论，2026新年之初，新的风向就已显现苗头—— 就在被认为将揭开新一年科技趋势的CES 2026上，一个共识正在各大厂商的行动中浮出水面： 所有的设备都会变成AI设备。 不仅是传统的手机、PC，呈现出更加智能的形态： 智能硬件新物种，呈现的方式也越来越五花八门： 甚至更深层的变化，也开始暗潮汹涌…… 比如英伟达老黄、AMD苏妈都在cue的关键词「物理AI」。 以及其背后更广泛意义上的， AI「超级入口」新竞争 。 在过去的移动互联网时代，超级App们，就是「超级入口」的代表：具有强大的流量聚集能力，连接用户与多种场景，成为用户进入数字世界 的主要通道。 历史的经验已经证明，移动互联网时代的赢家们，关键胜负手，正是通过软硬件结合和服务生态，牢牢掌握住了操作系统和应用分发的入口， 在「超级入口」竞争中跻身上游。 来到大模型时代，同样的底层逻辑正在显现。 当基础模型竞争逐渐收拢，AI领域的排位赛开始从底层技术导向，向应用层，乃至更全面更复杂的生态层延伸。 前有OpenAI、Qwen推进C端应用，后有英伟达老黄 ...

文 | 奇点研究社，作者｜孟雯 在互联网的浩瀚进程中，有些战争是为了疆土，有些战争是为了生存。 2025年12月10日，阿里巴巴正式官宣，改版上线仅23天的千问App月活跃用户（MAU）突破3000万， 刷新了全球AI应用的增长纪录。 看似辉煌的数据背后，是一场由阿里董事会主席兼CEO吴泳铭亲自带队、集结了100多名工程师秘密研 发数月的"闪击战"。 这种罕见的紧迫感折射出一个事实：在经历了移动互联网时代的流量起伏后，阿里正以"全村的希望"之 姿，将千问推向决定未来十年命运的战场。 AI技术已在C端市场全面渗透。 据CNNIC《生成式人工智能应用发展报告（2025）》披露，截至2025年6月，中国生成式AI用户规模已 达5.15亿，其中超过三分之一网民使用过AI应用，且"多数用户更偏好国产模型"。 AI助手正逐渐超越传统的App，成为新一代"超级入口"。 掌握了入口，就等于掌握了流量的"税收权"。阿里巴巴CEO吴泳铭在财报会上明确表示，阿里将在AI to B和AI to C两条战线同时发力，尤其是C端，要打造"AI原生超级应用"。 一场不容有失的"入口保卫战" 阿里巴巴对「入口」的执念，是刻在骨子里、由数次阵 ...

Core Insights - AI has become the central theme in China's technology industry throughout 2025, viewed as a key driver for the next productivity revolution and business model iteration [1] - Major internet companies like Tencent, Alibaba, and ByteDance are intensifying their focus on AI, with significant investments and strategic moves to secure their positions in the market [1][3] Investment Trends - Alibaba announced a substantial investment of 380 billion yuan over three years for cloud and AI hardware infrastructure, which is more than its total investment in the past decade [3] - ByteDance plans to increase its capital expenditure to 150 billion yuan in 2025, marking an 88% year-on-year growth, with expectations to further raise it to 160 billion yuan in 2026 [3] - Tencent's capital expenditure surged by 386% to 36.58 billion yuan in Q4 2024, with projections for 2025 to reach around 100 billion yuan [4] Financial Performance - Alibaba's cloud revenue grew by 34% year-on-year in Q3 2025, with AI-related product revenue maintaining triple-digit growth for nine consecutive quarters [4] - Tencent's marketing services revenue increased by 21% in Q3 2025, driven by improvements from AI-driven advertising platforms [4][7] - Baidu's AI business revenue grew over 50% year-on-year, becoming a core growth engine despite a decline in overall revenue [4] AI Integration in Business - Major companies are embedding AI into their core business lines to enhance operational efficiency and create new growth avenues [6] - Alibaba's AI tools improved operational efficiency for 500,000 merchants during the "Double 11" shopping festival, with average efficiency gains of 1.5 times [7] - Tencent's advertising revenue growth is attributed to AI-driven precision targeting, enhancing the effectiveness of ad placements [9] New Market Opportunities - The competition for AI-native applications is intensifying, with companies vying to establish new "super entry points" in the AI era [10] - The market has seen the emergence of leading AI applications, with ByteDance's Doubao and Tencent's Yuanbao gaining significant user bases [10] - The hardware sector is also evolving, with AI smart glasses and phones being developed to serve as new interaction points for users [12] Future Outlook - The AI landscape is still in its early stages, with ongoing competition among companies to validate user engagement and commercial viability [13] - The integration of AI into both enterprise and consumer applications is expected to blur the lines between the two, driving further innovation and growth [11]

Core Insights - ByteDance and ZTE launched the "Doubao Phone," intensifying competition for the next-generation super entry point in the industry [2] - Lenovo plans to unveil its first global "AI Super Intelligent Body" at CES, directly competing with Doubao Phone Assistant, focusing on cross-device connectivity [2][3] - Lenovo's strategy emphasizes a hybrid AI approach, integrating personal, enterprise, and public intelligence over the next decade [6] Group 1 - Lenovo is collaborating closely with ByteDance's Doubao and Volcano Engine, indicating a partnership despite potential competition with the new AI Super Intelligent Body [4][5] - The smartphone market is experiencing a "Matthew Effect," where Lenovo must leverage AI opportunities to enhance Motorola's global market position [8] - ByteDance's strategy involves pre-installing AIGC plugins on partner devices, empowering more hardware terminals with Doubao Phone Assistant's capabilities [9] Group 2 - Vivo and ByteDance are reportedly in discussions for collaboration, while Meizu has expressed interest in working with Doubao [10] - OPPO has denied rejecting collaboration with ByteDance, emphasizing an open approach to partnerships based on user experience [10][11] - The competition for AI super entry points is just beginning, with hardware manufacturers concerned about becoming mere hardware providers if they overly rely on ByteDance's Doubao Assistant [13]

Core Insights - The article discusses the evolution of AI technology and its commercialization potential, emphasizing the shift from text-based models to native multimodal models that can understand and process various types of data simultaneously [5][8][14]. Group 1: AI Technology Evolution - AI technology has faced challenges in finding practical applications, but advancements in models like DeepSeek and OpenAI's GPT-4o are reshaping user perceptions of AI's value [6][7]. - The introduction of native multimodal models, such as Baidu's Wenxin 5.0 and Google's Gemini 3, is expected to enhance AI's understanding of images, videos, and audio, thereby improving its commercial viability [12][14]. Group 2: Commercialization Challenges - The high cost of reasoning in AI models has been a barrier to widespread adoption, with predictions that reasoning tasks will consume over 50% of token usage by 2025 [17]. - Companies are focusing on reducing reasoning costs through full-stack optimization, which includes advancements in algorithms, architectures, and hardware [20][21]. Group 3: Competitive Landscape - The competition in the AI industry is evolving from merely scaling models to providing deeper intelligence at lower costs, with companies like Baidu and Google leading the charge [21][24]. - The concept of a "super entrance" is emerging, where companies are transitioning from traditional app-based platforms to intelligent multimodal assistants that can interact with users in more sophisticated ways [22][23]. Group 4: Strategic Developments - Baidu is leveraging its technological foundation to create a comprehensive ecosystem that integrates its AI capabilities with various applications, positioning itself as a leading player in the AI landscape [24]. - Tencent is also ramping up its AI efforts by establishing new departments and recruiting top talent to enhance its research and development capabilities [26].

曾就读于清华大学，获得电子工程学士学位和硕士学位。 一、从上千万到几万元 具身智能价格下探应用场景更多元 第一财经 高 远 嘉宾简介： 夏志进 淡马锡旗下 祥峰投资管理合伙人 2010年加入祥峰投资，主导创新科技领域的投资，重点关注半导体芯片、人工智能、先进制造等领域的 创业项目。 投资案例包括地平线[9660.HK]、极智嘉[02590.HK]、錼創科技[TWSE: 6854]、云英谷、松灵机器人、 移芯通信、中科原动力等。 加入祥峰之前，夏志进曾在汤姆逊研究与创新部担任研究工程师，致力于视频处理技术并获得多项国际 专利，也曾在泰山国际投资公司担任分析师。 祥峰投资 夏志进 我记得在十年前Boston Dynamics（波士顿动力）出了人形机器人，他们给高校科研使用一台，价格是 1000万人民币。所以您想想看，过去十年从1000万到现在几万人民币，机器人价格已经是多大的变化或 者降幅。 其实仔细来看，我觉得机器人不应该是一个价格很昂贵的产品，我们去看它的本体、电机，无非是一些 材料的组装。所以我觉得整个价格的下降趋势还会加快。 现在这个时间点上，有很多可以落地的场景。我上周在深圳还碰到一个美国的创业者，他们 ...

专题：未竟之约：张小珺访谈录 由新浪财经 、微博着力打造，微博财经 × 语言即世界工作室联合出品的泛财经人文对话栏目《未竟之 约》首期深度访谈即将上线。主持人张小珺对话金沙江创投主管合伙人朱啸虎，直面AI浪潮下的激流 与暗礁。 以下为对话实录： 张小珺：Hello，大家好，我是小珺。欢迎来到微博财经与语言及世界工作室联合出品的高端人物访谈 节目《未尽之约》，我们希望和还未完成愿望的人一起去抵达还未完成的旅途。 2024年3月，我曾经发表过一篇报道，叫作《朱啸虎讲了一个中国现实主义AIGC故事》，那使他以犀利 的观点为人所熟识。那么今天，我们将继续记录他在这波全球AIGC浪潮中的新鲜的辛辣的观。 辛辣的观察。 新鲜的辛辣的观察。 今天，我们将继续记录他在这波全球AIGC浪潮中的新鲜的、辛辣的观察。 哈喽Allen，先给观众朋友们打个招呼。 朱总：大家好。 张小珺：这是我们近两年的第三次聊天，也是《朱哮虎讲了一个中国现实主义AIGC故事的第三次连 载》，我们想持续记录你在这波AI浪潮中的观察笔记。 朱啸虎：chatGPT会成为一个超级入口，对META构成威胁 张小珺：ChatGPT会成为一个新的超级入口吗？ 那从 ...

Core Insights - The global race for a "super application" in the AI field is intensifying, with significant investments focused not on the technology itself but on identifying the "super entry point" [1][2] - Alibaba's launch of the "Qianwen" project marks its entry into the AI to C (consumer-facing AI) market, positioning it as a competitor to ChatGPT [1][3] - The competition is characterized by two main fronts: the "engine war" focusing on the capabilities of large models, and the "entry war" which emphasizes the importance of productizing AI capabilities into indispensable tools [3][4] Industry Dynamics - The emergence of AI applications is creating strong ecological barriers, making them essential bridges between users and the digital world [2][4] - The uncertainty surrounding the form of the "super entry point" fuels a competitive arms race among AI players, with various potential forms such as advanced chatbots or integrated assistants [2][5] - The future of AI applications is expected to revolve around creating an "AI Agent" that not only provides information but also executes tasks, thus enhancing user experience [4][5] Competitive Landscape - Major players like OpenAI and Google are actively developing their ecosystems, with OpenAI introducing APIs and collaborative features, while Google integrates its AI models into existing platforms [4][5] - Alibaba aims to leverage its existing e-commerce, life services, and cloud computing ecosystems to create a comprehensive integration path with its "Qianwen" project [4][5] - The competition for the "super entry point" is critical, as it will determine the allocation of trillion-dollar market values and shape the global technology landscape for the next decade [5]

Core Insights - The competition among internet giants in the automotive sector is intensifying, with JD.com announcing a collaboration with GAC and CATL to launch the "National Good Car" during the Double 11 shopping festival, marking a significant entry into the automotive market [1][2] Group 1: JD.com's Strategy - JD.com is positioning itself not as a car manufacturer but as a seller, focusing on understanding consumer preferences and acting as an exclusive 4S store [2][3] - The collaboration with GAC and CATL reflects a clear division of responsibilities: GAC manufactures the vehicle, CATL provides battery technology, and JD.com handles sales and customer insights [2][3] - JD.com has been gradually building its automotive business since 2015, with significant milestones including the establishment of an automotive division and the expansion of its car maintenance services [3] Group 2: Competitive Landscape - The current strategies of internet giants in the automotive sector can be categorized into three types: direct manufacturing (like Xiaomi), providing technology solutions (like Huawei), and offering data and algorithm support (like Alibaba) [2] - JD.com’s approach emphasizes service and supply chain capabilities rather than manufacturing, leveraging its extensive resources to provide a full range of automotive services [4] Group 3: Future Outlook - The competition in the electric vehicle market is shifting from technology to ecosystem battles, focusing on comprehensive service offerings throughout the vehicle lifecycle [4] - JD.com is also exploring autonomous driving technologies, having registered a trademark for "Joyrobotaxi" and developed logistics vehicles with advanced autonomous capabilities [4] - The automotive sector is seen as the next "super entry point" for internet companies, with the potential to reshape consumer engagement and ecosystem integration [5] Group 4: Challenges Ahead - Despite the opportunities, JD.com faces challenges such as brand commitment risks, where any issues with the vehicle could reflect poorly on JD.com despite its non-manufacturing role [6] - The automotive industry requires significant investment and has long return cycles, posing a challenge for internet companies accustomed to lighter asset models [6] - The competitive landscape is crowded, with established players like Huawei and Tesla creating significant barriers in technology and data [6]

Summary of Key Points from Conference Call Records Industry and Company Overview - **Industry**: AI Applications and Social Media - **Company**: Sora App and its impact on traditional social media and short video platforms Core Insights and Arguments 1. **Sora 2's Technological Advancements**: Sora 2 enhances AI video realism by simulating physical world laws and synchronizing audio and video output, lowering the barrier for user-generated AI videos [1][2][4] 2. **Impact on Traditional Platforms**: Sora 2 may significantly affect traditional social media and short video platforms by attracting users with its low content creation threshold and innovative social features, potentially threatening platforms like Douyin and Kuaishou [4][10] 3. **AI Integration in Daily Tools**: The integration of ChatGPT with third-party applications signifies a deeper embedding of AI technology into everyday tools, prompting domestic companies to accelerate their AI ecosystem development [5][6] 4. **Chinese Internet Companies' Growth**: Chinese internet companies excel in scaling products, with e-commerce retail size three times that of the U.S. and mobile gaming contributing nearly 40% of the global market share [7] 5. **Domestic AI Developments**: Domestic companies have made significant strides in AI-generated images and videos, with Douyin's trending topic generating over 14 million views [8][9] 6. **Sora's Market Position**: Sora's rapid popularity has helped cultivate user habits for AI video consumption, serving as an incremental growth factor rather than a disruptive force for existing platforms [10] 7. **OpenAI's Shift to Monetization**: OpenAI is transitioning from model advancement to productization and monetization, with expectations that revenue from subscriptions will decrease while income from API calls will rise [11] 8. **Potential for Super Apps in China**: Major Chinese companies like Tencent, Alibaba, and ByteDance have the potential to create super apps that could rival the complete mobile application ecosystem [12] 9. **IP Revenue Sharing**: The use of classic anime IPs in AI-generated content raises copyright issues, with potential revenue-sharing agreements between OpenAI and IP holders [13] 10. **Kuaishou's Recovery**: Kuaishou's C-end app revenue has stabilized and is slowly recovering, with a new model reducing costs by 30% and achieving global leadership in image-to-video capabilities [14][15] Additional Important Insights 1. **Keling's Competitive Edge**: Keling's traditional DIT architecture offers unique advantages in video modeling, with a market potential of $140 billion, where AI could penetrate $50 billion in the short term [16] 2. **Sora's Limitations**: Sora's lack of detail control makes it less suitable for artistic creators compared to Keling, which supports collaborative workflows [17] 3. **Keling's Growth Prospects**: Keling's revenue is expected to grow significantly, with a forecasted annual revenue of 2 billion RMB, driven by overseas market expansion [18] 4. **Kuaishou's Advertising Growth**: Kuaishou's advertising business is projected to grow steadily, with a target market value of 500 billion RMB by 2026 [19] 5. **Bilibili's AI Integration**: Bilibili is testing a new AI tool that helps creators generate videos quickly, potentially increasing active creator numbers significantly [20][21] 6. **Bilibili's Game Development**: Bilibili's upcoming game "Three Kingdoms Hundred Generals" is expected to become a major revenue driver, with a projected annual revenue of 1 to 2 billion RMB [23] 7. **Meitu's Overseas Strategy**: Meitu's overseas revenue has grown significantly, with a 56% year-on-year increase, driven by its core products [29] 8. **Focus on AI in E-commerce**: Worth Buying is enhancing its AI capabilities, including a new shopping assistant app, which is expected to improve user experience and drive revenue growth [36][37] 9. **Wondershare's AI Revenue**: Wondershare's AI revenue is projected to double this year, driven by updates to existing applications and a growing mobile user base [38] 10. **Guangyun Technology's Growth**: Guangyun Technology's core agent products are experiencing rapid growth, with a focus on e-commerce platforms and a positive outlook for profitability [39][41]