上周市场震荡，上证指数下跌0.45%，深证成指上涨1.14%，创业板指上涨1.00%，沪深300指数下跌 0.57%，品牌工程指数上涨0.60%，报2059.78点。 □本报记者 王宇露 上周市场震荡，中证新华社民族品牌工程指数上涨0.60%，报2059.78点。长电科技、华润微、广联达等 成分股上周表现强势；开年以来，安集科技、中微公司、长电科技等成分股涨幅居前。展望后市，机构 认为，2026年宏观政策和产业层面积极因素居多，出现系统性风险概率较低，市场整体具备趋势向上的 基础，结构选择仍是投资关键。 多只成分股表现强势 上周市场情绪出现较大波动。星石投资表示，短期来看，在监管政策的导向下，市场炒作情绪较前期有 所下行，叠加当前A股整体融资余额占比处于相对合理水平，此次调整对市场场内流动性的实际影响不 大，更多是对短期情绪端产生影响。这种政策信号大概率不会影响市场方向，只是对资金流动方向存在 影响。向后看，随着年报业绩预告披露开启，行业景气线索明晰，资金对具有业绩支撑板块的关注度会 有所提升。 星石投资认为，中期维度下，不同行业表现差异较大，部分板块仍处于偏低位置，随着国内政策发力带 动供需平衡、价格回升， ...

Group 1 - Keda Xunfei's Chairman Liu Qingfeng stated that the domestic AI infrastructure has taken initial shape, with domestic large models matching international standards despite having half the parameters [2] - Michael Burry warned that the era of tech giants earning huge profits with minimal investment is ending, primarily due to AI, and investors should focus on Return on Invested Capital (ROIC) rather than revenue growth [3] - A BlackRock survey revealed that while investors are optimistic about AI, they are shifting their investment focus towards energy and infrastructure suppliers, with only one-fifth considering large US tech companies as attractive investment opportunities [4] Group 2 - Demis Hassabis, CEO of Google DeepMind, indicated that Chinese AI models are only a few months behind those in the US and Western countries, with significant advancements made by Chinese developers [5] - DeepSeek released a new paper on conditional memory, significantly improving model performance in various tasks, and has open-sourced a related memory module [6] - Wang Xiaochuan, CEO of Baichuan Intelligent, mentioned that the company has 3 billion yuan on hand and may initiate an IPO plan in 2027 [7] Group 3 - Zhiyu and Huawei launched the first domestically trained multimodal SOTA model on local chips, achieving a full training process on the Ascend Atlas 800T A2 device [8] - Kuaishou announced that Keling AI's revenue exceeded $20 million in December 2025, with an annual recurring revenue (ARR) of $240 million [9] - Yongyou Network projected a net loss of 1.3 to 1.39 billion yuan for 2025, although it expects to reduce losses compared to the previous year [10] Group 4 - JD.com and Lenovo deepened their "hybrid AI" cooperation, launching new products at CES 2026, with a focus on strategic collaboration around smart devices and services [11] - Alibaba's Qianwen app integrated with various services, allowing users to order food and book flights through AI, marking a significant upgrade in functionality [12] - Alipay and partners released China's first AI commercial agreement, aimed at creating a universal language for AI tasks across platforms [13] Group 5 - Yunhai Medical launched the "YunJian AI Spirit," a product that reduces long-term costs for users by offering unlimited access to traditional Chinese medicine infrared algorithms [14] - Zhiyuan purchased thousands of hours of robot training data for various tasks [15] - Meituan released the open-source "ReThink" model, achieving state-of-the-art performance in several benchmarks [16] Group 6 - Teslian introduced the upgraded T-Cluster 512 super node architecture, designed for high-performance AI model training, with a total computing power exceeding 500 PFlops [17] - Keda Xunfei launched a marketing AI platform based on the "SuperAgent" framework, enhancing efficiency in marketing strategies [18] - The first domestically trained text-to-image model was released by Zhiyu and Huawei, completing the entire training process on local chips [19] Group 7 - Tencent Cloud ADP launched the first "AI-native Widget," enhancing task delivery experiences through natural language interaction [20] - Anthropic implemented stricter measures to prevent third-party tools from bypassing rate limits, affecting several developer projects [21] - Google announced a partnership with Walmart to expand AI model shopping capabilities, allowing direct transactions through its AI application [22] Group 8 - Mark Zuckerberg initiated the "Meta Compute" project, aiming to build substantial AI infrastructure by 2030, with a focus on collaboration with governments [23] - Meta plans to lay off hundreds of employees in its Reality Labs department, shifting focus from the metaverse to AI [24] - Alphabet's market value surpassed $4 trillion for the first time, joining a select group of companies [24] Group 9 - Nvidia and Eli Lilly will jointly invest $1 billion to establish an AI drug laboratory over the next five years [26] - The US relaxed export controls on Nvidia's H200 chips to China, potentially impacting the AI hardware market [27] - Microsoft announced a plan to limit the impact of data center energy costs and water usage on local communities [29] Group 10 - OpenAI is reportedly seeking US hardware suppliers for its planned consumer devices and cloud data center expansion [32] - Elon Musk's lawsuit against OpenAI is set to go to trial in late April [33] - OpenAI and Cerebras announced a partnership worth over $10 billion to deploy a large-scale AI inference platform [34] Group 11 - Zivariable Robotics completed a 1 billion yuan A++ round of financing, backed by major investors including ByteDance and Meituan [35] - Qiangnao Technology submitted a confidential IPO application in Hong Kong [36] - OpenAI agreed to acquire the AI health application Torch for approximately $100 million [37] Group 12 - K2 Lab, founded by a former DingTalk executive, secured tens of millions in seed funding to develop an AI-driven content e-commerce agent [38] - Alibaba Cloud completed a strategic investment in ZStack, achieving a controlling stake [39] - Skild AI raised nearly $1.4 billion in funding, reaching a valuation of over $14 billion [40] Group 13 - WeLab completed a $220 million D-round strategic financing, the largest single round since its inception [41] - Merge Labs, a brain-machine interface startup, raised $252 million in seed funding, with OpenAI as a major investor [42] Group 14 - A report indicated that by 2026, the Chinese tech giants index is expected to surpass the US tech giants in profitability growth for the first time since 2022 [43] - China is accelerating the establishment of a data property registration system to enhance data circulation and value [44] - Storage prices are expected to surge by 40%-50% in Q4 2025 and again in Q1 2026 due to increased demand from AI and server capacity [45] Group 15 - A new AI model developed by US researchers can predict the risk of approximately 130 diseases based on sleep data [46] - Foreign investment firms are increasingly incorporating AI into their research processes in the Chinese market [47] - UBS believes the probability of an AI bubble in China is low, with monetization relying on cloud services and advertising [48] Group 16 - The number of AI companies in China has exceeded 6,200, with applications expanding across various industries [49]

数智周报将整合本周最重要的企业级服务、云计算、大数据领域的前沿趋势、重磅政策及行研报告。】 观点 科大讯飞刘庆峰：自主可控的AI基础设施已初步成型 科大讯飞董事长刘庆峰在第九届全球深商盛典暨中国企业家俱乐部20年活动上表示，全国产算力平台上，国产大模型在参数小一倍的情况下可对标国际领先 水平。在芯片受限的背景下，自主可控的AI基础设施已初步成型。 "大空头"Michael Burry：科技巨头赚取巨额利润的时代将终结，AI时代的关键指标是ROIC 知名投资者、"大空头"Michael Burry警告，大型科技公司靠相对少的投资赚取巨额利润的时代正在结束，AI是主因。他认为投资人应关注投入资本回报率 (ROIC)，而非营收增长或市场规模。他指出，AI正推动微软、谷歌和Meta等公司，从过去轻资产的软件模式，转向由数据中心、芯片和能源主导的资本密 集型的硬件公司。即使AI帮助科技巨头扩大了市场，ROIC下降仍可能在未来数年对股价造成压力。 贝莱德投资者调查：尽管投资者看好人工智能前景，但将投资重点转向能源和基础设施供应商 贝莱德的调查报告显示，尽管投资者看好人工智能前景，但将投资重点转向能源和基础设施供应商；贝莱 ...

作者丨欧雪 编辑丨袁斯来 随着AI与营销加速融合，单一功能的Agent已难以满足复杂营销场景需求。 这种重构首先体现在协作方式的改变上。传统营销往往存在部门壁垒——数据分析、内容创意、媒介投 放各自为政。而智能营销时代，需要的是跨职能的协同工作流。"优化师能快速定位策略问题，创意团 队能知道哪个素材跑得好，这种统一视角正在改变传统的协作模式。"王云霞指出。 更深层的变革在于价值衡量体系的重建。当62%的用户在获取AI答案后不再点击传统链接，传统的点击 率、转化率等指标已无法全面反映营销效果。行业需要建立新的度量标准，能够衡量品牌与消费者之间 建立的长期信任和价值连接。 聚焦AI的未来机遇，Ronen Mense表示，AI在营销行业中的定位，正从可选择的效率辅助工具，转化为 深度赋能核心业务流程的基础支撑。 此外，科大讯飞副总裁、AI营销业务群总裁李平在峰会中提出了一个前瞻判断：SuperAgent将驱动营 销关系从B2A（Brand to Agent）演进至B2A2C（Brand to Agent to Consumer）。在此模式下，AI不 再仅是品牌的提效工具，更成为连接品牌与消费者的智能桥梁。 "数据割 ...

Core Viewpoint - The A-share market experienced moderate fluctuations with a total trading volume returning to over 3 trillion yuan, indicating a shift in capital from popular sectors like AI applications and communications to storage chips, automotive chips, and robotics actuators [1][4][6]. Market Performance - The Shanghai Composite Index fell by 0.26% to 4101.91 points, while the ChiNext Index decreased by 0.2% to 3361.02 points, and the Shenzhen Component Index dropped by 0.18% [6]. - The total trading volume across the three markets reached 3.06 trillion yuan, with margin trading balances increasing to 2.72 trillion yuan as of January 15 [6][16]. Sector Performance - The storage chip sector rose by 5.54%, third-generation semiconductors by 3.88%, automotive chips by 4.94%, and robotics actuators by 4.26% [8]. - In contrast, sectors such as media and computing saw significant declines, with media down by 4.84% and computing by 2.23% [9]. Capital Flow and Investment Strategy - Capital is being reallocated due to regulatory measures aimed at risk prevention and monetary policies supporting liquidity, leading to a diversified market structure [4][16]. - Investment strategies should focus on policy guidance, performance support, and valuation matching, particularly in sectors benefiting from domestic substitution and global chip technology breakthroughs [4][22]. Future Market Outlook - The market is expected to continue its oscillation around the 4100-point mark, with potential for structural opportunities in policy-supported sectors and high-performing stocks [19][20]. - Analysts suggest maintaining a neutral position while avoiding high-volatility stocks and focusing on sectors with strong fundamentals [19][22].

Core Insights - Ant Group's AI health application "Afu" gained significant market attention with a monthly active user (MAU) count of 30 million within a month of its December 2025 release, indicating a strong interest in AI applications in health management [2] - Major tech companies like Baidu, JD Health, ByteDance, and others are increasingly active in the medical AI sector, reflecting a resurgence of interest in this field [3] - The strategic focus of these companies has shifted from merely replacing healthcare professionals to enhancing and empowering them, aiming for an integrated service model that connects medical, pharmaceutical, insurance, and testing services [3][4] Company Strategies - Ant Group's "Afu" offers three core functions: health companionship, health Q&A, and health services, leveraging its ecosystem to provide end-to-end service from consultation to payment [5] - Baidu's "Wenxin Health Manager" utilizes its search engine traffic and AI technology but faces challenges in converting users from information seekers to service users [6] - JD Health's "Kangkang" has achieved stable profitability, primarily through pharmaceutical retail, while its AI services enhance efficiency [6] Market Dynamics - The medical AI sector is characterized by a divide between horizontal platform players (like Ant Group and Baidu) and vertical specialists (like ByteDance and iFlytek), each pursuing different strategic paths [4][7] - The demand for AI in healthcare is driven by the need for efficiency in a system facing resource distribution challenges, with 71% of Chinese clinicians relying on AI tools to alleviate work pressure [8][9] - AI applications are expanding from disease treatment to proactive health management, creating broader opportunities for user engagement [8] Challenges and Opportunities - Despite the potential, the commercialization path for medical AI remains unclear, with issues such as low willingness to pay in primary care and regulatory hurdles [15][16] - The integration of AI in healthcare requires high-quality, standardized data, which is often difficult to obtain due to privacy and sharing constraints [13][16] - The sector's complexity necessitates a deep understanding of medical industry regulations and ethical considerations, making it a challenging landscape for tech companies [16]

Group 1: Industry Trends - The domestic AI large model industry is experiencing a critical turning point with intensified competition for C-end traffic and clearer commercialization paths [2][3] - Major companies are shifting focus from B-end empowerment to comprehensive efforts in the C-end market, leading to the emergence of "AI native super apps" [2][3] - The rapid growth of user engagement is evident, with ByteDance's Dola achieving over 10 million daily active users and the Kimi model from Moonlight achieving a monthly user growth rate of 170% [3][4] Group 2: Capital and Financing - The AI large model sector has seen significant capital activity, with Moonlight completing a $500 million Series C funding round, raising its valuation to $4.3 billion [4][5] - The industry is projected to generate over 10 billion in revenue by the end of 2025, indicating a shift from merely burning cash to demonstrating real monetization capabilities [4][5] - Companies are adopting differentiated capital strategies, with some focusing on immediate funding through technological advancements while others pursue long-term IPOs [4][5] Group 3: Company Developments - Alibaba's Qwen team launched several new models and applications, including the Qwen-Image-Edit model and the Qwen-Image-Layered model, enhancing capabilities in image generation and editing [11][12][14] - ByteDance's Dola and the Beanbag model have shown remarkable growth, with the latter's daily token usage surpassing 50 trillion, reflecting a tenfold increase year-on-year [9][20] - SenseTime's Kapi camera app has reached over 10 million users, becoming a leading choice in the photography app market [34] Group 4: Market Dynamics - The competition is shifting from simple parameter comparisons in chip performance to a focus on overall computational efficiency and cost-effectiveness across chips, systems, and software [6][7] - The AI large model industry is entering a phase characterized by differentiated competition and a focus on commercial performance, moving away from the narrative of merely burning cash [5][6] - The emergence of AI native applications is expected to enhance user experience and promote healthier business ecosystems [3][5]

五年级二班的讲台，50岁的黄丽琴有种老语文教师的沉稳和知性。她穿着深色长款羽绒服，配双红色棉 鞋。唯一显得"突兀"的，是手上拿着的不是课本，而是一块黑色平板。 小镇很偏远，从北京过来要花近一天时间：先是4小时高铁，再是3小时自驾。公路盘旋上升，温度渐 低。唯一的街道上，老鹰在头顶盘旋，鲜有车辆经过。 这里是安徽省金寨县天堂寨镇，是刘邓大军千里跃进大别山的前方指挥部所在地，海拔1700多米。小镇 最喧闹的地方正是我此行的目的地——天堂寨同心小学。午后，学生们正欢呼着把一批新捐赠的桌椅推 进教室，金属刮过水泥地，发出清脆的响声。 这座高山寒区里的小学，七八年前就被聚光灯照亮：作为全国最早的试点之一，它引入基于人工智能的 智慧课堂——学生可以对着平板背单词，老师用AI批改作业。 在"AI场景应用"如火如荼的今天，走进这所大山深处的小学，我看到，AI早已不再是新闻，而是一间间 教室里的日常。这些日常，试图回答那个人们都在关心的问题——AI到底能否弥补城乡教育鸿沟，让 教育，这件关乎所有人的事，变得更公平？ 这堂课要上《我的"长生果"》。 这是本学期最后一篇课文，要认的生词不少。前一天，黄丽琴已让学生预习。随着她的手指 ...

| 计算机行业资金流入榜 | | --- | | 代码 | 简称 | 今日涨跌幅（%） | 今日换手率（%） | 主力资金流量（万元） | | --- | --- | --- | --- | --- | | 002230 科大讯飞 | | 0.49 | 7.18 | 73894.05 | | 600131 国网信通 | | 2.59 | 5.45 | 14184.16 | | 000977 浪潮信息 | | -0.06 | 3.43 | 13300.58 | | 301095 广立微 | | 5.34 | 7.09 | 10925.09 | | 002331 皖通科技 | | 9.98 | 11.40 | 9563.39 | | 002180 纳思达 | | 2.33 | 2.54 | 8419.78 | | 002920 德赛西威 | | 1.00 | 2.29 | 8142.69 | | 301269 华大九天 | | 1.58 | 1.54 | 6894.44 | | 301638 南网数字 | | 2.49 | 26.05 | 6829.84 | | 000938 紫光股份 | | 0.67 | 3. ...

李平表示，解决这类复杂问题，Agent需要具有智能规划、多方协同、闭环迭代的能力。对此，基于 SuperAgent应用框架的营销智能体平台应运而生。 【破解复杂营销难题 科大讯飞推出SuperAgent智能体平台 】AI与营销正持续深度融合。2026年1月15 日，讯飞AI营销（AIMarX）提出营销SuperAgent的应用框架，并推出基于SuperAgent构建的营销智能 体平台。 "Agent已经全面进入了应用阶段，从应用价值层面来看，AI已经从早期的Copilot模式全面进化到了 Agent模式。"科大讯飞副总裁、AI营销业务群总裁李平指出，Agent已经应用在了营销场景的全链路 中，包括市场洞察阶段的受众分析和竞品分析，策略规划阶段的用户刻画和意向识别，内容创意阶段的 创意生成和创意结构，广告投放阶段的智能竞价和跨场景触达等，也极大推动了营销效率和效果。但 Agent只能解决一些简单的需求，面对营销场景中的复杂需求，现有的Agent仍显乏力。这种复杂需求并 非简单任务的总和，而是亟须行业认知的综合性难题。在未来，这类难题将成为品牌精准营销，攻克全 球市场的关键壁垒。 展望未来，李平认为，AI技术依然 ...