在人工智能技术快速迭代、大模型成为产业焦点的背景下，全球AI竞争格局逐渐清晰，技术多样性与商业化落地成为行业的核心议题。在这一背景 下，中国大模型企业的技术选择、组织路径与长期判断，逐渐成为观察全球AI竞争格局的重要样本。 谈及目前的通用大模型竞争格局，张鹏表示 "不会只剩一家"。相反，模型体系的多样性是技术持续进步的重要动力。 他强调，大模型行业早中期技术革新路径多元、市场空间广阔，不同玩家的差异化探索将形成各自的生存空间。虽然最终可能会向头部收敛，但基 于通用模型的生态应用将蓬勃发展，且当前行业技术溢价较高、创新活跃，不会出现大规模低价竞争。 作为通用大模型领域的代表性公司之一，张鹏表示，智谱自成立之初便以AGI为长期目标，专注通用大模型研发，拒绝非核心业务分散精力，同时 坚持研发投入优先保障技术领先，算力是研发成本的核心构成。 从1958年感知机奠定机器学习基础，到1966年首个聊天机器人Eliza验证图灵测试逻辑，从GPU技术突破解决算力瓶颈，到2017年Transformer论文 成为大模型技术基石，AI七十余年的发展历程，始终在目标与资源的动态不平衡中螺旋上升。 在节目中，张鹏将AI的技术演进类比 ...

Core Insights - The report emphasizes the potential for investment opportunities in commodities driven by geopolitical tensions, particularly in the Middle East, which may lead to rising prices for oil and precious metals in 2026 [2][5][6] - The A-share market is expected to experience short-term volatility due to geopolitical factors, but will likely refocus on domestic fundamentals and policy expectations in the medium term [3][15] - The report highlights a significant investment in AI applications by major domestic internet companies, indicating a competitive landscape focused on user habit formation and commercial viability [9][12] Market Overview - The report lists a "March Gold Stock Portfolio" featuring companies such as Poly Real Estate Group, CITIC Hanzhong, and Mindray Medical, indicating a focus on sectors like real estate, transportation, and healthcare [1][7] - The A-share market indices showed mixed performance, with the Shanghai Composite Index closing at 4162.88, up 0.39%, while the Shenzhen Component Index fell by 0.06% [1] - The report notes that the steel industry performed well, with a 3.37% increase, while sectors like construction materials and telecommunications saw declines [1] Commodity Insights - The report anticipates that geopolitical events will significantly impact oil and certain petrochemical product prices, with a focus on the implications of the closure of the Strait of Hormuz [5][29] - It is projected that Brent crude oil prices could exceed $80 per barrel due to potential supply disruptions from Iran, with historical comparisons to the 2022 Ukraine conflict [5][29] - The chemical industry is advised to focus on low-valuation leading companies and sectors benefiting from price increases under the "anti-involution" policy [28][33] AI Industry Developments - Major domestic internet companies invested over 4.5 billion yuan in promoting AI applications during the Spring Festival, marking a shift towards practical applications and user engagement [9][12] - The report highlights the rapid evolution of domestic AI models, with significant advancements in performance and market application, indicating a dual development path towards general models and vertical industry applications [10][12] - Concerns about AI replacing human jobs are noted, but the report emphasizes that current AI capabilities are more about enhancement rather than replacement [11][12] Investment Recommendations - The report suggests focusing on companies in the AI sector and those involved in the development of general models and industry-specific AI agents, such as MINIMAX-WP and iFLYTEK [13][12] - It also recommends monitoring traditional chemical leaders that are adapting to new materials and benefiting from improving industry conditions [33]

计算机 | 证券研究报告 — 行业点评 2026 年 3 月 1 日 AI 应用持续落地 计算机行业事件点评 国内互联网大厂于春节期间投入超 45 亿元推广 AI 应用，标志着 AI 竞争已进入应用 生态争夺与用户习惯培养阶段。国内大模型在性能与市场应用端也不断加速追赶。尽 管市场对 AI 取代人类产生阶段性担忧，但行业转向理性务实，更关注商业化落地。 我们认为，未来行业有望向"通用大模型"与"垂直行业应用"双线发展。 支撑评级的要点 投资建议 建议关注大模型及细分行业 AI Agent 领域企业，包括智谱、MINIMAX-WP、科 大讯飞、拓尔思、汉得信息、中控技术、宝信软件、索辰科技、万兴科技、金山 办公、合合信息、用友网络、道通科技、石基信息、税友股份、安恒信息等。 评级面临的主要风险 产业推进不及预期，技术研发进展不及预期，国际合作难以推进。 证券分析师：杨思睿 (8610)66229321 sirui.yang@bocichina.com 证券投资咨询业务证书编号：S1300518090001 证券分析师：郑静文 jingwen.zheng@bocichina.com 证券投资咨询业务证书编号：S1 ...

Core Viewpoint - The article discusses how NetEase has chosen a pragmatic approach in the AI era, avoiding the costly competition of developing general-purpose AI models while focusing on application-level innovations and maintaining a strong R&D investment [3][20]. Group 1: Market Context - During the recent Spring Festival, major tech companies like Alibaba, Tencent, ByteDance, and Baidu spent over 4.5 billion yuan on "red envelopes," marking one of the most expensive tech competitions in history [2]. - NetEase, however, did not participate in this "red envelope war" or the race for large AI models, leading to questions about whether it is falling behind in the AI era [2][3]. Group 2: Business Strategy - NetEase's strategy is characterized by a focus on practical applications rather than competing in the foundational AI model space, which is seen as a more sustainable approach for most companies [3][20]. - The company has maintained a consistent R&D expenditure of over 15% of its revenue, with a projected R&D budget of 17.7 billion yuan for 2025, focusing on application layers rather than general model training [4][20]. Group 3: AI Integration in Products - NetEase has developed thousands of AI production pipelines that enhance various aspects of game development, achieving significant efficiency improvements, such as a 70% increase in design efficiency and a 50% boost in development efficiency through AI tools [6][8]. - The company has also applied AI in its educational and music platforms, enhancing user experience and operational efficiency without pursuing a general-purpose model [6][8]. Group 4: Financial Performance - In 2025, NetEase's total revenue is expected to reach 112.6 billion yuan, with operating profit at 35.8 billion yuan, driven primarily by its gaming segment, which saw a net revenue increase of 11% year-on-year [13][20]. Group 5: Future Growth Potential - The potential for growth in the gaming industry is seen in AI-native games, which are expected to generate over 30 billion yuan by 2027, contributing to a 10% market increase [13][20]. - NetEase's focus on integrating AI into its gaming products positions it well to capitalize on this emerging market, as it transitions from traditional gaming to AI-driven experiences [13][20].

在吴欣鸿看来，应用开发者在每个时代阶段都有空间，关键在于对高价值垂直场景的深度挖掘。这类场 景普遍存在非常刚性的需求和成本高效率低的问题，客户愿意为此付费，服务后也能创造出高弹性的增 长空间。而应用与通用大模型的竞争壁垒，关键在于能否建立"这个垂直场景我最专业"的心智，解决最 后一公里与长尾需求。吴欣鸿认为，当下通用大模型的对话式交互具有局限性，且提炼垂直行业能力的 门槛较高，需要垂直应用来释放大模型的能力。 同时，吴欣鸿也明确，美图公司正致力于成为一个不断产生优质影像应用的平台，打造更多垂直场景的 影像产品。他透露，即便是在Nano Banana发布之后，美图公司的应用数据依然在快速增长。 中证报中证网讯(记者黄一灵)近日，美图公司董事长兼CEO吴欣鸿在公司年会内部分享中表示，通用大 模型和应用之间并不是覆盖的关系，而是协同、相辅相成的关系。他把通用大模型比喻为"瑞士军刀"， 能够满足大部分的通用或者日常需求，而垂直应用就好比剪刀、剃须刀等专业工具，能满足不同场景下 的特定需求。 ...

Core Insights - The article discusses the evolution of artificial intelligence (AI) from a technical concept to a competitive landscape in key sectors such as healthcare, finance, and industry, highlighting a "arms race" among major players like Alibaba, ByteDance, and Tencent [1] - Ant Group's AI assistant "Afu" has made significant upgrades, including a "senior mode" targeting the elderly market, differentiating its strategy from competitors focused on general models [1][2] - Ant Group's CEO emphasizes the need for continuous innovation and warns against complacency, stating that the company is still a "follower" in the AI space [2][3] Ant Group's Strategy - Ant Group is focusing on three core areas: payment, finance, and healthcare, which are seen as essential for its future development [3] - The company aims to leverage its long-standing experience in payment and healthcare to create a unique advantage in the vertical model space [1][4] - Ant Group's AI assistant "Afu" has achieved over 10 million daily health inquiries, with a significant portion coming from users aged 50 and above, capitalizing on the aging population [1][3] Market Dynamics - The competition among major players is characterized by a focus on foundational models, with a significant emphasis on parameter scale, computing power, and ecosystem influence [4] - Ant Group's approach contrasts with the broader strategies of competitors, as it seeks to build deep understanding and service loops within the healthcare sector [4][5] - The healthcare market presents substantial opportunities, with a notable shortage of quality medical resources and a fragmented patient demand [5] Organizational Changes - Ant Group is enhancing its internal incentives for teams making innovative contributions in AI, indicating a commitment to fostering innovation [2][9] - The company has restructured its health business into an independent unit alongside payment and finance, reflecting a strategic focus on health [5][9] AI Integration in Payment - Ant Group is integrating AI into its payment systems, exemplified by the "Alipay AI Pay" feature that allows users to place orders using natural language commands [6][7] - The introduction of the ACT protocol aims to create a standardized framework for AI and e-commerce collaboration, enhancing service efficiency and user experience [6][8] Future Outlook - Ant Group's strategic evolution is closely tied to the concept of "inclusive finance," with significant investments in AI and technology infrastructure planned for the coming years [9][10] - The competitive landscape is intensifying, with major companies like Alibaba and Tencent also ramping up their AI investments, indicating a broader industry shift towards AI capabilities [10][11] - The article concludes with a perspective that Ant Group has the potential to change the competitive landscape, moving from being a variable in the AI ecosystem to a key player [13]

经济观察网 据美图公司微信公众号2月4日消息，美图CEO吴欣鸿在2026美图年会上表示，自己每次体 验通用大模型新版本的时候，也都在绝望和希望中反复横跳。吴欣鸿表示，通用大模型和应用之间并不 是覆盖的关系，而是协同的关系，是相辅相成的关系。应用与大模型的竞争关键点，在于解决各类垂直 场景下的复杂需求，占据"这个场景我最专业"的心智。 ...

Core Viewpoint - The discourse surrounding large models consuming applications has raised market concerns, leading to a collective setback in the AI application sector. Despite the release of Nano Banana, Meitu's application data continues to grow rapidly, indicating a synergistic effect between general large models and applications [1][3]. Group 1: Insights from CEO Wu Xinhong - Wu Xinhong, CEO of Meitu, acknowledges that while general large models are "omnipotent," the space left for application layers is diminishing. However, the efficiency of general large models in specific vertical scenarios is not very high [1][3]. - He compares large models to a "Swiss Army knife," capable of handling general needs and daily tasks, while vertical applications are likened to specialized tools that meet specific demands in various scenarios [1][3]. - Wu emphasizes that application developers always have opportunities at different stages, with the key being the deep exploration of high-value vertical scenarios that exhibit rigid demand and high costs, where customers are willing to pay, thus creating high elastic growth potential [1][3]. Group 2: Limitations and Focus Areas - Wu believes that the conversational interaction of current general large models has limitations, and the threshold for extracting vertical industry capabilities is high, necessitating vertical applications to unleash the potential of large models [2][4]. - Specific vertical application scenarios such as industry SOPs, vertical creator communities, high-precision editors, consistent batch outputs, material asset management, and team collaboration are areas where general large models may not perform well [2][4]. - Meitu is committed to becoming a platform that continuously generates high-quality imaging applications, focusing on creating more vertical scenario imaging products [2][4].

2026年开年，港交所的铜锣被敲得震天响，几家头部大模型独角兽的市值接连冲破千亿大关。 资本市场的空气里弥漫着金钱和多巴胺的味道，香槟开启的爆裂声此起彼伏。在满屏飘红的K线图下，投资人们笃定地相信，自己正在见证下一个万亿时代 的开端。 但如果我们极其扫兴地把这一层金粉刮掉，逼着所有人看一眼最枯燥、最没有想象力，但最诚实的"营收栏"，你会发现一个尴尬到令人窒息的断层。 翻开行业头部的招股书，你会看到一种普遍的"巨婴式繁荣"：营收规模往往只有几亿元，而同期的净亏损却高达几十亿，亏损额往往是营收的十倍。这意味 着每赚1块钱，就要赔进去9块5。这是一张典型的"互联网烧钱换规模"的旧船票。 而在时间轴的另一端，没有任何光环、甚至被主流科技圈视为"传统派"的云知声，却在角落里默默交出了一份令所有人侧目的成绩单：在2025财年统计口径 下，其大模型相关业务的预期收入已逼近1亿美元，约合6.0亿至6.2亿元人民币。 反观云知声，它的生意经完全不同。它的收入来自哪里？来自北京友谊医院的私有化部署，来自吉利、平安等车企的定制化座舱。 通用大模型充其量是企业的"高配文员"，它们在业务的表层打转，写写邮件、做做客服。但云知声切入的是 ...

（原标题：湖北省人大代表、中贝通信董事长李六兵：建议多措并举推动湖北AI产业高质量发展） 经济观察网 程久龙 1月27日上午，湖北省第十四届人民代表大会第四次会议在武汉开幕。"湖北两会"期 间，湖北省人大代表、中贝通信董事长李六兵在接受经济观察网采访时表示，湖北科教资源富集、产业 雄厚、应用场景多元，在全国AI发展格局中地位重要，湖北省应多措并举，大力推动AI产业高质量发 展。 二是、场景融合。推动6G与低空经济、应急救援等领域深度融合，探索"6G+卫星+AI"模式，带动芯 片、器件、终端应用等上下游企业协同发展，培育千亿级产业集群。 三是、平台赋能。在武汉都市圈、长江航运等区域部署试验平台，开展真实场景规模化试验，打通 从"实验室"走向"应用场"的关键转化通道，加速技术商业化落地。 在算力集群和算法领域，李六兵建议采用"场景化布局+平台化共享"模式，贴近教育AI、自动驾驶、智 能工厂等核心场景，推动龙头企业布局E级算力集群建设；设立算法创新专项，对共性技术研发平台给 予补助。 据经济观察网了解，中贝通信作为武汉本土企业，已构建"通信+AI+新能源"核心业务板块，具备AI全 链发展能力。公司还组建研发团队，计 ...