Core Viewpoint - The establishment of the Jiangsu Provincial Government Cloud marks a significant milestone in the province's digital government construction, transitioning from a fragmented system to a unified cloud platform [1] Group 1: Government Cloud Infrastructure - The government cloud serves as a core foundational platform for digital government, utilizing cloud computing and intelligent tools to provide computing resources, storage, service support, and security for various business applications [1] - The Jiangsu Provincial Data Bureau oversees the government cloud, with the Provincial Big Data Management Center managing operations and the Provincial Data Group acting as the cloud service provider [1] Group 2: Integration and Efficiency - The official launch of the provincial government cloud enables 100% of the core business systems of 95 provincial units to migrate to the cloud, indicating a shift to a "one cloud" integration phase [1] - The Provincial Government Cloud Dispatch Center acts as the command center for cloud operations, ensuring unified management and flexible allocation of cloud service resources, enhancing the efficiency of government services and decision-making [1]

Fixed Income - The report analyzes the new regulatory framework for insurance companies, highlighting five major changes aimed at improving asset-liability management [9][10] - The insurance industry's solvency indicators declined in Q3 2025, attributed to increased equity capital usage and dual pressure on liabilities [11] - Investment returns improved due to rising long-term bond yields and favorable A-share performance, reducing overall industry risk [12] - A selection of insurance subordinated bonds with specific criteria is recommended for investment, including those from major state-owned insurance companies [14] Pharmaceutical Industry - The pharmaceutical index fell by 3.31% recently, with a notable trend of Chinese innovative drugs expanding into international markets [15][17] - Companies like China Biopharmaceutical are enhancing their R&D capabilities through acquisitions and self-developed products, positioning themselves for potential overseas opportunities [18] - A focus on innovative drugs is recommended, with specific companies highlighted for their strong fundamentals and potential for stock price recovery [19] New Consumption - Ruoyu Chen and Wancheng Group are projected to see significant profit growth in 2025, driven by their proprietary brand business and operational efficiency improvements [21][23] - The government is promoting new service consumption growth points, aiming to enhance service supply and consumer experience [25][26] - The snack retail sector is evolving, with companies like Mingming Hen Mang rapidly expanding and innovating to meet consumer demands [24] North Exchange - The North Exchange indices saw over 6% growth in January 2026, indicating a market rebound and highlighting specific sectors for investment opportunities [28][29] - The overall market liquidity remains ample, with a focus on structural investment opportunities in specialized and innovative enterprises [29][30] Media - Alibaba's FY2026Q3 revenue is expected to reach 288.1 billion yuan, with cloud business growth anticipated despite a slowdown in e-commerce [32]

付强建议，全省国资企业尽快摸清全省国资系统现有算力资源(CPU、GPU、存储、网络)总量与分布， 建立算力资产台账。提出2026年、2027年、2028年三个关键节点的算力发展目标，特别是AI算力占比 与增长路径。根据综合办公、经营管理、生产运营等不同场景，划分通用算力、智能算力、高性能算力 等类型，实现差异化配置与精准供给。 付强建议，明确客户范围，将省属企业、市属国企、参股企业及符合条件的产业链上下游合作单位纳入 目标客户体系，分阶段开放接入权限。优化考核机制，将"使用云算力中心资源比例"纳入相关企业年度 绩效考核指标，对优先采用中心算力资源的企业给予加分激励。推动"以用促建"，通过实际应用反馈持 续优化平台功能、服务能力和用户体验，形成"建设—使用—优化"的良性循环。推进考核松绑，将国资 企业采购云算力中心服务的资本性支出在年度经营业绩考核中予以合理扣除，不计入当期利润考核压 力。推动正向激励，对承担基础设施建设、平台开发、运维保障等重点任务的企业，在科技创新、数字 化转型等专项考核中给予加分，激发企业主动作为。 付强建议，以信创标准为引领，构建自主可控的云底座，全面适配国产芯片、操作系统、数据库与中间 ...

财经频道更多独家策划、专家专栏，免费查阅>> 责任编辑：山上 据"微讯江苏"公众号，2月1日，重大基础设施工程材料全国重点实验室理事会2026年第一次会议及2025 年度学术委员会在南京召开。现场，东南大学全球首发"砼真砼知"混凝土材料科学大模型。该大模型已 应用在南京北站建设项目中。该模型由重大基础设施工程材料全国重点实验室联合东南大学大数据中 心、阿里云计算有限公司共同打造，是国内首个混凝土材料科学大模型。 ...

证券研究报告|行业研究周报 [Table_Date] 2026 年 2 月 1 日 算力供给紧张，成为 2026 年主线 [Table_Title2] 计算机行业周报 [Table_Summary] 本周观点: 算力供给紧张，成为 2026 主线 全球 AI 基建需求依然强劲，26 年海外 AI 算力依然是主要配置方 向。TrendForce 预测 2025 年全球八大主要 CSPs 资本支出(CapEx) 总额约 4,300 亿美元(yoy+65%),预期 2026 年 CSPs 仍将维持积极 的投资，资本支出将进一步推升至 6,000 亿美元以上(yoy+40%), 展现出 AI 基础建设的长期成长潜能。 海外大厂创造 2026 算力新格局 根据财经宇宙，微软 CEO Satya Nadella 表示未来微软从 OpenAI 的"独家运营商"，转型为全球 AI 资源的总调度商，让企业客户 能根据成本与性能需求，在 GPT 5.2、Claude4.5 等多款大模型间 灵活选择。微软正通过这种饱和式的基建投入，试图在 AI 从技术 神话下沉为工业底座的过程中，完成从一家"软件服务商"向未来 "算力基础服务商" ...

Investment Rating - The report suggests a positive outlook for the AI-driven sectors, particularly in servers and IDC, driven by domestic and international AI developments [4]. Core Insights - Major companies like Industrial Fulian, Meta, and Microsoft are showing strong growth in their earnings, driven by AI investments and demand for high-performance computing [1][53]. - The report highlights a significant increase in capital expenditures by tech giants, indicating a robust investment trend in AI infrastructure [1][61]. - The demand for optical modules and servers is expected to rise due to the ongoing expansion of AI capabilities and infrastructure [2][35]. Summary by Sections Industry Overview - The telecommunications sector is experiencing steady growth, with a reported revenue of 16,096 billion yuan, reflecting a 0.9% year-on-year increase [3][14]. - The optical module exports have seen a decline, with a 24% drop in November, attributed to domestic companies establishing overseas factories [3][35]. Company Performance - Industrial Fulian anticipates a net profit of 351-357 billion yuan for 2025, marking a 51%-54% increase year-on-year, driven by strong growth in cloud services and AI servers [1][53]. - Meta's revenue for Q4 reached $59.893 billion, a 24% increase year-on-year, supported by recovering advertising business and AI investments [1][53]. - NewEase is expected to achieve a net profit of 94-99 billion yuan in 2025, reflecting a growth of 231.24%-248.86% due to increasing demand for high-speed products [1][55]. Market Trends - The server index decreased by 5.70% this week but has increased by 1.20% for the month, indicating volatility in the market [2][6]. - The optical module index rose by 10.07% this week, reflecting strong demand for high-speed optical devices [2][6]. - The IDC index increased by 3.50% this week, with a notable 11.04% rise for the month, driven by advancements in AI models [2][9]. Future Projections - Companies are expected to continue investing heavily in AI infrastructure, with Alibaba planning to increase its investment in AI and cloud computing from 380 billion yuan to 480 billion yuan over the next three years [1][12]. - The report anticipates that the demand for AI-related services and products will continue to grow, leading to further investment opportunities in the sector [4][35].

Investment Rating - The report maintains a "Buy" rating for key companies in the sector, including Zhongji Xuchuang, Xinyi Sheng, and Tianfu Communication, among others [11]. Core Insights - The transition to the "Agent" era is accelerating, with AI agents like Clawdbot and Claude Excel evolving from simple conversational tools to essential productivity tools, significantly increasing demand for cloud computing resources [1][20]. - The AI application landscape is fundamentally shifting from simple interactions to deep integration into work processes, marking a new paradigm where AI becomes a core executor in workflows [2][21]. - The emergence of AI agents is driving a qualitative change in underlying computing power demand, as these agents transition from auxiliary tools to autonomous entities [3][22]. - The competition for computing power is intensifying, with rising cloud service prices and continuous growth in capital expenditures (capex) from cloud providers, highlighting the increasing importance of scarce computing resources [4][23]. Summary by Sections Investment Strategy - The report suggests focusing on the computing power sector, particularly in optical communications, with recommended companies including Zhongji Xuchuang, Xinyi Sheng, and Tianfu Communication [14]. - It also highlights the importance of liquid cooling and edge computing platforms, recommending companies like Yingweike and Meige Intelligent [14]. Market Review - The communication sector has seen an increase, with optical communications performing particularly well, as evidenced by significant stock price increases for companies like Tianfu Communication and Zhongji Xuchuang [16][17]. Computing Power Demand - The report emphasizes that the new operational modes of AI agents lead to a dramatic increase in token consumption, necessitating high concurrency and continuous online capabilities from computing services [5][25]. - Major cloud providers like Google Cloud and Amazon AWS have announced price increases for their services, indicating a shift from a long-term trend of decreasing prices [10][23]. Key Companies and Performance - Zhongji Xuchuang and Xinyi Sheng have shown strong growth, with Zhongji Xuchuang maintaining its leading position in the industry [25]. - The report recommends monitoring the performance of companies involved in the computing power supply chain, including those in optical modules and liquid cooling sectors [25].

Core Viewpoint - Baillie Gifford continues to focus on technology and growth sectors, maintaining a global perspective despite a reduction in total assets held to $120 billion, down $15 billion from the previous quarter [1]. Holdings Overview - As of Q4 2025, Baillie Gifford holds a total of 265 companies with assets amounting to $120 billion, reflecting a decrease of $15 billion from Q3 [1]. Top Holdings - The firm remains heavily invested in global technology and consumer giants, with a notable emphasis on Latin American e-commerce platform MercadoLibre, and Nvidia leading the portfolio, indicating confidence in the long-term potential of AI infrastructure [4]. Top Buys - Significant increases in holdings include: - Axon Enterprise (AXON) with an additional investment of $860 million - Medline Inc (MDLN) with $437 million - Rocket Lab (RKLB) with $399 million - Google-A (GOOGL) with $373 million - Duolingo (DUOL) with $357 million - The firm shows strong interest in vertical technology sectors, particularly in safety technology and education [5][6]. Top Sells - The firm has reduced its positions in: - Datadog Inc (DDOG) by $1.01 billion - Cloudflare (NET) by $990 million - Shopify (SHOP) by $680 million - BioNTech SE (BNTX) by $630 million - Meta Platforms (META) by $590 million - This indicates a strategic retreat from cloud computing and vaccine-related companies, likely based on valuation and growth switching logic [7][8]. New Positions - New investments include: - Medline Inc (MDLN) with $437 million - United Therapeutics (UTHR) with $157 million - MongoDB (MDB) with $112 million - BillionToOne (BLLN) with $104 million - Silvergate Capital (SVM) with $35 million - The new positions reflect Baillie Gifford's exploration in emerging fields such as rare disease treatment and databases, showcasing a commitment to long-term tracking in technology and healthcare [9][10].

Core Insights - The acceptance of large models by financial institutions is continuously increasing, with 587 projects across various sectors including banking, securities, insurance, and more [2] - The banking sector remains the primary adopter of large models, projected to have 290 projects by 2025, accounting for 49.4% of the total [2] - Financial applications are the leading demand for large models, with 312 projects expected by 2025, representing 53% of the total [3] Group 1: Market Trends - The top five companies in terms of project bids include Yudu, Keda Xunfei, Huoshan Engine, Zhongguancun KJ, and Awang Cloud, with bid amounts of 602.1 million, 588.1 million, 530 million, 186.5 million, and 308 million respectively [1] - The application of AI in finance is becoming the primary direction for large model implementation [2] Group 2: Application Scenarios - The leading application scenarios for large models in finance include intelligent customer service and digital humans (81 projects), knowledge Q&A and platforms (35 projects), intelligent auditing and decision-making (28 projects), intelligent programming (15 projects), and content generation (14 projects) [3] - A significant increase in internal model service usage has been reported, with daily token usage surpassing 10 billion, indicating a shift from pilot phases to large-scale implementation [3] Group 3: Technological Advancements - Financial institutions are increasingly seeking specialized models for credit risk control, transaction monitoring, customer service, and compliance review [5] - Baidu Intelligent Cloud has gained a competitive edge in the financial sector due to its comprehensive AI cloud stack capabilities, providing system-level optimization solutions [6] - Collaborations with major banks, such as the partnership with China Merchants Bank, have led to enhanced performance in multi-modal data analysis and intelligent customer service applications [6]

Core Viewpoint - The report from CITIC Securities highlights a strong demand for inference and training computing power overseas, leading to price increases from both Amazon Cloud and Google Cloud [1] Group 1: Inference Demand - The launch of Agent products like MoltBot and Claude Code has significantly increased the demand for cloud computing resources [1] - Token call volume has been experiencing rapid growth for 2-3 consecutive weeks since early 2026 [1] Group 2: Training Demand - Models such as Grok-5 and Veo4 are continuously iterating, with the industry exploring scaling limits to support training computing power demand [1] Group 3: Future Outlook - Although large-scale commercialization of AI applications remains uncertain, the next 3-6 months are expected to see intensified AI application deployment on the inference side, coupled with ongoing model iterations on the training side [1] - CITIC Securities predicts that computing power demand is likely to rise further, and confirmation of this demand during the U.S. earnings season could alleviate previous concerns about a "computing power bubble" affecting sentiment and valuations in the sector [1] - The computing power industry chain may enter a new phase of growth [1]