回顾2025年： AI引领、流动性新特征 驱动结构性分化行情 2025年全球主要市场的涨幅与AI算力进展呈现出显著正相关。AI 发展最为领先的美股市场表现最佳； 亚太市场中的韩国与日本也因AI相关产业活跃而涨幅居前。与此同时，AI云业务占比较高的恒生科技 指数也获得了内外资金的积极配置。在流动性方面，险资、被动基金及高净值个人投资者的共同参与 推动了市场资金面走强。受流动性改善支撑，A股微盘股及银行板块持续表现强势。值得关注的是， 本轮市场的增量流动性主要来自被动基金，这与过去两轮牛市有明显区别：2014-2015年主要由杠杆资 金驱动，2019-2021年则更多依赖主题/主动基金入市。 展望2026年A股：三大分化趋势 ——行业分化、盈亏分化、内外需分化 我们预计2026年A股基本面结构分化将进入下半场，A股将继续呈现三大分化趋势：（1）行业分化， 科创板块的利润贡献将持续挤占金融地产的份额；（2）盈亏分化，尽管全A 利润增长，但超过半数 公司的利润增速可能下滑；（3）内外需分化，海外营收占比较高的企业（超20%）其净利润贡献已超 过25%。这些分化意味着，由科技创新驱动的经济转型、上市公司经营的头部集中化 ...

Summary of Key Points from Conference Call Records Industry Overview - The focus is on the traditional computing power sector, with significant institutional buying in pure equity funds and a notable decline in sentiment towards convertible bond funds [1][2] - The precious metals market remains a key focus, with COMEX gold and silver experiencing volatility but recovering due to news of Zijin Mining's overseas acquisitions [2] Market Dynamics - The stock market experienced significant selling pressure early in the day, with over 4,500 stocks declining, but later recovered to close up 0.14% [1] - The overall trading volume decreased by 360 billion, falling to 2.92 trillion [1] - Growth sectors such as technology and emerging industries saw substantial gains, while traditional sectors like CPO, semiconductor, and storage rebounded sharply [1][2] Investment Trends - Institutional funds showed a continued increase in positions in equity assets, with a shift towards pure equity funds, indicating a potential influx of higher-risk capital [2][3] - The net subscription ratio for passive equity funds was reported at 1.61%, with significant buying from wealth management subsidiaries [3] - The net subscription ratio for active equity funds was slightly higher at 1.62%, with all institutions increasing their positions [3] Sector-Specific Insights - The automotive and machinery sectors received 40% of the inflows, followed by electronics, communications, and new energy [4] - The top sectors for increased investment included electronics, banking, automotive, new energy, and non-bank financials [5] - The convertible bond market saw a net subscription ratio of 0.27%, but overall buying volume significantly decreased compared to the previous two weeks [5] Risk Considerations - The convertible bond market is facing supply-demand imbalances, which could lead to significant volatility if demand weakens [2] - The lithium battery sector experienced a sharp decline due to a sudden drop in lithium carbonate prices, highlighting localized risks [2] Conclusion - The market is currently characterized by a return to traditional computing power investments, with a mixed sentiment in the convertible bond space and a focus on growth sectors. The overall trading environment remains cautious, with attention to sector-specific dynamics and potential risks.

Investment Rating - The report maintains an "OUTPERFORM" rating for Baidu Group [2] Core Views - Baidu's operating cash flow is expected to remain positive in Q4, supported by prudent spending and the integration of AI tools in operations [4][13] - The listing of Kunlun Chip (KLX) is anticipated to unlock significant valuation for Baidu, with potential revenue comparable to Cambricon [5][14] - Apollo Go is set to accelerate its global expansion in 2026, which may further support Baidu's stock performance [15] Financial Summary - Baidu's revenue for FY25 is projected at RMB 128.7 billion, a decrease of 2.4% year-on-year, with a net profit of RMB 18.6 billion, down 31% [3][11] - The core business is expected to generate an adjusted operating profit of RMB 14.1 billion, corresponding to a margin of 13.8% [11] - For Q4, total core revenue is estimated at RMB 25.8 billion, reflecting a year-on-year decline of 6.9% but a quarter-on-quarter increase of 4.5% [12] Valuation and Recommendations - The report employs a Sum-of-the-Parts (SoTP) valuation method, estimating a total market cap of approximately HK$481.8 billion or a target price of HK$178, corresponding to a 24x PE for FY26 [6][14] - The valuation breakdown includes traditional search ads, AI cloud, AI SaaS, AI ads, and other segments, with significant contributions from the Robotaxi and Kunlun Chip businesses [6][15]

Group 1: Macroeconomic Insights - The forecast for the US GDP growth in 2026 has been raised to 2.6%, up from a previous estimate of 2.3%, driven by factors including the upcoming midterm elections and potential policy changes from the Trump administration [2][3] - The report highlights a potential rebound in inflation in the second half of 2026, despite a slight downward adjustment in inflation predictions due to various economic factors [2][3] - Structural issues in the US economy, such as asset price inflation and income inequality, are expected to worsen, indicating a "K-shaped" recovery [2][3] Group 2: Real Estate Policies - The Trump administration has announced a series of real estate policies aimed at stimulating demand, including a $200 billion mortgage-backed securities purchase plan and restrictions on large institutional investors in the single-family housing market [3][4] - These measures are expected to provide marginal support to the real estate market but may not effectively address housing affordability issues [3][4] Group 3: Employment Data - In December, the US added 50,000 non-farm jobs, falling short of the Bloomberg consensus estimate of 70,000, with the unemployment rate decreasing to 4.4% [5] - The labor participation rate declined to 62.4%, while hourly wage growth showed a slight increase, indicating a mixed employment landscape [5] Group 4: Sector Performance - The report suggests that the A-share market is experiencing a "spring rally," with a focus on sectors such as gaming, duty-free, batteries, engineering machinery, and agricultural chemicals for potential investment opportunities [7] - The Hong Kong stock market is expected to benefit from a combination of factors, including improved liquidity and upward revisions in profit expectations [8] Group 5: REITs Market - Recent policies from the China Securities Regulatory Commission and stock exchanges are expected to enhance the REITs market, promoting high-quality development amid OCI disturbances [22] - The report indicates that the REITs market may experience improved sentiment and quality as it enters a new phase of development [22] Group 6: Semiconductor Industry - The report emphasizes the growing demand for cleanroom facilities in the semiconductor industry, driven by increased capital expenditure from global tech giants [23] - The cleanroom engineering services sector is expected to see a rise in order rates, benefiting from the demand for advanced manufacturing processes [23]

周一（12月22日），据CNBC报道，富国银行发布最新报告中指出，凭借其AI云基础设施的强劲潜力，甲骨文股价隐含近50%的上涨空间。 这一乐观展望是在科技板块面临剧烈波动、市场对AI驱动型企业估值过高及增长放缓的担忧日益加剧的背景下提出的。富国银行的报告强调，甲骨文在 被低估的AI云业务敞口将是支撑其股价看涨的核心驱动力。 截至2025年12月21日，甲骨文股价收于190美元左右。富国银行的最新预估暗示其潜在目标价可达285美元。 富国银行报告指出，凭借强劲的AI云基础设施潜力，甲骨文股价存在近50%上涨空间，目标价可达285美元。该行强调，甲骨文在被低估的AI云业务敞口 将是支撑其股价看涨的核心驱动力。分析指出，如果AI的应用加速落地，这将意味着该股面临显著的反弹机会。 富国银行认为，市场当前围绕人工智能（AI）投资的悲观情绪显得"过头"了，被严重低估的甲骨文正迎来重大的价值重估机会。 AI悲观情绪"过头"与被低估的云资产 富国银行在分析中详细阐述了看涨逻辑，核心观点在于甲骨文稳健的云基础设施和AI集成是被市场低估的资产。 分析指出，如果AI的应用加速落地，这将意味着该股面临显著的反弹机会，这与当前市场普遍 ...

这一乐观展望是在科技板块面临剧烈波动、市场对AI驱动型企业估值过高及增长放缓的担忧日益加剧 的背景下提出的。富国银行的报告强调，甲骨文在被低估的AI云业务敞口将是支撑其股价看涨的核心 驱动力。 截至2025年12月21日，甲骨文股价收于190美元左右。富国银行的最新预估暗示其潜在目标价可达285美 元。 富国银行认为，市场当前围绕人工智能（AI）投资的悲观情绪显得"过头"了，被严重低估的甲骨文正迎 来重大的价值重估机会。 周一（12月22日），据CNBC报道，富国银行发布最新报告中指出，凭借其AI云基础设施的强劲潜力， 甲骨文股价隐含近50%的上涨空间。 分析指出，如果AI的应用加速落地，这将意味着该股面临显著的反弹机会，这与当前市场普遍存在的 谨慎情绪形成了鲜明对比。 AI悲观情绪"过头"与被低估的云资产 富国银行在分析中详细阐述了看涨逻辑，核心观点在于甲骨文稳健的云基础设施和AI集成是被市场低 估的资产。 据CNBC引述的报告内容，甲骨文近期在AI数据中心的扩张及合作伙伴关系，使其在与竞争对手的博弈 中占据了有利位置。 分析师认为，当前市场对AI领域的悲观情绪主要源于对科技巨头高估值的审查，但这掩盖了甲骨 ...

Investment Rating - The report maintains an "Accumulate" rating for the industry [6] Core Insights - In recent years, several Bitcoin mining companies have partnered with AI cloud providers, utilizing part of their electricity for GPU hosting and developing AIDC (AI Data Center) businesses, with contracts totaling $53.6 billion and involving 1.959 GW of power [1][11] - The market perceives that Bitcoin mining companies primarily have an advantage in electricity resources compared to other AIDC competitors, but many have prior experience in self-operated cloud services [4][10] - The report suggests that companies with more AI electricity capacity tend to have higher market valuations, with average valuations of $4.73 million/MW for total electricity, $12.26 million/MW for mining electricity, and $144 million/MW for AI electricity [4][55] Summary by Sections 1. Current Status of Bitcoin Mining Transformation to AIDC - As of November 3, 2025, six Bitcoin mining companies have signed contracts with at least five cloud providers, totaling $53.6 billion and involving 1.959 GW of power [1][11] - Over 50% of the AI cloud providers that signed contracts with Bitcoin mining companies are emerging firms, with Coreweave being the largest [2][13] - The average contract duration for AIDC projects is 14 years, with a minimum of 5 years and a maximum of 25 years [3][24] - The average price for key IT load contracts is $1.73 million/MW/year, indicating a consistent pricing structure across projects [3][27] 2. Reasons and Models for Transformation - The transition to AIDC is driven by the uncertainty in revenue and profitability from Bitcoin mining, which faces challenges such as fluctuating prices and increasing competition [33][35] - AIDC offers higher gross margins compared to Bitcoin mining, with various projects reporting net operating income margins between 80% and 88% [40][39] 3. Self-Operated Cloud Services by Mining Companies - Companies like Iris Energy, Hive Digital, and Bitdeer are actively developing self-operated AI cloud services, with Iris Energy expected to operate 140,000 GPUs by the end of 2026 [43][47] - Bitdeer plans to operate 1,160 GPUs by the end of 2025, while Hive Digital aims to exceed 11,000 GPUs by the end of 2026 [44][51] 4. Future Outlook for AIDC Transformation - The report anticipates an acceleration in the transformation of Bitcoin mining companies to AIDC, as companies with more AI electricity capacity are valued higher in the market [55] - The report highlights the importance of companies having substantial electricity reserves and experience in AI cloud services for future investment opportunities [4][55]

Core Insights - Goldman Sachs reports that Kingsoft Cloud's Q3 revenue growth accelerated to 31% year-on-year, surpassing the previous quarter's 24% growth and meeting market expectations [1] Revenue Performance - AI cloud business revenue grew by 7% quarter-on-quarter and 116% year-on-year, accounting for 32% of total Q3 revenue [1] - Revenue from related parties, such as Xiaomi and the Kingsoft ecosystem, increased by 84% year-on-year, up from 70% in the previous quarter, driven by stronger AI training demand, representing 28% of total revenue [1] Profitability Metrics - Adjusted EBITDA reached 652 million yuan, exceeding both Goldman Sachs' and market expectations by 8% and 9% respectively, after excluding approximately 170 million yuan in one-time government subsidies [1] - Adjusted EBITDA margin improved by 14 and 9 percentage points year-on-year and quarter-on-quarter, reaching 26.3% [1] Future Projections - The firm raised its revenue forecasts for Kingsoft Cloud for 2026 and 2027 by 3% to 5%, and adjusted EBITDA forecasts by 5% to 7% [1] - The target price for US stocks was slightly increased from $13.5 to $13.8, maintaining a "Neutral" rating [1]

Summary of Baidu's Q3 2025 Earnings Call Company Overview - **Company**: Baidu - **Quarter**: Q3 2025 - **Total Revenue**: 31.2 billion RMB, a decrease of 7% year-over-year [2][13] - **Core Revenue**: 24.7 billion RMB, with online marketing revenue at 15.3 billion RMB, down 18% year-over-year [13] Key Highlights Financial Performance - **Non-Advertising Revenue**: 9.3 billion RMB, up 21% year-over-year, driven by AI cloud business [2][13] - **AI Cloud Infrastructure Revenue**: 4.2 billion RMB, up 33% year-over-year [2][13] - **Subscription-based AI Accelerator Revenue**: Increased by 128% year-over-year [2][4] - **Operating Loss**: 15.1 billion RMB, with a net loss of 11.2 billion RMB, translating to a diluted loss per share of 33.88 RMB [14][21] AI Cloud Business - **Growth Drivers**: Strong demand for AI products and solutions, with a shift towards sustainable recurring revenue models [4][16] - **Long-term Asset Impairment**: 16.2 billion RMB due to outdated infrastructure not meeting current efficiency requirements [4][14] Autonomous Driving (Apollo Go) - **Service Growth**: Provided 3.1 million fully autonomous rides, a 212% increase year-over-year [2][6] - **Geographic Expansion**: Coverage in 22 cities, with international expansion into Switzerland, Abu Dhabi, Dubai, and Hong Kong [6][18] Ernie Model Development - **Ernie 5.0 Launch**: Introduced the first native full-language model, excelling in language understanding and creative writing [7][15] - **Impact on Business**: Supports various applications, enhancing enterprise efficiency across sectors [15] Agent and Digital Human Technology - **Revenue Growth**: Generated 2.8 billion RMB, a 262% increase year-over-year [8][14] - **Applications**: Utilized in healthcare, automotive, and legal services, improving lead conversion rates for advertisers [8] Mobile Ecosystem Innovations - **AI-Generated Content**: Approximately 70% of mobile search results feature AI-generated content, enhancing user experience [12][19] - **User Engagement**: Increased user retention and interaction time, with a 6% rise in AI search queries [12][19] Future Outlook AI Business Growth - **Investor Transparency**: New classification views for AI applications to enhance visibility of underlying value drivers [17][18] - **Sustainable Revenue Growth**: Anticipated increase in revenue and profit margins from AI applications and services [17][18] Shareholder Returns - **Stock Buyback**: $2.3 billion worth of stock repurchased in 2023, with plans for future buybacks and potential dividend policies [4][21] Capital Expenditure - **Investment in AI**: Over 10 billion RMB invested in AI technologies, with plans for continued investment to enhance operational efficiency [20][21] Conclusion - **Overall Assessment**: Despite a decline in total revenue, Baidu's strategic focus on AI and cloud services, along with significant growth in autonomous driving and digital human technologies, positions the company for future growth and innovation in the rapidly evolving tech landscape [2][4][18]

Core Insights - Jinglin Asset's overall holdings have increased to $4.44 billion, a significant rise of over 50% compared to the previous quarter, with a total of 30 stocks in its portfolio [1][2] Holdings Overview - The top ten holdings of Jinglin Asset include META, NetEase, NVIDIA, Pinduoduo, Google-A, Manbang, Futu Holdings, Alibaba, Nebius Group, and Qifu Technology, with Chinese stocks making up a substantial portion [2] - The firm has expressed strong optimism about China's future development, citing competitive advantages in cost efficiency, talent reserves, and industry chain collaboration [2] Investment Strategy - Jinglin Asset has notably increased its investment in technology stocks, with a 277% increase in NVIDIA holdings, raising its market value to $444 million, making it the third-largest holding [2] - The firm has also increased its stake in Google-A by 852% [2] - Alibaba saw an 81% increase in holdings, with 843,400 shares valued at $151 million, driven by strong revenue growth and a significant rise in stock price [3] Additional Investments - The firm has made substantial investments in Atour, increasing its holdings by nearly tenfold, and has also built a significant position in WeRide, with 2.68 million shares [3] - WeRide is recognized as a leading player in China's autonomous driving sector, with a notable revenue increase in its Robotaxi business [3] Reductions in Holdings - Jinglin Asset has significantly reduced its holdings in Futu Holdings, Beike, NetEase, and Intel, with Futu Holdings seeing a 34.94% decrease [4] - The firm has completely divested from seven stocks, including Microsoft and Daqo New Energy [4]