大规模量产 触发此轮行情的直接导火索是Neuralink的最新动态。当地时间1月1日，马斯克在社交媒体宣布， Neuralink将于2026年启动脑机接口设备的大规模生产，并推进几乎完全自动化的手术方案。与此同时， OpenAI首席执行官萨姆·阿尔特曼支持的Merge Labs正分拆独立，主攻利用超声波读取脑部活动的新技 术路线。 两大科技巨头的动作表明，脑机接口正从单一的科学实验走向多元化的商业竞速。脑机接口(Brain- Machine Interface，BMI，也称BCI)是指在人或动物大脑与外部设备之间创建的直接连接，实现脑与设 备的信息交换。Neuralink是马斯克于2016年创立的脑机接口公司，致力于开发高带宽、低侵入性的脑机 接口技术，旨在帮助治疗神经系统疾病并最终实现人类与AI的共生。 马斯克明确"2026年开始大规模生产"脑机接口设备后，脑机接口相关概念股热度快速攀升，反映出市场 对行业商业化的强烈预期。该公司于2024年1月28日完成了首例脑机接口设备人体移植。截至2025年9 月，全球已有12人植入Neuralink设备，累计使用时间达2000天，总使用时长超过1.5万小时。 该公司 ...

Core Insights - AI is expected to enhance global trade in the long term by reducing trade costs and increasing productivity, with WTO projecting a growth in global trade volume by 33.7% to 36.7% from 2025 to 2040, driven by lower trade costs, improved AI services, and productivity in tradable sectors [1] Group 1: AI-Related Product Exports - Global AI-related product exports are projected to reach $3.1 trillion in 2024, a 10.1% increase from 2023, accounting for 14.5% of total exports, up 1.2 percentage points [2] - The top three regions for AI-related product exports are Mainland China ($633.6 billion), Hong Kong ($355.4 billion), and Taiwan ($310.6 billion), with Mainland China's share of total exports at 17.7% [2] - The export structure from Mainland China is balanced between intermediate goods (68%) and equipment (31%), compared to other major economies [2] Group 2: AI-Related Product Imports - The U.S. has seen a significant increase in AI-related product imports, with a year-on-year growth of 27% in August 2025, driven by a surge in private sector investment in information processing equipment [3] - AI-related products accounted for 20.7% of total U.S. imports in August 2025, an increase of 5.5 percentage points from the previous year, with a notable rise in the share of AI-related equipment imports [3] Group 3: China's AI-Related Product Export Trends - From December 2024 to November 2025, Mainland China's AI-related product exports are expected to reach $690.3 billion, a 9.7% year-on-year increase, although still below the levels seen in July 2022 [4] - The contribution of AI-related products to China's monthly export growth is projected to be approximately 1.7 percentage points in 2025, up from 1.4 percentage points in 2024, with a contribution rate of 32% [4] - Direct trade between China and the U.S. in AI-related products is relatively low, with only 8% of U.S. imports coming from Mainland China [4] Group 4: AI-Related Services Exports - AI-related services, which include telecommunications, computer, and information services, are also experiencing growth, with exports reaching $70.8 billion by October 2025, a 13.5% increase year-on-year [5] - China is positioned to gain competitive advantages in AI-related services in developing economies, with 81% of low-income economies using open-source AI models from China [5]

Core Viewpoint - The article discusses the impact of AI on global trade, highlighting its potential to reduce trade costs and enhance productivity, which is expected to drive trade growth significantly from 2025 to 2040, with global trade volume projected to increase by 33.7% to 36.7% [2] Group 1: AI-Related Product Exports - Global AI-related product exports are projected to reach $3.1 trillion in 2024, a 10.1% increase from 2023, accounting for 14.5% of total exports, up by 1.2 percentage points [3] - The top three regions for AI-related product exports are Mainland China ($633.6 billion), Hong Kong ($355.4 billion), and Taiwan ($310.6 billion), with Mainland China's share of total exports at 17.7%, relatively low among major economies [3] - The export structure from Mainland China is balanced between intermediate goods (68%) and equipment (31%), although intermediate goods remain dominant [3] Group 2: AI-Related Product Imports - The United States has seen a significant increase in AI-related product imports, driven by a surge in private sector investment in information processing equipment, with a year-on-year increase of 26% in Q2 2025 [4] - By August 2025, AI-related product imports in the U.S. grew by 27% year-on-year, with these products constituting 20.7% of total imports, an increase of 5.5 percentage points from the previous year [4] Group 3: China's AI-Related Product Export Trends - From December 2024 to November 2025, Mainland China's AI-related product exports are expected to reach $690.3 billion, a 9.7% increase year-on-year, although still below the levels seen in July 2022 [5] - The contribution of AI-related products to China's monthly export growth is modest, with a year-on-year increase of approximately 1.7 percentage points in early 2025, compared to 1.4 percentage points in 2024 [5] - In comparison to other economies, AI-related products have driven nearly all export growth in Taiwan, Malaysia, Singapore, and Thailand, while countries like South Africa and Brazil have seen minimal impact [5] Group 4: AI-Related Services Exports - AI-related services, which include telecommunications, computer, and information services, are also experiencing accelerated growth, with exports reaching $70.8 billion by October 2025, a 13.5% increase year-on-year, surpassing the growth rate of AI-related products [6] - China is positioned to gain a competitive advantage in AI-related services in developing economies due to lower unit costs, with 81% of low-income economies using open-source AI models from China [6]

Core Viewpoint - The AI hype is currently in the early stages of a bubble, with expectations that the U.S. stock market will significantly underperform non-U.S. stocks and gold assets by 2025 [1][3]. Group 1: Market Performance - Major U.S. stock indices are expected to record double-digit gains in 2025, marking the third consecutive year of growth, similar to the trend observed from 2019 to 2021 [3]. - Last year, gold prices surged over 60%, and emerging markets had a strong performance, with the FTSE 100 index outperforming major global markets [3]. Group 2: Investor Behavior - Investors are increasingly favoring non-U.S. stocks and bonds over U.S. equities and dollar cash, indicating a shift in investment preferences [3][4]. - Concerns about a potential AI bubble have led to market volatility and increased risk aversion among investors [3]. Group 3: Federal Reserve Policy - The newly appointed Federal Reserve Chair and the Federal Open Market Committee (FOMC) are likely to favor lowering both nominal and real interest rates, which could support asset prices but also exacerbate the bubble [4]. - Analysts suggest that as concerns about the AI bubble grow, investors will actively seek undervalued investment opportunities in the financial markets [4]. Group 4: Bridgewater Performance - Bridgewater's flagship fund, Pure Alpha II, achieved a record return of 34%, marking a significant recovery from previous years of low returns [4]. - The Bridgewater China Macro Fund and All Weather Strategy Fund also reported returns of 34% and 20%, respectively [5].

2025 年香港经济表现超预期，全年 GDP 增速有望达 3.2%，连续第三年增长且基本恢复至疫情前水平，前三季度同比增长 3.3%，呈现 "加速上行" 态势，主 要得益于 "物流（G）" 和 "资金流（C）金融项" 的强劲拉动。 从增长动能来看，依据 "HGC 三元流动模型"，资本市场交投活跃推动 "资金流（C）金融项" 拉动 GDP 增长约 1.2 个百分点；美企 "抢进口" 效应及 AI 相关 产品需求带动，"物流（G）" 贡献 1.1 个百分点的增长；"人流（H）" 相关产业正向拉动 0.35 个百分点，但 "资金流（C）地产项" 因建造活动疲弱拖累增长 超 0.3 个百分点。具体而言，2025 年访港旅客量同比增约 12%，零售业自 5 月起恢复正增长；进出口货值同比分别升 13.8%、13.6%，机场客货运量稳步回 升；恒指累计上涨 28.9%，港股 IPO 募资额重回全球榜首，银行保险业景气度提升，但地产市场供需分化，建造业表现疲弱。 展望 2026 年，偏乐观情形下香港 GDP 有望增长 3.0%，"物流（G）" 和 "资金流（C）" 仍将是核心引擎，"人流（H）" 贡献保持稳健。利好因素包括 ...

对冲基金经理雷伊・达利欧发文警告，曾推动华尔街科技股走高的人工智能（AI）热潮目前已进入泡 沫初期阶段。他指出，2025年美国股市表现显著落后于非美股市及黄金资产。"显然，相较于美国股 票，投资者更愿意配置非美股票；同理，他们也更倾向于持有非美债券，而非美国债券与美元现 金。""当然，未来美联储的政策走向以及生产力增速，都存在巨大的不确定性。" 达利欧表示。"种种 迹象表明，新任美联储主席及联邦公开市场委员会，大概率会倾向于压低名义利率与实际利率。这一举 措虽会对资产价格形成支撑，但也会进一步吹大泡沫。" 市场有风险，投资需谨慎。本文不构成个人投资建议，也未考虑到个别用户特殊的投资目标、财务状况或需要。用户应考虑本文中的任何 意见、观点或结论是否符合其特定状况。据此投资，责任自负。 风险提示及免责条款 ...

Core Viewpoint - The article discusses the optimistic outlook for China's future in mathematics, predicting that it will become a mathematical powerhouse within the next 5 to 10 years, as stated by the first Chinese Fields Medalist, Shing-Tung Yau, during the 10th International Conference of Chinese Mathematicians (ICCM) held in Shanghai [3][4]. Group 1: Talent Development - Yau emphasizes the importance of nurturing mathematical talent, noting that Chinese students around the age of 12 exhibit exceptional creativity and intuition comparable to their peers globally [7]. - He expresses concern that the current educational focus on standardized testing and rote learning ("刷题") stifles innovation and critical thinking among students as they progress into high school [6][8]. - The goal of specialized programs is not only to produce mathematicians but to provide a solid mathematical foundation for various scientific disciplines, allowing students to pursue their interests later on [7]. Group 2: AI and Mathematics - Yau argues that while AI enhances research efficiency, it does not revolutionize scientific paradigms, and breakthroughs in AI will ultimately require a return to mathematical fundamentals [6][10]. - He refutes the notion that AI will replace mathematicians, asserting that significant mathematical problems require years of deep contemplation and cannot be solved merely by searching existing knowledge [10]. - Yau highlights that the current reliance of AI on vast computational power reflects a lag in foundational mathematical theories, suggesting that advancements in AI will necessitate a deeper mathematical understanding [10].

"ICCM二十八年前大会刚创办时，中国数学家涉猎的领域还很狭窄。如今，数论、代数、几何等各个方 向都已呈现出多姿多彩的局面。"会议期间，菲尔兹奖首位华人得主、大会主席丘成桐接受第一财经在 内的媒体采访时给出一个判断：未来5到10年，中国将成为数学强国。 未来AI若想突破能耗与算力的天花板，必须回归数学。 距离2026年菲尔兹奖公布时间还有一个多月时间，为期5天的第十届世界华人数学家大会（ICCM）3日 在上海开幕。 除了乐观的判断，丘成桐对于数学人才的培养和人工智能（AI）的发展也有冷静的思考。 在他看来，AI改变的仅仅是科研效率，而非范式革命，AI的突破仍需要回归数学。而在人才培养方 面，不能在标准化的"刷题"与功利化的KPI中让年轻一代失去创新能力。在第十届世界华人数学家大会 "这个群体数量巨大，如果其中有三分之一能最终成才，就是国家了不起的财富。"丘成桐说，少年班的 核心目标并非只培养数学家，而是为理科学习打好数学基础，后续他们可根据兴趣选择不同学术方向。 然而，这个群体12岁之后的成长期，特别是一旦进入高中阶段，一些原本敢想敢问的学生开始变得不敢 提出试卷以外的问题，思维逐渐趋同。这种对不确定性的规避 ...

Group 1 - The Nikkei average index broke through the 51,000-point mark, which was previously seen as a resistance level, closing at 51,832 points on January 5, 2026, with an increase of 1,493 points (3%) from the previous trading day [2][4] - The market sentiment is optimistic for the 2026 stock market, driven by strong demand for semiconductors, particularly in AI-related stocks, with SoftBank Group (SBG) seeing a peak increase of 7% during the trading session [4] - Tokyo Electric Power Holdings surged by 9% due to plans for large-scale investments to meet the rising electricity demand driven by AI proliferation, enhancing its presence as an AI-related stock [4] Group 2 - The semiconductor sector is experiencing increased buying interest, with companies like Advantest, Tokyo Electron, and Lasertec seeing significant demand [4] - Analysts suggest that overseas capital inflows are contributing to the rise in semiconductor stocks, indicating a broader trend of investment in this sector [4]

Group 1: Market Overview - The global stock market is experiencing a surge driven by artificial intelligence (AI) enthusiasm, but inflation risks may threaten this growth [1] - Major tech companies contributed to half of the gains in the US stock market last year, with significant increases in stock indices due to AI and monetary easing expectations [1] - Wall Street anticipates that government stimulus and AI prosperity will inject new growth into the global economy in 2026 [1] Group 2: Inflation Concerns - Fund managers are preparing for a potential resurgence of inflation, as economic growth from AI may lead central banks to end the interest rate cut cycle [1][2] - Tightening monetary policy could reduce investor interest in speculative tech stocks, increase financing costs for AI projects, and cut into tech companies' profits and stock prices [2] - Analysts predict that inflation rates will remain above the Federal Reserve's 2% target due to substantial corporate investments in AI [2] Group 3: Cost Pressures - Rising costs associated with chip and energy consumption are expected to contribute to inflation, as major tech firms invest heavily in new data centers [2][6] - Oracle's stock dropped due to rising expenditure, while Broadcom warned of profit margin pressures, indicating early signs of market tension regarding cost increases [4] - HP anticipates experiencing price and profit pressures in the latter half of 2026 due to increased demand for storage chips driven by data center needs [4] Group 4: Investment Strategies - Investment firms are increasingly concerned about inflation risks, prompting some to shift towards inflation-protected bonds [5] - The potential for rising interest rates may lead to a decrease in the price-to-earnings ratios for large AI stocks [5] - Deutsche Bank forecasts that capital expenditures for AI data centers could reach $4 trillion by 2030, raising concerns about supply bottlenecks and spiraling investment costs [6]