Core Viewpoint - NVIDIA shares have experienced a modest rebound, increasing approximately 2% in Monday morning trading after a challenging start to 2026 [1] Company Summary - NVIDIA's stock performance indicates a recovery phase, suggesting potential investor confidence following recent volatility [1]

作者｜冬梅 北京时间 2026 年 3 月 17 日凌晨两点半，当英伟达 CEO 黄仁勋穿着那件标志性的黑色皮衣踏上 SAP 中心的舞台时，台下近万名开发者心里清楚：这一次，老黄要讲的不是某个单一芯片，而是一 整套 AI"全家桶"。 这场演讲之前，市场早已躁动不安。黄仁勋曾在 2 月预告将发布一款"前所未见的芯片"，被普遍认为 是采用台积电 1.6nm 制程、引入光通信技术的下一代 Feynman 架构。 今天揭晓的 Feynman 架构、Vera Rubin 平台的量产进展，以及面向企业级自主代理的开源平台 NemoClaw，不过是这座基础设施落地所需的"钢筋水泥"。黄仁勋用两个小时向资本市场证明了一件 事：英伟达早已不是那个卖显卡的公司，现在的英伟达是一家为"数万亿美元 AI 基建时代" 搭建完整 技术栈的"总包工头"。 回顾 CUDA 20 年：安装基数引爆飞轮，GPU 算力成本持续下降 演讲刚开始，黄仁勋用近十分钟篇幅，回溯了 CUDA 架构诞生 20 年的演进历程。他将这套软件生 态定义为英伟达一切业务的"中心"，并直言：真正难以复制的壁垒，是底层的安装基数。 "二十年来，我们一直致力于这种革命性架 ...

This panel brings together leaders from AITEM, CytoReason, Superluminal Medicines, and Therapyside, showcasing how NVIDIA inference microservices (NIMs), AI blueprints, and BioNeMo are driving advances in diagnostics, therapy workflows, and drug development. Through a technical and forward-looking discussion, we’ll explore how these innovators are integrating AI into their systems, overcoming challenges, and shaping a future where technology fundamentally transforms patient care and advances the potential f ...

Core Insights - Nvidia and Eli Lilly announced a collaboration to invest up to $1 billion over the next five years to establish a research lab focused on AI-assisted drug discovery models [2] - The partnership aims to leverage Nvidia's BioNeMo software platform and Vera Rubin accelerator to develop necessary infrastructure, talent, and computational resources for biological and chemical modeling [2] Group 1: Collaboration Details - The joint innovation lab will be located in the San Francisco Bay Area, bringing together top biologists and chemists from Eli Lilly with Nvidia's software engineers and model developers [2] - Eli Lilly's CEO, David Ricks, emphasized that combining their extensive data and scientific knowledge with Nvidia's computational power could revolutionize drug discovery [2] Group 2: Technology Platforms - BioNeMo is an open-source framework launched by Nvidia in the fall of 2022, specifically designed for building and training deep learning models for drug discovery [4] - The Vera Rubin computing platform, recently unveiled at CES, promises five times the performance of Nvidia's previous Blackwell GPU, providing essential computational power for training new foundational models [3][4] Group 3: Broader Applications - The lab's focus extends beyond AI drug discovery; researchers will also explore AI applications in clinical development, manufacturing, and commercial operations [3] - Eli Lilly is investigating Nvidia's Omniverse robotics platform to optimize manufacturing facilities and increase production of high-demand drugs [3]

Group 1: Data Center Cooling Stocks - Data center cooling stocks experienced a significant decline, with companies like Johnson Controls and Trane Technologies dropping by 10%, Modine falling by 20%, and Carrier and Vertiv Technologies decreasing by over 5% [4] - The drop was influenced by Jensen Huang's comments at CES regarding new server racks using the Rubin chip, which can operate without water cooling, requiring similar airflow to racks using the Blackwell chip [4] Group 2: Semiconductor Stocks - Semiconductor stocks saw a broad increase, with Microchip Technology rising by 11.65%, reaching its highest level since July, NXP Semiconductors increasing by nearly 10%, and TSMC gaining over 1.5% [4] - SanDisk experienced a significant surge of 27.56%, marking its largest intraday gain in ten months, while Micron Technology's stock rose by 10.02%, reaching a historical high of $343.43 per share, with a total market capitalization of $386.5 billion [4] Group 3: Chinese Concept Stocks - The Nasdaq Golden Dragon China Index fell by 0.78%, with mixed performance among popular Chinese concept stocks [5] - Companies like Hesai surged over 10%, Pony.ai increased by over 5%, and Huazhu, UMC, and others rose by over 4%, while BOSS Zhipin dropped over 6%, Tencent Music fell by over 4%, and Alibaba decreased by over 3% [5] Group 4: Precious Metals - Spot gold increased by 1.07%, reaching $4,496.66 per ounce, while spot silver rose by 6.15%, reaching $81.32 per ounce [6]

Group 1 - Data center cooling stocks experienced a significant decline, with companies like Johnson Controls and Trane Technologies dropping by 10%, and Modine falling by 20% following comments from Jensen Huang regarding new cooling technologies [2][3] - Huang stated that servers equipped with the new Rubin chip could be cooled without water cooling systems, requiring similar airflow to those using the Blackwell chip [2] - The overall stock market showed mixed results, with the Dow Jones Industrial Average up by 0.15%, the S&P 500 up by 0.31%, and the Nasdaq Composite up by 0.38% [1][2] Group 2 - Nvidia's core mission is to build a full-stack AI platform, enabling participation in the AI revolution across various sectors [3] - Nvidia has released a comprehensive toolchain, including NeMo for large language models and Clara NeMo for medical AI, supporting the entire lifecycle from data processing to deployment [3] - Semiconductor stocks saw a rise, with Microchip Technology increasing by 7.7%, reaching its highest level since July, and other companies like NXP and TSMC rising over 2% [3]

Group 1: AI Drug Development Industry - SandboxAQ, an AI startup supported by Alphabet and Nvidia, released a large-scale synthetic dataset to accelerate global drug development by simulating interactions between drug molecules and proteins [2] - Major pharmaceutical companies like Merck, Pfizer, Eli Lilly, and BMS have invested hundreds of billions in AI drug-related companies, with over $50 billion in significant transactions occurring in the last five years [3] - The demand for AI-assisted drug development from traditional pharmaceutical giants is increasing, as evidenced by large orders and collaborations with AI drug companies [3] Group 2: Company Updates - Crystal Holding (02228) expects a significant increase in revenue, projecting at least RMB 500 million for the first half of 2025, a year-on-year increase of approximately 387%, marking its first half-year profit [4] - The revenue growth for Crystal Holding is attributed to its collaboration with DoveTree Medicines, which has led to a substantial contribution from a $51 million upfront payment [5] - Viya Bio (01873) has been involved in AI drug development for five years, with AI-related orders now accounting for 12% of new orders, indicating a growing trend [5]

Group 1: Industry Overview - SandboxAQ, an AI startup supported by Alphabet and Nvidia, launched a large-scale synthetic dataset to accelerate global drug development by simulating drug molecule and protein interactions [1] - Major pharmaceutical companies like Merck, Pfizer, Eli Lilly, and BMS have invested hundreds of billions in AI drug-related companies, indicating a strong demand for AI-assisted drug development [1] - The total value of significant AI drug development transactions has exceeded $50 billion in the last five years, highlighting the growing interest and investment in this sector [1] Group 2: Company Developments - Crystal Tech Holdings expects to achieve a consolidated revenue of at least RMB 500 million by mid-2025, a year-on-year increase of approximately 387%, driven by a partnership with DoveTree Medicines [3] - Via Biotechnology has seen AI-related orders account for 12% of new signed orders, indicating a growing trend in AI integration within their drug development processes [3] - Via is transitioning from a focus on computational methods to AI-driven drug design, aiming to change the paradigm of drug design [3] Group 3: Market Trends - AI pharmaceutical companies are increasingly securing large orders from multinational corporations, validating the profitability of AI drug development through technology licensing and revenue-sharing models [2] - The first approved AI-driven drug is anticipated to be a significant milestone, with both AI companies and traditional pharmaceutical leaders competing for breakthroughs in this new field [2] - The entry of various players, including cross-industry companies, into the AI drug development space is expected to enhance competitive barriers over time [2]

Core Insights - The AI-driven pharmaceutical industry is on the verge of a breakthrough, with the first AI-developed drug expected to be approved soon, making it a focal point for investment opportunities [1][2] - The entry of both traditional pharmaceutical companies and new tech firms into the AI drug development space is creating a competitive landscape, emphasizing the importance of robust pipelines and proven capabilities [1][2] Timing - The application of AI in pharmaceuticals has transitioned from concept to reality, with multi-omics development expected to enhance efficiency by 1000 times, indicating a significant shift in the industry [2] - The timeline for AI's impact began with the Turing Test in 1950, gaining momentum with DeepMind's AlphaGo in 2016, and is expected to culminate in notable achievements like the Nobel Prize for breakthroughs in protein structure prediction by 2024 [2] - The EU's AI Act is set to eliminate AI drug discovery systems that rely on opaque models, ensuring that only verifiable and replicable AI models remain in the market [2] Essence - Cloud computing and data limitations are being overcome, leading to a resurgence in innovation and faster returns on research and development [3] - Major tech companies like Amazon, Google, and Microsoft provide ample cloud computing resources for pharmaceutical companies, enhancing their AI capabilities [3] - The shift from deep learning to federated learning is breaking down data silos, allowing for better model development and collaboration across organizations [3] - The development of generative AI models is crucial for maintaining competitive advantages, as efficiency in model iteration and training experience becomes increasingly important [3] Industry Dynamics - Tech giants are actively entering the AI pharmaceutical space, with significant investments and developments from companies like NVIDIA and Google [4] - Pharmaceutical companies are also ramping up their AI initiatives, with major players like Merck, Pfizer, and Eli Lilly investing hundreds of billions into AI-related ventures [4] - Recent data indicates that over $50 billion has been invested in AI drug development projects in the last five years, highlighting the growing importance of AI in the pharmaceutical industry [4]

Investment Rating - The report suggests a strong investment outlook for the AI pharmaceutical industry, highlighting the imminent approval of the first AI-driven drug as a pivotal moment for investment opportunities [5]. Core Insights - The application of AI in drug development is transitioning from concept to reality, with multi-omics development expected to reduce costs and increase efficiency by 1000 times, marking the dawn of a new era in innovative drug development [12][16]. - The report emphasizes the importance of computational power, data quality, and model development as critical factors driving the success of AI in pharmaceuticals [3][48]. - Major technology companies are entering the AI pharmaceutical space, with significant investments from top pharmaceutical firms, indicating a robust shift in the industry landscape [4][5]. Summary by Sections AI Application and Industry Transformation - AI applications in pharmaceuticals are moving towards practical implementation, with significant milestones such as the success of AlphaFold in protein structure prediction [12]. - The report notes that the AI pharmaceutical sector is on the verge of a transformative phase, driven by advancements in multi-omics applications [16]. Computational Power and Data Utilization - The availability of cloud computing resources from major tech companies like Amazon and Google is enhancing the computational capabilities necessary for AI applications in drug development [3]. - Innovations in federated learning are breaking down data silos, allowing for better data sharing while maintaining privacy, which is crucial for AI model training [37][39]. Industry Dynamics and Major Players - The entry of tech giants like NVIDIA and Google into the AI pharmaceutical space is reshaping the industry, with substantial investments in AI drug development [4]. - Leading pharmaceutical companies are also heavily investing in AI-related initiatives, with over $50 billion in significant transactions occurring in the past five years [4]. Investment Strategies - The report recommends focusing on companies with rich pipelines and strong validation capabilities, such as Insilico Medicine and Crystal Holding, as they are poised to benefit from the upcoming breakthroughs in AI drug approvals [5]. - It also suggests monitoring traditional pharmaceutical companies that are making significant strides in AI, such as CSPC Pharmaceutical Group and Fosun Pharma, for potential high returns [5].