February 10, 2026 Forward Looking Statements and Risk Factors This document contains "forward-looking statements" as defined in the Private Securities Litigation Reform Act of 1995. When Welltower uses words such as "may," "will," "intend," "should," "believe," "expect," "anticipate," "project," "pro forma," "estimate" or similar expressions that do not relate solely to historical matters, Welltower is making forward-looking statements. These statements include, among others, management's expectations regar ...

This presentation contains declarations that constitute forward-looking statements regarding the intent, belief or current expectations of value creation, customer base dynamics, estimates regarding future financial results and other aspects of the activities. Classificado como Uso Interno Disclaimer Analysts and investors are cautioned not to place undue reliance on those forward-looking statements, which speak only as of the date of this presentation. TIM S.A. undertakes no obligation to release publicly ...

Core Insights - Spotify is enhancing its platform through a focus on artificial intelligence and user engagement to improve user experience and drive growth [1] - The company competes with other streaming services such as Apple Music and Amazon Music [1] Stock Performance - Evercore ISI set a price target of $700 for Spotify, suggesting a potential increase of about 47% from its trading price of $476.13 [2][5] - The stock has experienced a 14.77% rise, indicating positive market sentiment [2][5] - Spotify's stock has fluctuated between $462.78 and $495.94, with a yearly high of $785 and a low of $405, reflecting investor reactions to its strategic focus [3] Market Capitalization and Trading Volume - Spotify's market capitalization is approximately $98 billion, highlighting its significant presence in the music streaming industry [3][5] - Today's trading volume for Spotify is 10.1 million shares, indicating strong investor interest [4]

Broadcom could be the biggest winner from Alphabet's surge in capex spending.Alphabet (GOOGL 1.77%) (GOOG 1.79%) shocked investors when the company set a capital expenditure (capex) budget of between $175 billion and $185 billion for 2026 when it reported its fourth-quarter results. That's a massive increase from the $91 billion in capex it spent in 2025.About 60% of its spending will go toward servers and about 40% toward longer-duration assets such as data centers and networking equipment. Within servers, ...

Core Viewpoint - Alphabet, Google's parent company, is entering the bond market with a 100-year bond, raising approximately $32 billion across multiple bond deals in various currencies, with a small portion of £1 billion dedicated to the century bond [1][2]. Group 1: Bond Characteristics - Century bonds are long-term debt instruments, with Alphabet's bond maturing in 2126, appealing to investors like pension funds and insurers who seek steady long-term payments [1][4]. - The bond market operates on the principle that bond prices and yields move inversely; as bond prices increase, yields decrease, and vice versa [8][9]. Group 2: Rationale for Borrowing - Despite having substantial cash reserves, Alphabet is opting to borrow due to the high costs associated with AI infrastructure development, taking advantage of historically low interest rates for flexible funding [5]. - Historical context shows that century bonds are not new, with previous issuances by companies like Coca-Cola and Motorola, indicating a trend in long-term financing strategies [6]. Group 3: Market Implications - Future observations should focus on demand for similar long-term tech bond offerings, pricing dynamics, and the narrative surrounding AI's transition from buildout to revenue generation, which could impact both stock and bond markets [10][11]. - The historical context of tech companies during the dot-com boom suggests that while cash-rich firms may not capture significant upside, they can experience prolonged downturns, a consideration for Alphabet's strategy [12].

Astera Labs reported fourth-quarter earnings that beat on top and bottom lines, but shares plunged as much as 10% as revenue came short of some analysts' expectations. Here's how the artificial intelligence infrastructure and semiconductor company did compared to LSEG consensus estimates:Earnings per share: 58 cents vs. 51 cents expectedRevenue: $271 million vs. $250 million expectedRevenue in the fourth quarter rose 92% year over year. However, some analysts were expecting the company to report revenue in ...

离春节不到一周，过去两年的经验告诉我们，国产AI新产品的爆发期又要来了。 Kimi K2.5在OpenRouter上一骑绝尘，Minimax也在官网上隐隐约约透露出M2.2的版本代号。 大家等得很焦急，智谱和DeepSeek年前是不是也得来点大动作？ 这不，短短48小时内，三个独立的技术平台接连出现了与智谱新模型GLM-5相关的线索，形成了一条 完整的信息链。 01 线索浮现：三平台齐曝 GLM-5 踪迹 2月7日，OpenRouter平台悄悄上线了一款代号为"pony-alpha"的模型。 根据实际测试的结果，它的思维链风格与智谱的GLM系列高度吻合，比如： 遇到常规回答，它会以"嗯，......"为开头进行思考； 遇到知识检索任务，就分点列出"1.分析请求：..."； 遇到代码任务，则会明确标注"用户要求......"。 而技术社区在把这个模型投入实际开发后，发现它在贪吃蛇、Minecraft等复杂的代码生成任务中表现相 当稳健，但缺乏图像等多模态输入支持。 接下来的2月9日，vLLM推理框架仓库出现了编号为34124的pull request，在代码中首次明确出现 了"GLM-5"标识。 更重要的是，代 ...

长江商报消息上交所再融资新政发布首日，中科曙光（603019.SH）抛出一份巨额融资计划。 2月9日晚间，中科曙光公告称，拟向不特定对象发行可转债募资不超过80亿元，用于面向人工智能的先 进算力集群系统项目、下一代高性能AI训推一体机项目、国产化先进存储系统项目。 长江商报记者注意到，中科曙光为国内算力龙头。2025年11月，其发布ScaleX 640超节点，实现单机柜 640卡互联，采用一拖二高密架构与静默式相变液冷技术，可提升大模型训练性能30%。 中科曙光在研发方面的投入力度领先。2020年—2024年及2025年前三季度，公司近六年累计投入研发费 用64.05亿元。 拟高位发80亿可转债 2月9日，沪深北交易所宣布优化再融资一揽子措施，主要涉及大力支持优质上市公司再融资、提高对创 新科技型企业的制度包容性、提高再融资制度便利性以及强化再融资全链条监管四方面内容。支持优质 上市公司将募集资金用于与主营业务有协同整合效应的新产业、新业态、新技术领域，投向第二增长曲 线业务，同时严防盲目跨界投资、多元化投资。 当日晚间，中科曙光公告称，拟向不特定对象发行可转债募资不超过80亿元，用于面向人工智能的先进 算力 ...

半导体 分析师：邹臣 登记编码：S0730523100001 zouchen@ccnew.com 021-50581991 海外云厂商 26 年资本支出再加速，半 导体产业链迎来全面涨价潮 ——半导体行业月报 证券研究报告-行业月报 强于大市(维持) 联系人：李智 投资要点： AI 驱动半导体周期持续上行，国内主要晶圆厂持续满载运行，部 分工艺节点的价格已经上涨；AI 服务器需求的爆发式增长使得封 测产能出现严重短缺，封测厂商也已开始调涨价格；由于上游成 本的上涨，国科微、中微半导、英集芯、富满微、英飞凌等厂商 陆续发布涨价函；由于 AI 推理需求的激增，拉动推理服务器的需 本报告版权属于中原证券股份有限公司 www.ccnew.com 请阅读最后一页各项声明 第 1 页 / 共 50 页 发布日期：2026 年 02 月 11 日 资料来源：聚源，中原证券研究所 -13% -2% 10% 21% 32% 44% 55% 66% 2025.02 2025.06 2025.10 2026.02 半导体 沪深300 相关报告 电话： 0371-65585629 地址： 郑州郑东新区商务外环路 10号 18楼 地址 ...

财通证券认为，临近春节，国产模型进入发布窗口期，国产模型快速迭代抢占用户AI交互入口，交互 方式也正在被塑造，对应推理侧算力需求有望迎来加速提升。同时2026年也是推理侧国产超节点上量元 年，目前已有大量国产厂商发布新一代超节点方案，华为Atlas 950/960，搭载8192/15488张算力卡，曙 光scale x 640、沐曦、昆仑芯、阿里磐久等均有超节点布局。供需两侧双向奔赴，产业链即将迎来放量 时点。 公司方面，据华鑫证券、华创证券表示， 据摩根大通在报告中预测：中国的AI推理Token消耗量预计将从2025年的约10千万亿，增长至2030年的 约3,900千万亿，五年间增长约370倍。 *免责声明：文章内容仅供参考，不构成投资建议 爆发式增长主要由两个因素驱动：一是AI在消费者和企业工作负载中的渗透率提高；二是应用场景从 简单的对话AI扩展到复杂的智能体和多模态输出。数据显示，截至2025年12月，字节跳动豆包系列模 型的日均Token消耗量已突破50万亿，同比增长超10倍。 *风险提示：股市有风险，入市需谨慎 协创数据：公司升级算力服务平台，通过异构算力纳管与用户租赁来实现不同类型算力资源的统一 ...