Anthropic 把销冠脑子里10年经验，一键变成全公司用的AI工具。 就在刚刚，Anthropic 用一个插件系统，重新解锁了 AI Agent 的新玩法，也将再次「杀死」一批 AI 创企。 2 月 24 日，Anthropic 更新了 Claude Cowork 插件系统，用户可以从零开始定制化打造 AI 插件，将 Claude 能力直接以工具包的方式融入工作流中，打开 整个企业级定制化插件市场。 不仅如此，Anthropic 还同步配置了插件创造、使用以及管理的工具平台。Claude 会通过提问，引导用户定制技能、设置相关命令并接入 MCP 协议。所有 新增的插件功能都可在新增的自定义菜单中查看、管理。团队和企业管理员还可以直接访问公司配置的插件以及 MCP 控制功能。 目前，Claude 可以接入企业已使用的工具，例如 Slack、Salesforce 和 Excel，Claude 的插件可以在 Cowork 以及任何基于 Claude Agent SDK 构建的系统中 运行。所有插件的用户体验更新均面向所有 Cowork 用户开放。 这次，Anthropic 还一口气推出 10 个横跨 HR、运营 ...

Core Insights - The core focus of major internet companies during the Spring Festival was on user acquisition, marking a significant shift towards AI application penetration in the market [1][2][3] User Acquisition and Engagement - Major companies like Tencent and ByteDance reported substantial increases in daily active users (DAU) during the Spring Festival, with Tencent's Yuanbao surpassing 50 million DAU and ByteDance's Doubao achieving 1.14 billion monthly active users [2] - The competition among AI applications intensified, with Doubao leading the App Store rankings, followed by Qianwen and Ant Financial's AI assistant [2][4] Market Dynamics and Competition - The Spring Festival saw a fierce competition primarily between ByteDance and Alibaba's AI assistants, with Qianwen experiencing a 727.7% increase in DAU on the first day of the red envelope campaign [4][5] - The rankings of AI applications fluctuated significantly during the campaign, indicating a dynamic market landscape [4][5] Long-term User Retention Challenges - Despite the initial success in user acquisition, industry experts highlight that user retention remains a critical challenge, with concerns about the sustainability of growth post-subsidy [6][11] - Companies are focusing on enhancing user experience and feedback rather than solely on DAU metrics, indicating a shift towards long-term engagement strategies [7] Technological Advancements and Strategic Differentiation - The competition is not just about user acquisition but also involves significant technological advancements, with companies launching new AI models and integrating them into their ecosystems [8][10] - Each major player is adopting distinct strategies: Alibaba is integrating its AI with e-commerce, ByteDance is leveraging its traffic and multi-modal capabilities, and Tencent is focusing on social integration [10][11] Future Outlook and Industry Trends - The Spring Festival marked a pivotal moment for AI applications in China, transitioning from novelty to mainstream usage, with a notable increase in older demographics engaging with AI [11][12] - The industry is entering a phase where the focus will shift from user acquisition to value creation and retention, indicating the beginning of an "AI oligopoly" era [13]

在全球视觉AI市场迎来爆发式增长的关键节点，软通动力正式推出历时三年打磨的"美通AI"超大分辨率 视觉模型。这款专为B端商业场景量身打造的技术成果，不仅直击行业痛点实现多重突破，更依托公司 坚实的研发投入与战略布局，成为软通动力在AI赛道开疆拓土的重要支点，彰显其"技术赋能商业"的核 心竞争力。 商业落地：头部客户密集签约，订阅模式打开增长空间 值得关注的是，美通AI已获得保利文化在内的行业巨头认可，客户覆盖零售、文化、酒店等多个高价 值领域，彰显出强劲的市场渗透力。作为软通动力AI战略布局的核心成果之一，经过三年持续迭代优 化，美通AI模型最终实现60K级超大分辨率视觉内容的高效生成，达到电影级视觉呈现标准。这一技术 突破使商业空间视觉内容制作周期从传统的数周缩短至最快15天，成本较传统制作模式大幅降低，为行 业效率革新树立了新标杆，目前已成功落地多个标杆项目。 在商业消费场景，海底捞子品牌捞宝街通过美通AI定制的戏曲戏剧、长白山雪景等十余套视觉主题实 现"数字换装"，有效提升客流吸引力；北京华联DT51、SKP、北京万达五棵松等顶级商业综合体，借 助其定制化动态内容激活公共空间，使传统广告位升级为场景化流 ...

Core Viewpoint - The current AI industry in the U.S. exhibits characteristics of a "rigid bubble," supported by deep integration into national strategy and strong policy backing, while also facing significant valuation pressures and competition between capital expenditure and output efficiency [1][2]. Group 1: Economic Importance and Policy Support - The development of AI has become central to U.S. national strategy and political correctness, receiving robust policy support and backing from major corporations [1]. - AI-related industries have rapidly increased their share of U.S. GDP from 5.35% in Q1 2020 to 7.36% by Q3 2025, indicating its role as a core growth pillar [1]. - The market capitalization of AI giants, represented by "MAG7," accounts for over 30% of the S&P 500 index, contributing significantly to market growth [1]. Group 2: Political Dynamics - The economic significance of AI has been deeply politicized, with influential figures like Musk and Sachs forming a closed loop of "personnel-policy-interest" through substantial political donations and direct participation [2]. - The U.S. government has systematically dismantled previous regulatory frameworks to facilitate AI industry growth, linking its political survival to the prosperity of the AI sector [2]. Group 3: Valuation and Market Dynamics - The valuation of AI stocks is currently high, with the Nasdaq index's forward P/E ratio stabilizing despite rising stock prices, indicating a rational basis for market growth [3]. - The 12-month forward EPS for the Nasdaq index has risen from approximately $610 at the end of 2024 to $826, suggesting strong earnings growth is absorbing stock price increases [3]. - Unlike the tech bubble era, current tech giants maintain steady revenue growth, supported by robust cash-generating core businesses [3]. Group 4: Constraints on AI Industry Expansion - The AI industry's expansion faces four significant constraints: a dangerous gap between capital expenditure and output efficiency, impending cash flow pressures, physical limitations in semiconductor production and energy supply, and unresolved competition in technology pathways [4][5][6]. - Major companies plan to increase capital expenditures to a total of 640.4 billion yuan by 2026, a 55% increase year-on-year, which may lead to a reliance on future revenue growth [5]. - The industry debt has rapidly escalated to over $150 billion, indicating a potential cash flow gap as companies strive to meet high shareholder return commitments [5]. Group 5: Investment Strategy Recommendations - The company suggests constructing a three-tier dynamic asset allocation strategy in anticipation of a narrative reversal in the AI bubble [7]. - The first tier focuses on "rock-solid" opportunities in internet giants with stable cash flows, providing downside protection and liquidity support [7]. - The second tier emphasizes "shovel-type" opportunities in computing infrastructure, benefiting from increased AI capital expenditure and potential profit increases due to supply constraints [8]. - The third tier targets "contrarian" opportunities in the software sector, where concerns about AI applications may be overblown, allowing for strategic positioning after market corrections [8].

Core Insights - The rapid development of AI is leading to strong demand for computing power, driven by continuous upgrades of large models [1] - The commercial monetization of large models is transitioning from a "low-cost/free user acquisition" phase to a "high-quality paid" phase, significantly increasing the demand for computing power [1] - Companies in the AI computing and optical communication sectors have shown strong performance in the US stock market during the holiday period [1] - Jensen Huang announced that a "world-first" chip will be unveiled at the GTC conference in March, indicating ongoing innovation in the AI industry [1] - The outlook for the AI industry chain, including large models and AI computing chains, remains positive [1]

每经记者｜陈婷 李宇彤 每经编辑｜金冥羽 魏文艺 记者|陈婷 李宇彤 编辑|金冥羽 魏文艺 杜波 校对|许绍航 用户留存率才是真正的"试金石" 随着春节假期结束，这场"烈火烹油"般的红包大战暂告一段落。 截至2月24日，互联网大厂们的大手笔投入都已陆续交出成绩单。《每日经济新闻》记者（以下简称"每经记者"）梳理发现，"拉新"无疑是今年春节期间 大厂们最为核心的成果。 2月23日，千问App（应用程序）发文称，春节期间，千问帮大家"一句话下单"近2亿次，全国平均每10人就有1人让千问下单。 此前的2月18日，腾讯宣布，元宝日活跃用户（DAU）超5000万，月活跃用户（MAU）已达1.14亿。 豆包官方则于2月17日透露，除夕（2月16日）当天豆包AI（人工智能）总互动19亿次，"豆包过年"活动在除夕帮助用户生成超过5000万张新春主题头像、 超1亿条新春祝福。 2026年春节之前，国内通用大模型行业已经历了数年的技术竞速与百模大战。如今，随着互联网大厂以"撒钱"姿态强势入局，AI应用的渗透率正以前所未 有的速度飙升至新高点，今年春节无疑成为中国AI从尝鲜走向全民普及的里程碑。 然而，在艾媒咨询CEO（首席执行官 ...

OpenAI周二宣布任命Arvind KC为首席人力资源官，这位前Roblox公司高管将负责招聘、入职培训和员 工发展，因为ChatGPT的制造商OpenAI正在不断扩张。据OpenAI发言人称，KC将向首席战略官Jason Kwon汇报工作。这家 人工智能初创公司目前正在进行新一轮融资，预计融资额将超过1000亿美元，并 且最近已采取措施推进上市计划。 ...

来源：环球市场播报 美股频道更多独家策划、专家专栏，免费查阅>> 埃隆-马斯克的人工智能初创公司xAI指控竞争对手山姆-奥特曼（Sam Altman）的OpenAI窃取其商业机 密，加州一名联邦法官周二驳回了xAI的诉讼。 责任编辑：山上 旧金山的美国地区法官Rita Lin表示，xAI可以重新提起诉讼，但目前未能指控OpenAI有任何不当行 为。今年9月提起的诉讼，声称xAI前员工在离职后前往OpenAI就职时带走了与其Grok聊天机器人相关 的源代码和其他机密信息。"值得注意的是，诉讼中没有对OpenAI本身的行径提出指控，" Lin说。"xAI 没有指控任何事实表明OpenAI诱使xAI的前员工窃取xAI的商业机密，也没有指控这些xAI的前员工在受 雇于OpenAI后使用了任何窃取的商业机密。"Lin曾在1月份的意见书表示她很可能会做出有利于OpenAI 的裁决。她要求xAI在3月17日前提交修改后的诉状。xAI 还另外起诉了前工程师李学勤，指控他将商业 机密提供给了 ChatGPT 制造商。在该案中，李学勤被阻止 与OpenAI分享xAI的技术，但OpenAI表示， 李学勤从未为该公司工作过，也从未获 ...

华西证券研报表示，近期中国AI大模型迎来"春节档"爆发，技术趋势已从通用聊天工具全面转向垂直生 产力工具与真实Agent落地，重点体现在工业级的视频生成（如Seedance 2.0的多模态叙事与音画同 步）、工程级编程（如智谱GLM-5的跨文件重构与系统工程能力）以及消费办公场景。当前时点，面 临近期的流动性担忧以及海外OpenAI资本开支预期的变化，反映市场对于当前AI发展逐渐聚焦于成本 效益比率，预计市场仍将保持震荡，板块建议相对谨慎，中性配置。目前AI发展阶段仍处于Scale up和 Scale out的关键加速期，伴随相关算力芯片的供应体系逐渐丰富，应用发展对于token需求量仍在加速增 加，相关底层算力基建仍然还在扩张期。仍坚定看好Scale up的柜内光升级，电源、液冷等产业领域向 国产厂商升级机遇，国产算力包括芯片、交换机、服务器等需求放量，以及CSP厂商资本开支加速带来 的算力租赁和AIDC相关市场机遇。 ...

Group 1 - The AI industry is experiencing rapid advancements, with large models continuously iterating and increasing demand for computing power [3][18][29] - Major AI companies are transitioning from a "low-cost/free user acquisition" phase to a "high-quality paid" phase, significantly boosting computing power demand [4][9][24] - During the holiday period, US AI computing stocks performed exceptionally well, with companies like Lumentum, Corning, and Coherent reaching historical highs [12][28][29] Group 2 - Anthropic launched the Claude Sonnet 4.6 model, which shows significant improvements in programming and reasoning capabilities, becoming the default model for its platforms [4][19] - Google's Gemini 3.1 Pro model outperformed several competitors in benchmark tests, achieving a score of 77.1% on the challenging ARC-AGI-2 test, doubling its predecessor's performance [6][21] - Zhizhu AI released the GLM-5 model, featuring a parameter scale of 744 billion and training data of 28.5 trillion tokens, achieving top scores in open-source model benchmarks [7][22] Group 3 - MiniMax introduced the M2.5 model, designed for agent scenarios, demonstrating significant improvements in programming tasks and cost efficiency [9][24] - Alibaba's Qwen3.5-Plus model outperformed similar open-source models, offering competitive pricing and extensive language support [10][25] - The domestic AI market saw a surge in user engagement during the Spring Festival, with significant increases in active users for various AI applications [12][27]