（原标题：AI时代的"降本增效浪潮"到来! 微软(MSFT.US)猛裁1.5万人 从客服到编程都交给AI） 智通财经APP获悉，有媒体援引知情人士透露的消息报道称，AI应用软件领军者微软(MSFT.US)近期加 快步伐向员工们展示与普及人工智能应用如何改变其内部工作方式，以及如何大举提高软件工程项目进 展与各项业务经营效率。 与此同时，该公司第二轮大裁员在此期间已经开始，今年以来微软已经开启两轮大裁员行动，预计累计 裁减员工数量将高达1.5万人。微软于上周开启2025年第二轮大规模裁员行动，预计削减大约9000个就 业岗位，占其全球员工总数的约5%。 在一些华尔街分析师看来，作为全球云计算与AI应用领军者开启的连续两轮大裁员以及加码布局内部 AI工具提高效率，意味着在生成式AI应用与AI智能体驱动之下，人工智能时代的"降本增效超级浪 潮"可能已经到来。 据知情人士透露，在本周的一次演示中，微软首席商务官贾德森·阿尔托夫(Judson Althoff)表示，基于最 先进AI大模型的人工智能应用工具正在从产品销售、客户服务到大型软件工程项目落地等公司核心业 务的方方面面大举提升生产力以及经营效率。 谷歌母公司Al ...

人类对AI最大的误会之一，就是以为它吃这套人情世故。 你跟Deepseek说"谢谢"，它会不会更听你的话？不会。它只会更耗电。 你以为是在打招呼，它以为你在提问。你说"谢谢你"，它琢磨半天："这是问题的一部分吗？这是请求结束了吗？这个人到底要我干嘛？" AI听不懂礼貌，它只识别信号。 别用社会那套来要求AI，它只认关键字和语义点 对AI说"谢谢"的人，大概分两类：一类是文明人，礼貌教育刻在骨子里；另一类，是在赌未来AI统治世界时，自己能被分到好人阵营。 但我要说句扎心的：AI不吃这套。 你说"谢谢"它不会感动，你说"我给你小费"，它也不会加班加点。你说"你是所有领域的专家"，它也不会脸红。 AI不是人，它没情绪，也没有被尊重的感觉。它的大脑里没有什么所谓的一些用户臆想出来的"感恩机制"，也不会"记仇"或者"记得你是个好人"。 当然，如果你真的哪天受不了，对它破口大骂，它也不会立刻把你拉黑（不过厂商可能会）。 AI到底是怎么理解我们说的话的？简单来说，它真的不懂。 它只是在识别你的词语模式：你说了哪些词？这些词和什么语义靠得最近？这些语义在它见过的数万亿条语料中，最可能对应怎样的回复？ 举个例子各位爷就明白了。 ...

Core Viewpoint - The partnership between OpenAI and Microsoft is facing unprecedented challenges as they negotiate the definition of "Artificial General Intelligence" (AGI), which involves significant commercial interests and control over AI technology [1][3]. Group 1: Investment and Agreements - Microsoft invested $1 billion in OpenAI in 2019, followed by an additional $13 billion in subsequent agreements, which included terms that allow OpenAI to restrict Microsoft's access to future technologies once AGI is achieved [3]. - The agreements aim to ensure OpenAI's technological independence and prevent monopolization by a single commercial entity [3]. Group 2: Definition and Standards of AGI - There is currently no clear technical definition of AGI, but it is generally believed to require cross-domain generality, autonomous evolution, consciousness, and self-awareness [3]. - OpenAI favors a "functionalism" definition based on the system's actual capabilities, while Microsoft advocates for a "gradual assessment" approach to avoid premature triggering of technology restriction clauses [3][4]. Group 3: Implications for Microsoft - Microsoft's reliance on OpenAI's technology is deeply integrated into its core business, with Azure providing computational support and OpenAI's API embedded in tools like Microsoft Copilot and GitHub Copilot [4]. - If OpenAI restricts technology sharing upon achieving AGI, Microsoft could face technological obsolescence, strategic disadvantages, and financial losses [4]. Group 4: Negotiation Stalemate - Negotiations between OpenAI and Microsoft have reached an impasse, with Microsoft proposing to transfer AGI determination rights to a third-party organization, which OpenAI has rejected due to concerns over compromising technical confidentiality [4]. - OpenAI is forming an independent committee with academic and ethical experts to establish AGI determination standards and plans to release a technical white paper by 2026 [4].

图片来源： Scale AI 据 OpenAI 发言人及近期与该初创公司洽谈的两家大企业高管透露 ， OpenAI 已开始对同意购买额外 AI 产品的企业客户提供 ChatGPT 应用订阅折扣。 • OpenAI 为购买附加产品的客户提供 10% 至 20% 的 ChatGPT 企业版折扣 • 该公司预计到 2030 年， ChatGPT 企业客户将贡献近 150 亿美元年收入 • 微软最大折扣仅5%-10%， 微软 销售人员因无法匹配 OpenAI 的折扣而错失交易 这一举措令微软的销售人员感到沮丧。作为 OpenAI 最大商业合作伙伴，微软销售与之竞争的应用和 模型，但通常不会提供同等力度的折扣。 此次降价表明， OpenAI 对企业客户的定价能力可能不如对消费者市场强势——后者目前贡献了其绝 大部分收入 。这也反映出 AI 公司在找到降低数据中心模型运行成本的方法后，正尝试将部分节省成 本让利给大客户。 OpenAI 对 ChatGPT 提供的折扣幅度可达 10% 至 20% ，前提是客户签署多年协议购买其他 OpenAI 工具，包括可生成完整报告的定制版 Deep Research 智能体，以及能自主 ...

如果只看热度，这几个产品中Seedance应该是被关注度最高的了。 它刚上线就登上了第三方榜单Artificial Analysis的文生视频和图生视频两项第一，超过了可灵2.0和谷歌的Veo 3。榜单 怎么评的我们可以再讨论，但至少说明在业内标准下，Seedance的生成效果已经达到了主流模型中的较好水平。 | Artificial Analysis Video Arena Leaderboard | | | | | | Artificial Analysis Video Arena Leaderboard | | | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | Text to Video | Image to Video | | | | Text to Video | Image to Video | | | | Creator | Model | Arena ELO | 95% CI | II Appearances | Creator | Model | Armina ELO | 95% CL | II App ...

数字化和人工智能(AI)正在创造可能改变化学工业格局的机遇和风险，这似乎成为全球化工行业业内人 士的共识。近期，麦肯锡公司高级合伙人兼全球化学品实践负责人克里斯·穆索在由法国化学工业协会 主办的"探索化学工业中AI机会"网络研讨会上表示，AI正极大地使化学公司受益，因为AI使这些公司 能够轻松借鉴过去的成果，并能够随时随地运用行业内部的研究进展。 美国国际香精香料公司(IFF)企业数据和AI主管莫伊恩·哈克表示，数字技术的真正优势在于加速发现和 实验过程，并有效地瞄准新想法。哈克在此次网络研讨会上表示："这些想法最好源于市场情报所提供 的市场洞察和消费者趋势。将这些洞察融入产品组合对于发现差异化需求至关重要。"此外，数据的数 字化正在加强知识管理，促进协作，改善信息共享，从而有可能增加价值并加快流程。当然，在更广泛 的范围内分享知识仍然是一个挑战。 穆索表示："我们正处在一场数字革命的边缘，将对化工行业产生重大影响。那些快速适应这种数字转 型的公司可能会收购那些不能适应的公司，并导致行业整合。这场数字革命将会决定未来的收购趋势， 塑造领导者并重塑行业。" 德国汉高黏合剂创新业务全球主管迈克尔·托德表示，新冠疫 ...

Core Insights - The report by Mary Meeker highlights the unprecedented speed and scale of AI adoption, indicating a transformative impact on technology history [3][6][22] - AI is experiencing exponential growth, with ChatGPT reaching 800 million users in just 17 months, surpassing any product from the internet era [3][8] - The report emphasizes a shift in AI development focus from academia to industry, driven by proprietary interests and competitive advantages [6][10] User Growth - ChatGPT achieved 800 million users within 17 months, with an annual recurring revenue growth rate that outpaces any product from the internet era [3][8] - The rapid user adoption of AI technologies is reshaping the landscape of digital interaction and functionality [8][18] Cost Dynamics - Training costs for AI models can reach up to $1 billion, but inference costs have decreased by 99% over two years [4][14] - The energy efficiency of GPUs has significantly improved, with NVIDIA's 2024 Blackwell GPU showing a 105,000-fold reduction in power consumption compared to the 2014 Kepler GPU [4][14] Competitive Landscape - The rise of Chinese firms in the AI space is notable, with open-source approaches enabling rapid advancements and global competition [4][10] - Closed-source models like OpenAI's GPT-4 and Anthropic's Claude dominate enterprise applications due to their superior performance, despite lacking transparency [6][10][13] Infrastructure and Investment - The demand for AI infrastructure is increasing, putting pressure on cloud providers and chip manufacturers [8][21] - Significant capital investment is required for AI development, with ongoing competition among companies for key technologies like chips and data centers [21][22] Job Market Impact - Since 2018, job vacancies related to AI have surged by 448%, indicating strong demand for talent in the AI sector [19][22] - AI is evolving roles in various professions, enhancing productivity rather than replacing jobs [18][22] Market Segmentation - The AI market is bifurcating into closed-source models, which are favored by enterprises, and open-source models, which are gaining traction among developers and startups [10][12][13] - Open-source models are becoming increasingly competitive, offering low-cost alternatives with robust capabilities [12][13] Strategic Implications - Companies are shifting from selling isolated software licenses to integrating AI functionalities across their technology stacks, focusing on delivering tangible outcomes [21][22] - The competition in AI is likened to a space race, highlighting the strategic importance of technological advancements in this field [21][22]

2025年，AI战场硝烟弥漫。 OpenAI的GPT-4.5已能自主编程、微软Copilot彻底重塑办公生态、Anthropic的Claude 4在长文本推理上独步天下，而DeepSeek则凭架构创新跻身世界前 列。 然而，在这场决定未来的AI军备竞赛中，曾经的王者——谷歌，却逐渐沦为配角。 三年前，谷歌还是AI领域的绝对霸主：Transformer架构的发明者、AlphaGo的缔造者、DeepMind与Google Brain的掌控者。 如今，它的Gemini大模型却始终未能像ChatGPT那样颠覆行业，甚至被诟病"缺乏创新""官僚迟缓""战略摇摆"。 诞生无数AI研究成果的谷歌一度成为杰弗里·辛顿等全球顶尖AI人才向往的麦加。 但变化莫测的棋局中，并没有永恒的胜利者。 谷歌，这个曾和OpenAI旗鼓相当的头号玩家，到底是如何从AI巅峰跌落，沦为追赶者的？ Al的变局与谷歌的迟钝 回顾人工智能的发展史，谷歌的名字曾与无数里程碑式的突破紧密相连。 2012年，谷歌研究的神经网络在YouTube视频中自发识别出一只猫，成为深度学习革命的起点；2016年，AlphaGo击败世界围棋冠军李世石，彼时OpenAI 刚 ...

3月，中国智能体初创公司Manus爆火，智能体（AI Agent）这个概念瞬间被推到了公众面前。 如果说AI大语言模型（LLM）是人的大脑，能存储知识和逻辑推理，但没法自己动手干活；那构建于大语言模型之上的智能体（AI Agent）则相当于一位 数字员工，能够自己执行任务，参与到日常工作的决策和执行中去。 如今互联网巨头纷纷"逐鹿"智能体，字节Coze、腾讯元器、百度文心等智能体平台开启混战，智能体正渗透至各行各业。 但目前，大多数人还只是听说过智能体，真正实际应用或从0到1开发智能体的并不多。 而企业对智能体人才却是求贤若渴。Boss直聘显示，不仅百度、字节等互联网大厂挂出了大量和智能体相关的在招岗位，就连小规模初创公司也急需智能 体开发工程师，月薪普遍都在2万元以上。 正是因为这样的供需不平衡，在智能体这条蓄势待发的赛道上，智能体就业培训师成了第一批追赶风口的人。 从"智能体来了"的课程内容可以发现，的确都包括了构建智能体的方法论、工作流结构构建、数据输入输出和插件的实战调用等方面。此外，线下课 是"项目制"，以实操为主。 | Ai智能体应用开发工程师2.5个月就业班培训课纲 | | --- | | 总体 ...

Core Insights - Microsoft Build 2025 developer conference emphasizes the theme of "Open Agentic Web," aiming to establish AI agents as a new foundational infrastructure for the internet [2] - Microsoft Copilot transitions from a functional tool to a development paradigm, while Windows becomes a habitat for AI agents and Azure serves as the training, deployment, and interaction hub for these agents [2] - Anker Innovations showcases its soundcore audio brand at the conference, highlighting the integration of AI technology with consumer audio devices [2] Company Highlights - Anker Innovations' President Zhao Dongping expresses pride in participating in the Microsoft Build conference, focusing on redefining smart audio products and enhancing audio interaction experiences across cultures [2] - Microsoft China Vice President Liang Yan praises Anker Innovations as a model for Chinese tech companies in globalization, emphasizing the importance of innovative collaboration in enhancing user experiences [2]