作者丨周芮 出品丨牛刀商业评论 2025 年末，科技圈被一则人事任命炸翻了天。 27 岁、OpenAI 前核心研究员、首席 AI 科学家，当这几个标签集中在姚顺雨身上，再加上 "统筹腾讯 AI 两大核心板块" 的重任，所有人都看懂了：腾讯要动真格，给深陷困境的 AI 业务换 "主心骨"。 彼时的腾讯 AI，早已没了巨头光环：事业群各自为战，每年白白烧掉 20 亿研发预算；技术榜单刷得漂 亮，产品却没人买单；手握微信 13 亿日活的王牌，却在 AI Agent 的浪潮中越跑越慢。 这场 "临危受命"，不是简单的人才引进，而是腾讯向 AI 下半场发起冲击的战略宣言。 姚顺雨主导的腾讯大模型研发新架构，能否在 AI 模型迭代效率与工程化落地能力上实现双重突破，助 力腾讯在 AI 竞争下半场成功突围，目前仍有待市场检验。 带着 OpenAI 的技术基因和清醒的财务思维，姚顺雨这位 27岁的"破局者"，到底要怎么盘活腾讯 AI 这 盘大棋？ 1 月 10 日，27 岁的腾讯首席 AI 科学家姚顺雨，通过线上连线的方式，完成了空降腾讯后的首次公开 发声，距离他打破行业常规接棒腾讯 AI 核心业务仅过去不久。 这位年轻科学 ...

从过去的BAT三巨头，到如今的腾讯、阿里巴巴、字节跳动三巨头，腾讯始终是最稳的那个。 而要看懂腾讯2026年甚至未来几年的发展前景，盘点腾讯2025年的财报就显得尤为重要了。腾讯在2025 年前三季度交出了一份"量价齐升"的成绩单：总收入突破5573亿元，同比增长14.26%；归母净利润高达 1665.82亿元，同比增长16.7%；三季度单季营收1928.7亿元，同比增长15.4%，这已经是腾讯连续多个 季度保持双位数增长。 三大引擎 腾讯能够持续多年保持业绩稳定增长，始终离不开核心业务引擎的发展。2025年前三季度稳定增长的背 后，同样离不开三大引擎。 引擎一：游戏业务华丽转身。从前三季度的数据来看，腾讯游戏已经成功从本土市场迈向国际市场，本 土游戏市场，腾讯依靠《王者荣耀》、《和平精英》等长青游戏，加上《三角洲行动》和《无畏契约： 源能行动》的爆发，依然稳步前行。 海外市场，得益于Supercell的持续贡献、新收购工作室的融合，以及《皇室战争》等老游戏的焕发生 机，腾讯国际游戏在2025年Q3首次突破200亿元大关，同比增长了54%。 引擎二：广告业务插上"AI翅膀"。在AI的赋能下，腾讯通过AI优化广 ...

在部分用户实际使用场景下，混元翻译1.8B和7B两个尺寸模型同时使用，可以实现端侧和云侧模型的 协同部署，提升模型的效果的一致性和稳定性。 据介绍，腾讯混元翻译模型此前不仅在国际机器翻译比赛拿下30个第1名，也在首次开源一周内便登上 了HuggingFace 模型趋势榜第一位。目前，混元翻译模型已经在腾讯内部多个业务场景落地应用，包括 腾讯会议、企业微信、QQ浏览器、客服翻译等。 HY-MT1.5-1.8B主要面向手机等消费级设备场景，经过量化，支持端侧直接部署和离线实时翻译，仅需 1GB内存即可流畅运行。HY-MT1.5-1.8B 的参数规模只有 1.8B，量化后仅需约 1GB 内存，和一款常见 手机应用体量大致相同。在参数量极小的前提下，在多项主流翻译测试集中，它的整体表现已经达到超 大尺寸闭源模型的90 分位水平，明显领先同尺寸开源模型与主流商用翻译 API。同时，模型也体现出 了极致的效率和性价比，与主流商用翻译模型API对比，HY-MT1.5-1.8B 推理速度更快，处理50个 tokens的平均耗时只有0.18秒，其他模型的时间在0.4秒左右。 HY-MT1.5-7B模型效果相比前一版本效果有较大 ...

代表人物之一是个荷兰人——Pieter Levels。 这兄弟也不融资，没团队，就一台电脑开整，但一年差不多整到270万美金。 不要以为这纯粹就是偶然的，下面这个表单大家如果仔细看，就会发现，至少还有一个 （排第五名 那个） ： | | | | LeanAlLeaderboard.com - Top Lean Al Native Companies Leaderboard : Official | | | | | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | Company | Description | Location | Annual Revenue | # of Employe | Revenue/Empl | Profitable | | 1 | | Telegram | Messaging | Dubai | $1,000,000,000 | 30 | $33.333.333 | Yes | | 2 | 17 | Midjourney | Image Generation | San Francisco | $500 ...

Core Insights - AI has become the central theme in China's technology sector throughout 2025, viewed as a key driver for the next productivity revolution and business model iteration [1] - Major internet companies like Tencent, Alibaba, and ByteDance are intensifying their focus on AI, with significant investments and strategic moves to secure their positions in the market [1][2] Investment Trends - Alibaba announced a three-year investment plan of 380 billion yuan for cloud and AI hardware infrastructure, which is more than its total investment over the past decade [2] - ByteDance plans to invest 150 billion yuan in 2025, marking an 88% year-on-year increase, with expectations to further increase spending to 160 billion yuan in 2026 [2] - Tencent's capital expenditure surged by 386% year-on-year to 36.58 billion yuan in Q4 2024, with projections for 2025 spending to reach around 100 billion yuan [3] Financial Performance - Alibaba's cloud revenue grew by 34% year-on-year in Q3 2025, with AI-related product revenue maintaining triple-digit growth for nine consecutive quarters [3] - Tencent's marketing services revenue increased by 21% year-on-year, driven by AI improvements in its advertising platform [3] - Baidu's AI business revenue grew over 50% year-on-year, becoming a core growth engine despite a decline in total revenue [3] AI Integration in Business - Major companies are embedding AI into their core business lines to enhance operational efficiency and create new growth engines [6][7] - Alibaba's AI tools improved operational efficiency for 500,000 merchants by an average of 1.5 times during the 2025 "Double 11" shopping festival [7] - Tencent's AI-driven advertising has led to a 21% increase in marketing service revenue, showcasing the shift from traditional to intelligent advertising [8] New Market Opportunities - The competition for AI-native applications is emerging as a new growth area, with companies vying to establish the next "super entrance" in the AI era [9][10] - The market for AI applications is evolving, with significant user engagement reported for products like ByteDance's Doubao and Tencent's Yuanbao [10] - The global smart glasses market saw a 64.2% year-on-year increase in shipments in the first half of 2025, indicating strong potential for AI hardware [11] Future Outlook - The competition among companies in AI applications, hardware, and native applications is still in its early stages, with the need for validation of user engagement and business models [12]

Core Viewpoint - DingTalk is expanding its AI strategy beyond office software, aiming to establish itself as an AI ToB entry point that integrates both software and hardware solutions [1][4]. Group 1: AI Product Development - DingTalk launched over 20 AI products during the AI DingTalk 1.1 product release and ecosystem conference, emphasizing its goal to connect AI with the physical world [1]. - The introduction of AgentOS marks DingTalk's first AI-focused intelligent operating system, which will serve as the foundation for all future AI agents on the platform [1]. - The core of AgentOS is to facilitate the operation and collaboration of AI agents while creating an open AI ecosystem to attract more industry players [1]. Group 2: Hardware Investment - DingTalk is accelerating its investment in hardware, highlighted by the launch of the enterprise-level AI hardware DingTalk Real, which aims to enable AI to autonomously connect and interact with the physical world [4]. - The previous hardware, DingTalk A1, has undergone over 2,000 co-creation sessions with clients in the past four months, maintaining a rapid iteration cycle of every two weeks [4]. - DingTalk Real is designed to ensure that AI operations remain within defined boundaries, maintaining human control over the system [4]. Group 3: Competitive Landscape - Alibaba has been focusing on AI hardware in the consumer sector, with the launch of Quark AI glasses and the integration of the Tmall Genie team with the Quark team to explore hardware directions [5]. - The competition among DingTalk, Feishu, and Tencent Meeting in the AI spreadsheet domain is expected to intensify, with both DingTalk and Feishu continuously enhancing their product capabilities [5]. - DingTalk is shifting its focus in AI spreadsheets to create an application development platform, allowing enterprises to generate AI applications from Excel-based data through various input methods [6].

转自：北京日报客户端 关于发布完成个人信息收集使用优化改进App清单的公告（2025年第6批） 为规范App收集使用个人信息行为，保护个人信息权益，推动形成全社会共同维护个人信息安全的良好 环境，中国网络空间安全协会组织指导酒店服务、网络社区、女性健康、远程会议和投资理财等5类6款 App运营方，对照《中华人民共和国网络安全法》《中华人民共和国个人信息保护法》《常见类型移动 互联网应用程序必要个人信息范围规定》等法律法规，重点针对超范围收集个人信息、过度调用敏感权 限、权限设置和账号注销不便等个人信息收集使用问题完成了优化改进。 6款完成个人信息收集使用优化改进App清单 中国网络空间安全协会 2025年12月23日 来源：中国网络空间安全协会 6款App运营方已在应用商店或者官网上架优化改进版本（名单及版本号附后），并承诺升级版本持续 保持合规水平。现将App清单公布如下： | 序号 | 类型 | App 名称 | 运营方 | 版本号 | | --- | --- | --- | --- | --- | | l | 酒店服务 | 华住会 | 汉庭星空(上海)酒店管理有限公司 | 9.38.1 | | 2 | ...

Core Viewpoint - Tencent's recent structural upgrade, including the establishment of AI Infra, AI Data, and Data Computing Platform departments, aims to comprehensively reconstruct its foundational capabilities for large models, reflecting a significant shift in management logic within the tech industry [5][6]. Group 1: Tencent's AI Strategy - Over the past year, Tencent has released over 30 versions of its Hunyuan large model, integrating AI capabilities into more than 900 business scenarios, indicating a shift from AI as an experimental project to a core component of business processes [8]. - The company faces substantial technical pressure due to the complexity of managing over 900 AI application scenarios, necessitating a focus on redefining problems rather than merely increasing model parameters [8][9]. - Tencent's aggressive talent acquisition strategy, characterized by offers up to 50% higher than previous salaries, reflects the urgency to secure top AI talent amid a competitive landscape [9]. Group 2: Leadership and Talent Dynamics - The appointment of 27-year-old Yao Shunyu as a key AI figure at Tencent signifies a departure from traditional expectations of extensive experience, emphasizing the importance of judgment and problem definition in the rapidly evolving AI landscape [10][12]. - Yao's background, including his contributions to AI frameworks and his critical questioning of AI's real-world utility, positions him as a leader capable of bridging technical innovation with practical business needs [11][12]. - The shift towards younger leadership in tech roles, particularly in AI, suggests a broader trend where experience is no longer the sole determinant of decision-making authority, highlighting the value of fresh perspectives in navigating technological disruptions [15][16].

【#腾讯回应OpenAI前研究员姚顺雨入职#：任总办首席AI科学家】 36氪智能涌现独家获悉，腾讯近期完成了一次组织调整，正式新成立AI Infra部、AI Data部、数据计算 平台部。 12月17日下午发布的内部公告中，腾讯表示，Vinces Yao将出任"CEO/总裁办公室"首席 AI 科学家，向 腾讯总裁刘炽平汇报；他同时兼任AI Infra部、大语言模型部负责人，向技术工程事业群总裁卢山汇 报。 腾讯并未披露Vinces Yao的中文名或过往履历。不过，《智能涌现》了解到，Vinces Yao即为数月前入 职腾讯的姚顺雨，他毕业于清华和普林斯顿大学，曾任OpenAI研究员，是OpenAI首批智能体产品 Operator与Deep Research的核心贡献者。 此次调整也是腾讯在今年紧锣密鼓的AI布局中，颇为重磅的一步。 此前，混元大模型团队虽是腾讯的公司级项目，拉通了各个BG的不同板块，就TEG内部而言，参与到 混元模型研发的就有大预语言模型部、AI Lab、机器学习平台等等部门。经过调整后，模型团队内部力 量会更加统一。 新成立的 AI Data 部、数据计算平台部，将分别负责大模型数据及评测体 ...

Core Viewpoint - Tencent has made significant adjustments to its AI model organization structure, focusing on enhancing foundational research and engineering capabilities in the AI era [1][2]. Group 1: Organizational Changes - Tencent has established new departments: AI Infra, AI Data, and Data Computing Platform, to strengthen its AI model research and engineering capabilities [1]. - Yao Shunyu has been appointed as the Chief AI Scientist, overseeing both the AI Infra department and the large language model department, reporting to the company's president [1][2]. - The restructuring emphasizes the integration of data capabilities, with the new AI Data and Data Computing Platform departments prioritizing data cleaning, evaluation systems, and big data platform integration [1]. Group 2: Strategic Focus - The recent adjustments mark a shift from algorithm development to a stronger emphasis on computing power and data, reinforcing Tencent's engineering advantages while enhancing AI model research capabilities [2]. - Tencent aims to improve the efficiency of AI model development and align with its broader AI strategic layout [2]. Group 3: Market Position and Competition - Tencent's mixed Yuan model has been implemented in over 900 internal business scenarios, including Tencent Meeting and WeChat [3]. - Over 90% of Tencent engineers are utilizing the internal code assistant CodeBuddy, with an AI code adoption rate of 50% [3]. - Tencent is actively upgrading its mixed Yuan model and expanding its top-tier R&D talent, while also integrating AI capabilities into major platforms like WeChat and QQ to increase user engagement [3]. - Competitors like Alibaba and ByteDance are also intensifying their investments in AI infrastructure and user engagement strategies, indicating a competitive landscape in the AI sector [3].