但斌表示，AI行业竞争激烈。亚马逊今年研发投入1250亿美金，谷歌900亿，微软联同Open AI要投 1000亿。一旦AI Agent实现，将来订飞机票可能只需要对着眼镜讲句话就搞定，连手机形态都可能消 失。到时候，连腾讯微信的商业模式都有机会发生大变化。 面对2022年开启的AI时代，但斌只做简单的战略性抉择，集中投资于以英伟达为代表的公司。今年， 东方港湾做出方向调整。"过去两年英伟达是第一大重仓，但今年初研究团队发现，AI对谷歌有推动作 用，并非负面影响，于是将谷歌加仓至第二大重仓。" 格隆汇12月2日｜私募大佬、东方港湾董事长但斌近日在一场活动上分享他的投资感悟时表示，如果AI Agent成真了，微信模式都要改写，即腾讯有可能被颠覆。 ...

截至9点56分，科创人工智能ETF（589010） 早盘受大盘及板块影响呈现震荡调整走势，现跌 0.88%，价格在日内均线下方运行。持仓股方面分化显著，晶晨股份表现强势，逆势上涨超2%，优刻 得、海天瑞声等维持红盘；芯原股份则领跌超6%。值得注意的是，虽然盘面短期波动，但流动性充 裕，开盘半小时成交额已超3300万元。结合昨日单日超1亿元的资金净流入情况来看，部分资金或正借 板块调整之机，在合理配置区间内积极布局，静待后续行情修复。 消息方面，12月1日，可灵AI正式发布其全新产品"可灵O1"，定位为首个大一统的多模态创作工 具。可灵O1基于全新的视频和图像模型，整合文字、视频、图片、主体等多模态输入，将所有生成和 编辑任务融合于一个全能引擎之中。可灵O1的推出，彻底解决了AI视频生成中角色、场景等一致性难 题，为影视、自媒体、广告电商等应用场景，提供了深度适配的一站式解决方案。 银河证券表示，AI Agent将在2026年呈现结构性加速落地的趋势。AI Agent领域当下呈现的"雷声大 雨点小"本质是技术成熟度与商业化节奏的阶段性错配，2026年加速落地趋势不变。从供给端看，MMC 调研显示，当前90%的A ...

互联网大厂入局，端侧 AI 硬件加速 20251201 摘要 端侧 AI 硬件在消费电子领域的发展背景是什么？ 端侧 AI 硬件的发展背景主要有两个逻辑。首先，从必要性角度来看，随着云端 算力的快速发展，将云端算力下沉到端侧成为一个重要方向和议题。在全球数 十亿台手机、PC 和平板的保有量下，消费群体庞大，因此在 To C 市场中，手 机与消费者日常生活紧密联系，各大巨头（无论是互联网公司还是硬件厂商） 都在尝试开发端侧 AI 硬件。其次，从可行性角度来看，云端算力的 token 消 耗量持续增加，如果未来在 To C 市场大规模推广，仅依靠云端算力调用会面 临成本和安全方面的问题。因此，端侧 AI 硬件作为未来 To C 应用和云端算力 之间的桥梁应运而生。未来，端云结合的算力部署和消耗分布将成为主流。 智能化功能对传统设备渗透率潜力巨大，若 2%的普通眼镜转变为智能 化设备，即可达到 4,000 万副市场规模。大厂推动消费者教育和成熟产 品发布将扩大智能设备在传统领域的渗透。 端侧 AI 加速落地推动传统蜂窝通信模组智能化升级，模组厂商角色日益 重要，业务延伸至解决方案方向。预计到 2030 年，嵌入式 ...

Core Insights - DeepSeek has launched two new models, DeepSeek-V3.2 and DeepSeek-V3.2-Speciale, marking a significant advancement in AI capabilities, particularly in reasoning and output efficiency [2][3] - The V3.2 model is positioned as the strongest open-source large model, outperforming competitors in various benchmarks while significantly reducing output length and computational costs [3][4] - The V3.2 model integrates a new sparse attention mechanism (DSA) to enhance performance in long-context scenarios, while also improving the model's ability to follow instructions and generalize in complex environments [8][9] Model Performance - In benchmark tests, DeepSeek-V3.2 achieved competitive scores against models like GPT-5, Claude 4.5, and Gemini 3 Pro, with notable strengths in specific areas [4][5] - The V3.2 model demonstrated superior performance in question-and-answer scenarios, providing detailed and accurate travel recommendations through advanced tool usage [5][6] - The V3.2 Speciale model focuses on maximizing reasoning capabilities, achieving results comparable to Gemini 3.0 Pro in mainstream reasoning benchmarks, although it requires a higher token cost and is not designed for everyday use [9][10] Development Focus - DeepSeek emphasizes practical usability and generalization in its models, aiming to overcome common pitfalls in AI interactions, such as making basic common-sense errors [6][8] - The company is committed to enhancing the reasoning abilities of its models, as evidenced by the integration of advanced mathematical reasoning capabilities from the recently released DeepSeek-Math-V2 [9][10] - The competitive landscape for large models is intensifying, with major players like GPT-5 and Gemini 3 pushing the boundaries of AI capabilities, suggesting a dynamic future for AI development [10]

Market Overview - The main board saw significant gains with notable increases in stocks such as Shida Group (+10.04%), Furong Technology (+10.01%), and others [1] - The ChiNext board led the gains with stocks like Guanghetong (+20.01%) and Kexiang Co. (+20.00%) [1] - The Sci-Tech Innovation board also experienced growth, highlighted by Yunzuka Technology (+20.00%) and Huaren Micro (+9.69%) [1] - Active sub-industries included SW Brand Consumer Electronics (+3.79%) and SW Communication Network Equipment and Devices (+3.52%) [1] Domestic News - According to the Ministry of Industry and Information Technology, in the first ten months, mobile phone production reached 1.25 billion units, a year-on-year decrease of 4.7%, while smartphone production was 1.02 billion units, showing a slight increase of 0.7% [1] - A new space data center construction plan was announced by the Beijing Municipal Science and Technology Commission, aiming to operate a centralized large-scale data center system in the Dawn-Dusk Orbit [1] - Omdia forecasts that by Q2 2025, the cloud infrastructure service market in mainland China will reach $12.4 billion, reflecting a year-on-year growth of 21% [1] - SEMI reported that Shaanxi Electronics has launched an 8-inch high-performance semiconductor production line, with an initial investment of 3.2 billion yuan and a designed monthly capacity of 50,000 wafers [1] Company Announcements - Huahai Chengke completed its share repurchase plan, acquiring 606,400 shares, which is 0.70% of its total share capital, for a total payment of 46.46 million yuan [3] - Liyang Chip announced that shareholder Zhang Liping reduced his holdings by 6,105,887 shares, exceeding the original plan by 2.9979% [3] - Hanshuo Technology reported a total of 680,000 shares repurchased, representing 0.1610% of its total share capital, with a total transaction amount of 36.25 million yuan [3] - Chip Origin announced a plan for shareholders to reduce their holdings, with a maximum of 5,258,582 shares to be sold [3] International News - Mira announced the completion of a $6.6 million seed round financing to develop a new AI smart glasses product [4] - SK Hynix plans to expand its production capacity for HBM memory and general DRAM memory in response to growing demand [4] - Merck Electronics has begun qualification procedures for its new chemical plant in Taiwan, which is set to start mass production in 2026 [4] - Omdia reported a 30% quarter-on-quarter increase in global DRAM sales in Q3, reaching $40.3 billion, with SK Hynix maintaining a market share of 34.1% [4]

Core Viewpoint - The article presents a comprehensive overview of the concept of an Egocentric Procedural AI Assistant (EgoProceAssist), which aims to assist individuals in performing daily procedural tasks from a first-person perspective. It identifies three core technical tasks necessary for this assistant: Egocentric Procedural Error Detection, Egocentric Procedural Learning, and Egocentric Procedural Question Answering [6][32]. Summary by Sections Motivation - The article emphasizes the prevalence of procedural tasks in daily life, which require a specific sequence of steps to achieve desired outcomes. It highlights the potential of an AI assistant to enhance safety and efficiency in performing these tasks, especially in high-risk scenarios [6][8]. New Classification System - A novel classification system is introduced, categorizing the three core tasks of the AI assistant and summarizing existing methods, datasets, and evaluation metrics relevant to each task [2][6]. Egocentric Procedural Error Detection - This section outlines the existing key technologies for detecting procedural errors from a first-person perspective. It differentiates between methods that require only video data and those that utilize multimodal data, emphasizing the unique challenges of procedural error detection compared to general anomaly detection [9][11][12]. Egocentric Procedural Learning - The article discusses various approaches to procedural learning, categorized by supervision levels: unsupervised, weakly supervised, and self-supervised methods. It highlights the importance of identifying key steps in procedural tasks to improve error detection and planning capabilities [14][16]. Egocentric Procedural Question Answering - This section summarizes current technologies for answering procedural questions from a first-person perspective, noting the challenges posed by occlusions and scene changes. It emphasizes the need for models to possess strong understanding and memory capabilities to effectively respond to user queries [17][20]. Supplementary Experiments - The article presents supplementary experiments that evaluate the performance of existing VLMs and AI agents in procedural error detection and learning tasks. The results indicate significant limitations in their ability to assist with first-person procedural tasks [23][25]. Challenges - The article identifies several challenges in developing the EgoProceAssist, including data scarcity, limited understanding of long-term procedural activities, and heavy reliance on manual annotations, which hinder real-time assistance capabilities [29][30][31]. Conclusion - The research concludes by reiterating the significance of the proposed AI assistant and its core tasks, while also addressing the ongoing challenges and limitations in the field. It aims to provide a foundation for future research directions in egocentric AI applications [32].

一、事件：2025火山引擎原动力大会 2025年火山引擎Force原动力大会将于2025年12月18-19日举行，主要聚焦人工智能技术与产业融合，可能发布新品。 公司与努比亚手机的合作主要涉及努比亚红魔8S Pro、8/8pro/8pro+、红魔9Pro、平板等机型。 刚刚重磅发布AI手机 除软件外，端侧硬件为字节持续布局的重点领域，AI时代产品包括智能体耳机、陪伴玩具等，以及12月1日刚刚发布合作的AI手机。 12月1日，字节跳动发布豆包手机助手技术预览版，嵌入中兴手机系统。实测显示，该AI具备惊人的"智能体"能力，可模拟人工跨App操作，实现"全网比价 后自动点外卖"、"根据屏幕内容回微信"等复杂指令。 豆包方面表示，这是一款需要和手机厂商一起完成的产品。豆包目前正与多家手机厂商洽谈助手合作，并没有自己开发手机的计划，后续将公布更多进展。 招商证券表示，相比目前主流AI手机，这款手机在字节和中兴努比亚团队的深度合作下，对硬件层、软件层、甚至操作系统层，都进行了更深入的整合和 重构，这被定义为一款"AI原生手机"（AI Native Phone）。其核心差异化在于高权限Agent能力。手机系统内嵌AI Ag ...

↑点击上面图片 加雪球核心交流群 ↑ 国泰海通证券近日表示，未来，豆包大模型有望深度集成到手机中，本质上是豆包尝试在操作系统底层的渗透。预计豆包手机不仅仅是对话，例如 其文生图、图生图能力可以直接在手机系统相册编辑时方便调用，同时具备读取屏幕内容、记忆用户习惯、调用其他应用接口的能力，推动Agent 形态的形成。此外，苹果siri有望加速向AI Agent形态迈进，带动手机换机以及ar眼镜等端侧产品渗透。 雪球App用户@ 小声比比 ： 今天三大指数齐涨，沪指放量涨0.65%收复3900点， 深成指涨1.25%，创业板指涨1.31%。 沪深两市成交额1.87万亿，较上一个交易日放量 2881亿，全市场近3400只个股上涨。 盘面上，有色金属、AI端侧方向等多点开花，两千亿中兴通讯涨停，白银领涨贵金属板块，此外受到 《疯狂动物城2》票房影响，中国电影也一字 涨停。 01 千亿中兴通讯爆拉涨停！ ai手机概念股大幅拉升，中兴通讯、福蓉科技、道明光学等多股涨停。带动消费电子板块集体集体走强。 消息面上，12月1日，字节跳动旗下豆包团队发布豆包手机助手技术预览版。豆包手机助手是以豆包APP为基础，与手机厂商在操作系统 ...

【环球网财经综合报道】日前，北京市科委、中关村管委会发布《北京人工智能产业白皮书（2025）》预测，各类AI Agent将迎来爆发式增长。具身智能的发展将实现从信息处理到物理作业的跨越。 白皮书指出，人工智能将促进技术普惠发展；端侧智能催生应用新蓝海，智能手机、个人电脑、智能汽车等端侧设备 将具备更强的智能处理能力。白皮书初步估算2025年全年，北京AI产业规模有望超过4500亿元。 NBCNEWS也发文提到，美国越来越多的硅谷公司正在采纳中国生产的性能强大且可免费定制的开源人工智能模型， 如DeepSeek的R1和阿里巴巴的Qwen，以降低成本和提高效率，这些中国模型在技术上接近甚至超越美国封闭模型。 这引发了关于美国人工智能行业可能依赖外国技术以及本土封闭模型策略是否合理的讨论，尽管美国公司仍保有在人 工智能能力前沿的优势，但中国模型的崛起和对外开放策略正挑战着美国在开源领域的领导地位。 新加坡海峡时报近日发文称，中国的人工智能发展迅速，特别是在生成式AI领域，挑战了美国的领先地位。AI将增加 下一个十年的生产力增长，推动经济增长；在最好的情况下，它将驱动人类经历类似于工业革命的转型。 ...

专题：2025分析师大会：资本市场"奥斯卡"启幕 东方港湾创始人兼董事长但斌出席并发表主旨演讲。但斌谈到AI眼镜的时候，表示自己很看好这个东西。 但斌表示，比如说我们看到了mate眼镜等AI眼镜，我很看好这个东西，包括马斯克说未来五年很可能手机这个形态就没有了，如果AI Agent实现的话，类似 于现在我要订一张飞机票，从上海飞深圳，过去还要通过手机、通过携程，未来我可能只需要说一句话，通过AI眼镜说一句话，就可以订这个票，或者不 是眼镜，是耳环、是个手表、或者是一个别针。而我的Agent可以帮我实现强联系，未来的形态变化是非常大的，人工智能的这场竞争任何一家公司都不可 能放弃，都要全力以赴。 责任编辑：张恒星 炒股就看金麒麟分析师研报，权威，专业，及时，全面，助您挖掘潜力主题机会！ 11月28日，2025分析师大会举行，专家学者、券商基金私募掌舵人、首席分析师等齐聚一堂，共寻穿越周期的投资真谛。 ...