来源 | youtu.be/0fXGtQoJgNo 出品丨AI 科技大本营（ID：rgznai10 0） 在 AI 圈，"深度学习三巨头"有着截然不同的晚年人设：Yann LeCun 是那个永远愤怒的乐观派，在推特上怼天怼地，坚信 AI 只是工具；Geoffrey Hinton 是那个突然觉醒的悲观派，为了发出警告不惜从谷歌辞职。 而 Yoshua Bengio，他是站在中间，带着一种近乎苦行僧般的冷静与忧虑。 作为图灵奖得主、深度学习三巨头之一，他的一生大半时间都在与数学公式和神经网络打交道。在很长一段时间里，他相信 "更聪明的机器"等于"更美 好的人类未来" 。这曾是他坚不可摧的信仰，也是他作为一名纯粹科学家的动力源泉。 编译 | 王启隆 但在 2023 年的某个时刻，这种信仰似乎崩塌了。 最新的一场在达沃斯论坛的采访，面对镜头的 Bengio 显得有些疲惫，但异常诚恳。 他总是频繁地提起他的孙子。 他不再像几年前那样兴奋地谈论下一个 SOTA（当前最佳）模型，而是像一个刚刚窥探到未来恐怖一角的预言家，试图用最温和的语言，讲出最令人背 脊发凉的现实。 他谈到了一个非常具体、却让人细思极恐的现象： Syco ...

野村发布研报称，金山云作为小米集团-W生态圈中唯一的人工智能云基础设施供应商，有望持续受惠 于小米锐意发展大语言模型（LLM）的决心，潜在的H200晶片进口或能缓解公司在2026财年面临的短 缺问题。该行认为，公司受惠于持续强劲的LLM训练需求，以及消耗更多Token的智能体等应用所带动 的推理需求增长所推动，相信今年中国AI投资周期仍在加速，相应将金山云2025至2027财年收入预测 上调1.4%至8.9%。 责任编辑：卢昱君 热点栏目 自选股 数据中心 行情中心 资金流向 模拟交易 客户端 金山云（03896）早盘股价上涨7.86%，报7港元，成交额3.10亿港元。 近期，谷歌云、亚马逊AWS等云服务商相继涨价。华创证券认为，云厂商涨价与AI开支扩张共同重塑 AIDC产业逻辑。涨价潮提升算力资产回报预期，需求爆发则抬升产业规模天花板，推动AIDC从重资产 行业升级为具备高壁垒、高确定性的核心基础设施赛道。具备技术迭代能力、资源整合效率的头部厂商 有望持续受益于结构性红利。 金山云（03896）早盘股价上涨7.86%，报7港元，成交额3.10亿港元。 近期，谷歌云、亚马逊AWS等云服务商相继涨价。华创证券认 ...

Core Viewpoint - The article discusses the shift from the first generation of AI models, primarily based on "next word prediction," to a second generation focused on "world modeling" or "predicting the next physical state," highlighting the limitations of current AI applications in the physical world [4][8]. Group 1: Current AI Paradigms - The first generation of AI models, exemplified by large language models (LLMs), has achieved significant success but struggles with real-world applications [4]. - The second generation, as proposed by Jim Fan, emphasizes world modeling, which involves predicting reasonable physical states under specific actions, marking a transformative shift in AI development [8]. Group 2: World Modeling Definition and Implications - World modeling is defined as predicting the next physical state based on specific actions, with video generation models serving as a practical example [8]. - The article anticipates that 2026 will be a pivotal year for large world models (LWMs) in robotics and multimodal AI, establishing a real foundation for future advancements [8]. Group 3: Comparison of AI Models - Visual language models (VLMs) are described as "language-first," where visual information is secondary, leading to a disparity in physical understanding compared to LLMs [9]. - The design of VLA (visual-language-action) models prioritizes language over physical interactions, resulting in inefficiencies in physical AI applications [10]. Group 4: Biological Insights and Future Directions - The article draws parallels between human cognitive processing and AI, noting that a significant portion of the human brain is dedicated to visual processing, which is crucial for physical interaction [11]. - The emergence of world modeling is seen as a response to the limitations of current AI paradigms, with potential for new types of reasoning and simulation that do not rely on language [12]. Group 5: Challenges and Future Research - The article raises questions about the future of AI, including how to decode action instructions and whether pixel reconstruction is the optimal goal for AI development [13]. - It emphasizes the need for further exploration in the field, suggesting a return to fundamental research principles as the industry seeks to advance towards a "GPT-3 moment" in robotics [13].

2026.01. 29 本文字数：1766，阅读时长大约3分钟 作者 | 第一财经 何涛 以大模型为代表的人工智能正从实验室走向实际应用，成为引领新一轮科技革命与产业革命的战略性 力量。在此背景下，智能经济30人论坛日前在深圳举行。 来自中国（深圳）综合开发研究院、清华大学、北京大学、中国信息通信研究院等机构的专家学者及 产业界代表，围绕AI对经济、就业、治理带来的深远影响展开深度对话。 专家学者们普遍认为，AI将对产业、社会带来多层次的影响，历史规律表明技术替代终将催生新需 求与新职业；大模型正从语言交互向"能说会做"升级；需通过完善治理平衡创新与风险，依托制度优 势化解技术冲击带来的社会矛盾。 樊纲认为，技术冲击的同时必然催生新产业与新需求，这正是人类经济发展的历史规律。尽管当前无 法预判传统产业被颠覆的规模，但按照历史路径，旧产业消亡与新产业诞生始终相伴相生，部分产业 对人力需求下降的同时，新的就业需求会在其他领域增长。 智能经济30人论坛日前在深圳举行 图/第一财经 清华大学人工智能国际治理研究院院长薛澜深入技术底层，描绘了智能经济的演进路径。他认为，当 前主流大语言模型（LLM）的核心能力仍停留在" ...

以大模型为代表的人工智能正从实验室走向实际应用，成为引领新一轮科技革命与产业革命的战略性力 量。智能经济30人论坛近日在深圳举行。专家学者们普遍认为，AI将对产业、社会带来多层次的影 响，但历史规律表明技术替代终将催生新需求与新职业；大模型正从语言交互向"能说会做"升级；同时 需通过完善治理平衡创新与风险，依托制度优势化解技术冲击带来的社会矛盾。 为把握这一历史性机遇，国家已全面部署"人工智能+"行动，旨在加快构建"人机协同、跨界融合、共创 分享"的智能经济和智能社会新形态。智能经济30人论坛近日在深圳举行，智能经济30人论坛由中国 （深圳）综合开发研究院联合发起，旨在汇聚全国人工智能、数字经济领域的专家学者与企业家，搭建 一个非官方、公益性、开放性的智库交流与合作平台。论坛上，来自中国（深圳）综合开发研究院、清 华大学、北京大学、中国信息通信研究院的专家学者以及产业界代表齐聚一堂，围绕AI对经济、就 业、治理带来的深远影响展开深度对话，为推动"十五五"时期智能经济高质量发展、打造中国式现代化 新引擎贡献智慧与力量。 "AI正在创造新产业，正在产生新需求" "科技进步首先会对经济产生一系列冲击，创新的本质就是 ...

中国应善用在应用能力方面的优势，把握智能经济机遇。 以大模型为代表的人工智能正从实验室走向实际应用，成为引领新一轮科技革命与产业革命的战略性力量。在此背景下，智能经济30人论坛日前在深圳举 行。 来自中国（深圳）综合开发研究院、清华大学、北京大学、中国信息通信研究院等机构的专家学者及产业界代表，围绕AI对经济、就业、治理带来的深远 影响展开深度对话。 专家学者们普遍认为，AI将对产业、社会带来多层次的影响，历史规律表明技术替代终将催生新需求与新职业；大模型正从语言交互向"能说会做"升级；需 通过完善治理平衡创新与风险，依托制度优势化解技术冲击带来的社会矛盾。 第一个层面是产业内部的冲击。部分传统企业将被淘汰，掌握新技术、具备更高效率的新企业会对老牌企业形成挤压，导致老牌企业的市场份额持续萎缩， 新企业份额则不断攀升；第二个层面是产业结构的重构，表现为新兴产业对传统产业的替代，推动产业体系发生根本性变革；第三个层面是社会层面的冲 击，反映在经济领域是收入差距的变化。 樊纲进一步阐释，随着工作时长缩短，人类闲暇时间大幅增加，由此催生旅游、健身、宠物经济等精神文化与体验类消费需求。这些领域高度依赖人际互 动、情感共 ...

Core Insights - The forum highlighted the transformative impact of AI, particularly large models, on the economy, employment, and governance, emphasizing its role as a strategic force in the new technological and industrial revolution [1][21]. Group 1: Economic Impact of AI - Experts agree that AI will have multi-layered effects on industries and society, with historical patterns indicating that technological replacement will ultimately create new demands and jobs [3][23]. - The economic disruption caused by technological advancement is characterized by three levels: internal industry shocks, restructuring of industry, and social impacts reflected in income disparity [3][23]. - Historical examples, such as the post-Napoleonic era, illustrate that technological progress can lead to increased employment and improved living standards, despite initial fears of job loss [4][24]. Group 2: AI Development and Governance - The evolution of AI is moving from merely predicting text to predicting real-world states, which will enhance its functional value [7][26]. - The unification of technical architecture and multi-modal integration is expected to lower the costs of AI application, transitioning it from a specialized tool to a universal infrastructure [7][26]. - AI risks are categorized into three types: malicious misuse, inherent technical flaws, and systemic social risks, necessitating careful governance to mitigate these challenges [8][27]. Group 3: China's Competitive Advantage in AI - China is positioned to lead in the global smart economy by leveraging its unique ability to convert technology applications into innovations, emphasizing the importance of data, computing power, and large models [10][29]. - The country’s unified market structure provides a significant advantage in achieving reverse innovation, which is crucial for economic development and technological advancement [10][29]. - The need for a cohesive approach among major tech companies is highlighted to avoid fragmentation and redundancy in the AI ecosystem [10][29]. Group 4: Opportunities and Challenges in Smart Economy - The rapid development of AI presents governance challenges that require proactive measures to ensure compliance and safety while promoting high-quality development [19][38]. - The importance of maintaining a balance between technological advancement and social equity is emphasized, particularly in addressing the impacts on employment and income distribution [19][38]. - The phenomenon of "investment waves" in the AI sector raises concerns about potential market bubbles, yet historical precedents suggest resilience in the face of such challenges [20][39].

Core Viewpoint - The article emphasizes that artificial intelligence, represented by large models, is transitioning from laboratory experiments to practical applications, becoming a strategic force driving a new wave of technological and industrial revolutions [1]. Group 1: Impact of AI on Economy and Society - AI is expected to have multi-layered impacts on industries and society, with historical patterns indicating that technological replacement will ultimately create new demands and jobs [1][3]. - The economic disruption caused by technological advancements occurs in three layers: internal industry shocks, restructuring of industry, and social impacts reflected in income disparity [4]. - Historical examples, such as the Luddites' protests during the Industrial Revolution, illustrate that while new technologies may threaten existing jobs, they also lead to overall job growth and improved living standards [5]. Group 2: Evolution of Intelligent Economy - The current capabilities of mainstream large language models (LLMs) are limited to text prediction, but future breakthroughs will involve predicting real-world states, enhancing their functional value [7]. - The unification of technical architecture and native multimodal integration is expected to significantly lower the costs of AI applications, transforming them from specialized tools to universal infrastructure [7]. - The development of AI will rely on effective computational resource management, with the potential for widespread access to computing power akin to utilities like water and electricity [7]. Group 3: Balancing Innovation and Governance - The rapid advancement of AI technology raises governance challenges, with 2026 seen as a critical year for achieving commercial viability [9]. - AI risks are categorized into three types: malicious misuse, inherent technical flaws, and systemic social risks that could affect employment and income distribution [9]. - The need for a collaborative approach among academia, industry, and government is emphasized to ensure the safe and effective development of AI technologies [10]. Group 4: Strategic Positioning in Global AI Competition - China is positioned to lead in the global intelligent economy, with major companies like Huawei, Tencent, and Baidu having substantial technological foundations [11]. - The article suggests that a unified national strategy is necessary to avoid fragmentation and redundancy in AI development, promoting collaboration among enterprises [11]. - The focus should be on application transformation and scaling, leveraging China's strong application capabilities to enhance its competitive edge in the global market [11].

Market Overview - The Hong Kong stock market experienced a decline, with the Hang Seng Index closing at 23,831 points, down 145 points or 0.61% [1] - The market turnover decreased to 82.799 billion, with net inflows of 484 million in the Hong Kong Stock Connect [1] - The US stock market also saw slight declines, with the Dow Jones down 0.17% and the S&P 500 down 0.06% [2] Key Insights - The report emphasizes the importance of "technological self-reliance" and AI applications as core themes for future growth in the Hong Kong stock market, with leading companies in these sectors expected to benefit in the medium to long term [3] - The semiconductor sector showed strong performance, with Huahong Semiconductor rising 6.3% and 7.4% on consecutive days, reaching a historical high [3] - The report suggests continued focus on sectors supported by government policies, including AI, semiconductors, and consumer services, as well as undervalued state-owned enterprises with high dividends [3] Company Highlights - Li Ning Company, a leading sports brand in China, reported a revenue of 14.817 billion for the first half of 2025, a year-on-year increase of 3.3% [10] - The company's gross margin was 50%, slightly down by 0.4 percentage points due to increased promotional competition [10] - Li Ning's net profit was 1.737 billion, a decrease of 11% year-on-year, with a net profit margin of 11.7% [10] - The company is expected to enhance its brand presence in professional sports through a partnership with the Chinese Olympic Committee starting in May 2025 [10] Industry Trends - The report highlights the rapid growth of AI in China, with the usage of large language models (LLMs) increasing by 460% in two months, capturing a market share of 13% globally [9] - The National Energy Administration announced that China's total electricity consumption is expected to exceed 10 trillion kilowatt-hours by 2025, indicating strong growth in energy demand [9] - The report recommends focusing on green power companies with stable growth and high dividends, such as Longyuan Power and CGN New Energy, which are currently undervalued [9]

Group 1 - The core theme of the events is the exploration of human agency in the context of AI, focusing on how to preserve meaningful human decision-making rights amidst the evolving landscape of artificial intelligence [2][4] - The first seminar titled "The Right to Work, Learn, Own & Choose" aims to integrate the technical AI community with AI governance to promote respect for human agency and protect rights related to work, learning, ownership, and choice [2][4] - The event features prominent speakers from various institutions, including Ashok Goel from Georgia Tech and Jungpil Hahn from the National University of Singapore [4] Group 2 - The second seminar, "Agentic AI meets Autonomous Agents and Multiagent Systems," focuses on advancements in intelligent agents based on large language models (LLMs) and the lessons learned in building and deploying these systems [11][13] - This seminar emphasizes the transition of modern "Agentic AI" systems from demonstrations to practical deployment, requiring capabilities in long-term planning, reliable tool usage, and robust interaction with humans and environments [13][14] - Notable speakers include Leslie Kaelbling from MIT and Bo Li from the University of Illinois at Urbana-Champaign, contributing to discussions on the long-term challenges in robotics and multi-agent systems [17]