以下文章来源于芒格书院 ，作者芒格书院小芒 芒格书院 . 由资深出版人施宏俊先生创立，定位于为终身学习者提供学习和思考的知识资源，推动认知升级和思想 分享。 接下来，我将介绍未来5年值得密切关注且可能具备投资价值的几大核心主题： 1.科技领域。第一个毫无疑问是科技领域，但这并不容易，因为科技类股票当前估值普遍较高。 2.金属与采矿业。这一领域可能有些不同寻常，我认为将来可能出现金属短缺，因此相关企业有望迎来爆发 式增长。 3."激情投资"。那些不涉及技术迭代、供给无法扩张的资产，在创新活动极为密集的时期，其价格往往会大 幅上涨。 9月20日下午，丹麦作家、企业家、投资人拉斯·特维德（Lars Tvede）携新作《超智能与未来》现身上海， 在芒格书院与中信出版作家演讲局联合组织的活动中，与书院会员们深度讨论了未来五年最具前景的投资 方向。 拥有横跨金融、科技创业与未来学领域的"跨界头脑"，特维德的经历堪称传奇：他不仅是Supertrends AG创 新地图公司、 Atlas Global Macro对冲基金、Fiftyfive Capital风投基金的联合创始人，更凭借《逃不开的经济 周期》《金融心理学》等著作 ...

AI竞赛，各国角逐 AI的脚步始于硅谷，但并未停滞在硅谷，随着技术和技术人才的迁移，全球AI竞赛已经进入白热化阶 段，而在这场科技与资本的角逐中，美国、中国、欧洲乃至中东，分别正在AI竞争之中扮演着何种角 色，是一个值得探讨的问题。 周炜认为，在AI领域未来只有中美两国是超级大国。回顾过去20年中美在投资和创业方面的变化，存 在一个有趣的现象，双方总是容易低估对方的优势。 在硅谷，许多人对医疗AI、图像识别AI等领域非常兴奋，甚至纷纷创办初创企业。然而，他们并未意 识到，中国在五年前就已经在这些领域有所突破，并开始应用大语言模型（LLM）实现了商业化。例 如，数坤科已在AI医疗图像评级领域成为全球领先企业。 美国可能低估了中国在应用层面的进展，而中国则往往低估了美国在基础模型方面的领先地位。虽然有 些人认为两国之间的差距仅为3-6个月，可是，实际上差距依然显著。 今天，全球AI赛场的竞争，不仅仅是中美两国的较量，新的力量正悄然崛起。而在未来的3-5年，谁能 在垂直领域找到那个隐藏的"金矿"？谁又能成为下一个改变世界的AI巨头？是时候重新审视这个时代的 机遇与挑战了。 美西时间2025年9月27–28日，钛媒体 ...

机器之心报道 机器之心编辑部 推理训练促使大语言模型（LLM）生成长思维链（long CoT），这在某些方面有助于它们探索解决策略并进行自我检查。虽然这种方式提高了准确性，但也增加了 上下文长度、token / 计算成本和答案延迟。 因此，问题来了：当前的模型能否利用其元认知能力，在这一帕累托前沿上提供其他组合策略，例如在降低上下文长度和 / 或延迟的情况下提高准确性？ 带着这一问题，Meta 超级智能实验室、伦敦大学学院、Mila、Anthropic 等机构的研究者进行了探索。从抽象层面来看，他们将 LLM 视为其「思维」的改进操作 符，实现一系列可能的策略。 研究者探究了一种推理方法家族 —— 并行 - 蒸馏 - 精炼（Parallel-Distill-Refine, PDR） ，该方法包含以下步骤：(i) 并行生成多样化草稿；(ii) 将其蒸馏成一个有 限的文本工作区；(iii) 在此工作区的基础上进行精炼，生成的输出将作为下一轮的种子。重要的是，通过调整并行度，PDR 能够控制上下文长度（从而控制计算 成本），并且上下文长度不再与生成 token 的总数混淆。 根据当前模型在 PDR 实例中的应用，它 ...

Core Insights - Large language models (LLMs) are reaching a performance bottleneck despite significant investments and data usage, leading to concerns about the sustainability of returns on investment [1][2][5] - Global spending on artificial intelligence (AI) is projected to reach nearly $1.5 trillion by 2025, a 50% increase from 2024, and could rise to $2 trillion by 2026, marking a further 37% increase [1][4] - Major tech companies are heavily investing in LLMs, but there is growing skepticism regarding the economic returns from these investments [1][4] Investment Trends - The competition among major tech firms like Google, Amazon, Meta, Microsoft, and OpenAI in LLM development is intensifying, with costs potentially reaching hundreds of billions [4][5] - In 2023, leading companies generated approximately $1 billion in public sales from LLM products, expected to grow to $4 billion in 2024 and potentially reach between $235 billion and $244 billion by 2025, although most of this revenue will be reinvested into infrastructure [4][5] - The UNCTAD forecasts that the AI market could reach $4.8 trillion by 2033, while CMR estimates global AI revenue could hit $3 trillion by then [4] Economic Viability - There is a significant gap between infrastructure investment and end-user software licensing revenue, raising questions about the sustainability of current investment levels [5][6] - The expectation that all major LLM companies will emerge as winners is based on the assumption that their core products are nearing the end of their useful lifecycle, which may not hold true for all [5][6] - The high training costs of new LLMs are increasing exponentially, with current costs reaching hundreds of millions, while performance improvements are becoming marginal [6] Market Sustainability - Deutsche Bank has raised concerns that the current AI investment boom may not be sustainable due to the difficulty in maintaining exponential growth in tech spending [7] - Bain & Company reports that AI may not generate sufficient revenue to support the required computational power, predicting a $800 billion funding gap by 2030 [7] - BCA Research warns of a potential shift from a shortage to an oversupply of computing resources, which could lead to a decline in capital expenditures [7] Long-term Outlook - Goldman Sachs remains optimistic, projecting that AI will significantly boost GDP growth, contributing approximately 0.4 percentage points annually in the coming years, with a cumulative potential of 1.5% growth in the long term [7] - UBS emphasizes that AI investment will be a key growth driver for investment portfolios in the medium to long term, with ongoing progress in monetizing AI solutions [7][8]

作者 | 汤拯 2025.09. 17 过去30年，互联网和智能手机迅速发展，成为现代社会的关键基础设施，深刻改变了人类的生活方式。然而，欧盟却未能孕育出可与美国硅谷或东亚 科技巨头比肩的新兴企业，与互联网移动时代屡屡失之交臂。 当前，人工智能（AI）正以惊人的速度成为下一代"数字基础设施"，其重要性可媲美当年的互联网和智能终端。这一次，欧盟各界弥漫着一股危机感。 从重新审视产业政策到加大研发投入，越来越多欧洲人意识到，AI时代很可能是重塑这片大陆竞争力的最后窗口期。如果再度错失良机，欧盟在未来 的产业与科技角逐中或将被彻底边缘化。 本文字数：3858，阅读时长大约6分钟 AI将成为未来社会的基础设施 首先，欧洲资本市场整体偏保守且碎片化，高风险创新项目难以获得持续大规模的融资。欧洲风险投资的规模和激进程度远不及美国。数据显示， 2018年至2022年欧洲深度科技创业公司仅吸引到约325亿欧元投资，同期美国这一数字超过1200亿欧元。欧洲社会和资本对失败的容忍度较低，更青 睐稳健经营，这种保守文化使很多创业者不敢或无力采取激进扩张策略。 其次，欧洲大型企业对初创公司的带动作用不足。统计显示，只有约12%的欧洲 ...

日本在左右AI性能的数据方面拥有高质量的资源，将利用这种优势推进研发日本国产AI…… 日本政府制定的人工智能（AI）开发和应用战略方案已经明确。日本在左右AI性能的数据方面 拥有高质量的资源。将利用这种优势推进研发日本国产AI。考虑到数据中心及半导体等相关投 资的需求，将通过增加预算以及放宽规定等提供支援。 日本政府近期将召开AI战略本部（本部长为首相石破茂）的首次会议，对AI基本计划的核心方 案展开讨论。该计划是依据9月1日全面施行的《AI推进法》制定的国家战略。目标是在2025 年年内敲定。 不仅是美国和中国，包括新兴市场国家在内各国都在竞争AI开发，日本则起步较晚。此次的计 划方案强调，由于AI技术日新月异竞争环境瞬息万变，因此"日本仍有迎头赶上的机会"。方 案将以下四项列为核心内容：（1）推进应用（2）强化开发（3）主导治理（4）推动面向AI 社会的持续性变革。 方案指出，日本在高质量AI数据及数据可靠性方面比其他国家有优势，并将其描述为"一条取 胜的道路"。方案还列举了从日本国内外吸纳具备专业知识的开发人才、完善和扩充日语数据 等具体措施。 目前，因篡改学习数据导致AI错误运行的"数据投毒（Data ...

神译局是36氪旗下编译团队，关注科技、商业、职场、生活等领域，重点介绍国外的新技术、新观点、新风向。 编者按：感觉良好、追求地位的学习，往往是无效的。真正的成长，始于直面"我很愚蠢"的痛苦。文章来自编译。 心态 学习 vs. 假装学习 工具 心态 学习 vs. 假装学习 如果我们总是放任自己，就会永远等到这样或那样的干扰结束，然后才能真正开始干活。唯一能取得巨大成就的，是那些对知识极度渴望，以至 于在条件依然不好的时候仍不懈追求的人。因为你永远也等不到条件好的时候。——C.S. 刘易斯 学习大量真正有用的信息极其重要。这也许是利用时间可做的最重要的一件事，它能改善生活，让你更有能力在世界上有所作为。哪怕你就读于顶尖学府或 从事对认知能力要求很高的工作，也很容易在不知不觉中让大量时间流逝，却没有真正学到很多东西。真正持续的学习需要高度的警觉、自我反省、坚韧的 心态、有效的策略和工具，以及问责体系。 根据我的经验，无论在哪里，人们都非常乐于见到那些真正花了很多时间和心思去学习某件事并进行深入思考的人，而不是那些只学习表面知识、懂得在适 当的时候说出正确"暗号"的人。我感觉有些人身上散发着一种独特的、充满活力的温暖 ...

Core Insights - The latest paper by Fei-Fei Li delineates the boundaries and establishes paradigms for the currently trending field of Agents, with major players like Google, OpenAI, and Microsoft aligning their strategies with the proposed capability stack [1][4] - The paper introduces a comprehensive cognitive loop architecture that encompasses perception, cognition, action, learning, and memory, forming a dynamic iterative system for intelligent agents, which is not only a technological integration but also a systematic vision for the future of AGI [1][5] - Large models are identified as the core engine driving Agents, while environmental interaction is crucial for addressing issues of hallucination and bias, emphasizing the need for real or simulated feedback to calibrate reality and incorporate ethical and safety mechanisms [1][3][11] Summary by Sections 1. Agent AI's Core: A New Cognitive Architecture - The paper presents a novel Agent AI paradigm that is a forward-thinking consideration of the development path for AGI, rather than a mere assembly of existing technologies [5] - It defines five core modules: Environment and Perception, Cognition, Action, Learning, and Memory, which together create a complete and interactive cognitive loop for intelligent agents [5][10] 2. How Large Models Drive Agent AI - The framework of Agent AI is made possible by the maturity of large foundational models, particularly LLMs and VLMs, which serve as the basis for the cognitive capabilities of Agents [11][12] - LLMs and VLMs have internalized vast amounts of common and specialized knowledge, enabling Agents to perform zero-shot planning effectively [12] - The paper highlights the challenge of "hallucination," where models may generate inaccurate content, and proposes environmental interaction as a key anchor to mitigate this issue [13] 3. Application Potential of Agent AI - The paper explores the significant application potential of Agent AI in three cutting-edge fields: gaming, robotics, and healthcare [14][19] - In gaming, Agent AI can transform NPC behavior, allowing for meaningful interactions and dynamic adjustments based on player actions, enhancing immersion [15] - In robotics, Agent AI enables users to issue commands in natural language, allowing robots to autonomously plan and execute complex tasks [17] - In healthcare, Agent AI can serve as a medical chatbot for preliminary consultations and provide diagnostic suggestions, particularly in resource-limited settings [19][21] 4. Conclusion - The paper acknowledges that Agent AI is still in its early stages and faces challenges in achieving deep integration across modalities and domains [22] - It emphasizes the need for standardized evaluation metrics to guide development and measure technological progress in the field [22]

Core Viewpoint - The article discusses the newly established policies regarding the use of large language models (LLMs) in academic research, particularly in the context of the ICLR conference, aiming to ensure academic integrity and mitigate risks associated with LLMs [2][4][14]. Group 1: ICLR Conference Policies - ICLR 2026 has introduced specific policies for the use of LLMs, which are based on the conference's ethical guidelines [2][4]. - The conference received 11,565 submissions in 2025, with an acceptance rate of 32.08% [2]. - The policies emphasize that any use of LLMs must be disclosed, and authors and reviewers are ultimately responsible for their contributions [6][7]. Group 2: Specific Policy Applications - Authors must disclose the use of LLMs in writing assistance, and they are responsible for all content, including any errors generated by the LLM [9]. - When LLMs are used for research ideas or data analysis, authors must verify the validity and accuracy of the contributions made by the LLM [9]. - Reviewers must also disclose their use of LLMs in writing reviews and are responsible for maintaining the confidentiality of submitted papers [11]. Group 3: Prohibited Practices - The article highlights the prohibition of "prompt injection," where authors manipulate the review process through hidden prompts, which is considered collusion and a serious academic misconduct [12]. - Violations of these policies can lead to severe consequences, including desk rejection of submissions [7]. Group 4: Broader Context - The article notes that ICLR is not alone in implementing such policies; other major conferences like NeurIPS and ICML have also established guidelines for LLM usage [13][15]. - The increasing reliance on LLMs raises concerns about academic integrity, including issues like false citations and plagiarism, prompting the need for clear guidelines [14].

Core Viewpoint - The belief among many AI entrepreneurs that model costs will decrease significantly is challenged by the reality that only older models see such reductions, while the best models maintain stable costs, impacting business models in the AI sector [6][20]. Group 1: Cost Dynamics - The cost of models like GPT-3.5 has decreased to one-tenth of its previous price, yet profit margins have worsened, indicating a disconnect between cost reduction and market demand for the best models [14][20]. - Market demand consistently shifts to the latest state-of-the-art models, leading to a scenario where older, cheaper models are largely ignored [15][16]. - The expectation that costs will drop significantly while maintaining high-quality service is flawed, as the best models' costs remain relatively unchanged [20][21]. Group 2: Token Consumption - The token consumption for tasks has increased dramatically, with AI models now requiring significantly more tokens for operations than before, leading to higher operational costs [24][26]. - Predictions suggest that as AI capabilities improve, the cost of running complex tasks will escalate, potentially reaching $72 per session by 2027, which is unsustainable under current subscription models [26][34]. - The increase in token consumption is likened to a situation where improved efficiency leads to higher overall resource usage, creating a liquidity squeeze for companies relying on fixed-rate subscriptions [27][34]. Group 3: Business Model Challenges - Companies are aware that usage-based pricing could alleviate financial pressures but hesitate to implement it due to competitive dynamics where fixed-rate models dominate [35][36]. - The industry faces a dilemma: adopting usage-based pricing could lead to stagnation in growth, as consumers prefer flat-rate subscriptions despite the potential for unexpected costs [39]. - Successful companies in the AI space are exploring alternative business models, such as vertical integration and using AI as a lead-in for other services, to capture value beyond just model usage [40][42]. Group 4: Future Outlook - The article emphasizes the need for AI startups to rethink their strategies in light of the evolving landscape, suggesting that merely relying on the expectation of future cost reductions is insufficient for sustainable growth [44][45]. - The concept of becoming a "new cloud vendor" is proposed as a potential path forward, focusing on integrating AI capabilities with broader service offerings [45].