这是辛顿首次访问中国并进行演讲。77岁的辛顿长期受腰椎间盘疾病的困扰，身体欠佳的他几乎无法坐飞机。 谷歌团队曾为邀请他去英国考察DeepMind团队特地包下私人飞机，并改造了座椅。 辛顿还在WAIC开幕前一天参加了第四届人工智能国际安全对话(International Dialogues on Al Safety,IDAIS)，并 与20余名人工智能行业专家联名签署发布了《AI安全国际对话上海共识》。 在最新的演讲中，他谈及"数字智能是否会取代生物智能"的问题，并讨论了AI可能带来的挑战与潜在的应对方 法。辛顿此前已多次在公开信和演讲中指出，当前AI系统已经具备自主学习和演化的潜能。 辛顿指出，在过去60多年里，AI发展存在两种不同的范式和路径--以符号型的逻辑性范式以及以生物为基础的 范式。1985年，辛顿做了一个小模型，尝试结合这两种理论，以解释人们如何理解词汇。 "我认为，如今的大语言模型就是我当年微型语言模型的衍生。"他表示，"它们使用更多词作为输入，采用更 多层的神经元结构，由于需要处理大量模糊数字，学习特征之间也建立了更复杂的交互模式。但和我做的小模 型一样，大语言模型理解语言的方式与人类相似-- ...

Core Viewpoint - Geoffrey Hinton, known as the "Godfather of AI," expresses regret over his life's work in AI, highlighting the potential risks and consequences of AI development, urging humanity to reconsider its direction [2][4][17]. Group 1: Hinton's Background and Career - Hinton joined Google to support his son, who has learning disabilities, and has since become a prominent figure in AI, winning prestigious awards like the Nobel Prize in Physics and the Turing Award [3][13][15]. - He initially focused on neural networks, a choice that was not widely accepted at the time, but has proven to be correct as AI has advanced significantly [9][10]. Group 2: AI Risks Identified by Hinton - Hinton categorizes AI risks into short-term and long-term threats, emphasizing the need for awareness and caution [21]. - Short-term risks include a dramatic increase in cyberattacks, with a reported 12,200% rise from 2023 to 2024, facilitated by AI technologies [22][25]. - The potential for individuals with basic biological knowledge to create highly infectious and deadly viruses using AI tools is a significant concern [26]. - AI's ability to manipulate personal habits and decisions through data analysis poses a risk of creating echo chambers and deepening societal divides [29][30]. Group 3: Long-term Risks and Predictions - Hinton warns of the emergence of superintelligent AI that could surpass human intelligence within 20 years, with a predicted extinction risk of 10%-20% for humanity [32][35]. - He compares humanity's relationship with superintelligent AI to that of chickens to humans, suggesting that humans may become subservient to their creations [37]. - The potential for widespread unemployment due to AI replacing cognitive jobs is highlighted, with recent layoffs at Microsoft exemplifying this trend [39][41]. Group 4: Recommendations for the Future - Hinton suggests that individuals consider careers in trades, such as plumbing, which are less likely to be replaced by AI [43][47]. - He advocates for increased investment in AI safety research and stricter regulatory measures to manage AI development responsibly [44][54]. - The importance of fostering unique personal skills and interests is emphasized as a way to thrive in an AI-dominated future [48][49].

Core Insights - The article highlights the journey of Fei-Fei Li from her early academic achievements to her current role as CEO of a company, emphasizing her passion for starting from scratch and building innovative solutions in AI [1][2][24]. Group 1: ImageNet and AI Development - ImageNet was conceived around 18 years ago to address the lack of data in AI and machine learning, particularly in computer vision, which was essential for the development of algorithms [4][6]. - The project aimed to download 1 billion images from the internet to create a global visual classification system, which became a cornerstone for training and testing machine learning algorithms [6][7]. - The breakthrough moment for ImageNet came in 2012 with the introduction of AlexNet, which utilized convolutional neural networks (CNN) and significantly reduced the error rate in image recognition tasks [8][10]. Group 2: Vision and Future of AI - Li emphasizes the importance of spatial intelligence for achieving general artificial intelligence (AGI), arguing that without it, AGI remains incomplete [14]. - The evolution of AI has progressed from object recognition to scene understanding and now to generating 3D worlds, which presents a new set of challenges [12][16]. - The integration of language models and visual understanding is seen as a critical area for future research and application, particularly in fields like robotics and the metaverse [20][21]. Group 3: Advice for Students and Researchers - Li advises students to pursue fundamental "North Star" problems in AI that are not necessarily tied to industrial applications, as academic resources have shifted significantly [34][35]. - She encourages interdisciplinary research in AI, particularly in scientific discovery, and highlights the importance of curiosity and problem-solving in graduate studies [38][39]. - The article underscores the need for a new generation of researchers who are fearless and willing to tackle complex challenges in AI [32][33].

AGI最新判断。 作者 | 闻乐 不圆 来源 | 量子位 （ID：QbitAI） 在我看来，没有空间智能，通用人工智能就不完整。 这是"AI教母"李飞飞在最新访谈中对AGI的判断——是的，李飞飞也开始谈论AGI了。 不过她有自己的表述，从进入人工智能领域开始，她就确定了她终身奋斗的梦想： 让智能体能够讲述世界的故事 。 而这，离不开 空间智能 。 正如她本人所说： 我整个职业生涯都在追逐那些极其困难、近乎疯狂的问题。 李飞飞如今聚焦于空间智能领域——这个人工智能最艰难的领域之一。 她认为 3D世界建模 对于实现AGI至关重要，并表示： 理解三维世界、生成三维世界、推理三维世界、在三维世界中做事，是人工智能的基本问题。 她的目标是创建一个超越平面像素、跨越语言障碍、能够真正捕捉三维世界结构和空间智能的 世界模型 。 在这次对话中，她从ImageNet的起源和影响说起，讲述了AI范式转变与关键突破，并提到了3D建模面临的挑战以及空间智能的数据 缺失问题。 量子位翻译并总结了全文，让我们一起来学习李飞飞的最新认知和分享。 ImageNet为现代计算机视觉搭建数据骨架 Q：你最早创建的项目之一是2009年的Image ...

编辑 | 褚杏娟 你说得对，我们大约在 18 年前就开始构想 ImageNet 这个项目。时间过得真快！那时我还是普林斯 顿大学一年级的助理教授。那时的人工智能和机器学习领域与现在完全不同，数据非常稀缺，在计算 机视觉领域，算法几乎是行不通的。那时也没有相关产业发展，公众根本不知道"人工智能"这个概 念。 但我们当时的那群人，从 AI 的奠基人 John McCarthy，到后来像 Geoffrey Hinton 这样的研究者， 都怀着一个共同的 AI 梦想：我们真的想让机器学会思考和工作。而对我个人而言，我的梦想是让机 器能够"看见"，因为视觉是智能的重要基石。视觉智能不仅仅是感知，它更在于理解世界并在世界中 采取行动。 主持人 ：后来很久才出现了一些有前景的算法。直到 2012 年，AlexNet 出现了。那才是实现 AI 的 第二个关键因素：投入足够的计算资源。当你看到你播下的数据种子开始发挥作用，开始取得更多科 研突破时，那一刻是什么感觉？ 李飞飞 ：没错，2009 年我们在 CVPR 会议上发表了一篇很简短的论文。2009 年至 2012 年这三年 里，我们坚信数据将驱动人工智能的发展，但当时几乎看 ...

来源丨Web3天空之城（ Web3SkyCity ） 著名AI科学家李飞飞 这一次在YC创业学院峰会的最新访谈深入探讨了她的职业生涯与前瞻思考。她 分享了创办ImageNet，这一引爆深度学习革命的关键项目的幕后故事，并阐述了计算机视觉从物体 识别到场景叙事，再到她当前所专注的"空间智能"的演进路径。 李飞飞认为，理解和交互于三维世界是实现通用人工智能不可或缺的一环，并为此创立了World Labs。访谈还触及了她的个人经历，从移民少年到创办洗衣店，再到成为顶尖学者和企业家的历程， 强调了"智识上的无畏"是推动创新和个人成长的核心动力。 核心观点 ImageNet的诞生与深度学习的黎明 李飞飞 : 我整个职业生涯都在追逐那些极其困难，近乎妄想的问题。对我来说，没有空间智能的AGI 是不完整的。我想解决这个问题。我就是喜欢当企业家。忘记你过去所做的一切。忘记别人对你的看 法。埋头苦干，努力建设。那是我的舒适区。 主持人 : 所以，我非常兴奋能邀请到李飞飞博士。她在人工智能领域有着非常长的职业生涯。 我相信你们很多人都认识她，对吧？请举手。我也认识。她被称为人工智能教母。飞飞你创建的第一 个项目之一是2009年的 ...

Core Viewpoint - The article emphasizes the importance of spatial intelligence in achieving Artificial General Intelligence (AGI), as articulated by AI expert Fei-Fei Li, who believes that understanding and interacting with the 3D world is fundamental to AI development [1][4][29]. Group 1: Spatial Intelligence and AGI - Fei-Fei Li asserts that without spatial intelligence, AGI is incomplete, highlighting the necessity of creating world models that capture the structure and dynamics of the 3D world [29]. - She identifies 3D world modeling as a critical challenge for AI, stating that understanding, generating, reasoning, and acting within a 3D environment are essential problems for AI [7][29]. - The pursuit of spatial intelligence is framed as a lifelong goal for Li, who aims to develop algorithms that can narrate the stories of the world by understanding complex scenes [20][29]. Group 2: Historical Context and Breakthroughs - The article discusses the inception of ImageNet, a pivotal project initiated by Li, which aimed to create a vast dataset for training AI in visual recognition, addressing the data scarcity issue in the early days of AI [11][14]. - The success of ImageNet led to significant advancements in computer vision, particularly with the introduction of AlexNet, which utilized convolutional neural networks and marked a turning point in AI capabilities [19][22]. - Li reflects on the evolution of AI from object recognition to scene understanding, emphasizing the importance of integrating natural language with visual signals to enable AI to describe complex environments [15][20]. Group 3: Future Directions and Applications - Li expresses excitement about the potential applications of spatial intelligence in various fields, including design, architecture, gaming, and robotics, indicating a broad utility for world models [35]. - The article mentions the challenges of data acquisition for spatial intelligence, noting that while language data is abundant online, spatial data is less accessible and often resides within human cognition [33][50]. - Li's new venture, World Labs, aims to tackle these challenges by developing innovative solutions for understanding and generating 3D environments, indicating a commitment to advancing the field of AI [29][35].

Core Insights - Embodied intelligence has gained significant attention in both industry and academia, particularly in humanoid robots, which integrate perception, movement, and decision-making capabilities [1][4][30] - The development of embodied intelligence is seen as a pathway towards achieving general robotics, with ongoing discussions about the challenges and milestones that lie ahead [1][30] Group 1: Current State and Future Prospects - The industry anticipates that 2025 may mark the "year of embodied intelligence," with significant competition emerging in the multimodal and embodied intelligence sectors [3][4] - NVIDIA's CEO Jensen Huang has proclaimed that the era of general robotics has begun, outlining four stages of AI development, culminating in "physical AI," which focuses on understanding and interacting with the physical world [3][4] - Experts believe that while progress has been made, the journey towards true general robotics is still in its early stages, with many technical and conceptual hurdles remaining [31][32] Group 2: Technical Challenges and Opportunities - The current landscape of embodied intelligence is characterized by a lack of comprehensive models and algorithms, with many systems still not achieving convergence [32][33] - Key technical challenges include the integration of sensory feedback, the development of robust algorithms, and the need for advanced perception capabilities, such as tactile sensing [33][34] - The industry is witnessing a shift where many researchers from the autonomous driving sector are transitioning to embodied intelligence, leveraging their expertise in perception and interaction [15][19] Group 3: Application Scenarios - Potential application areas for embodied intelligence include home care, household services, and industrial automation, which are seen as practical and immediate needs [41] - The focus on specific vertical applications rather than general-purpose robots is emphasized, as the technology is still maturing and requires targeted development to meet real-world demands [36][41] - The integration of embodied intelligence into existing industrial systems is viewed as a promising avenue for scalability and broader adoption [39]

在为期六周的会议中，与会者进行了深入的讨论、辩论与合作，奠定了人工智能作为一个正式学科的基础。他们 尝试定义人工智能的概念、明确其研究目标，并规划可能的研究方向。会议成员探讨了一系列深刻问题，包括问 题求解、机器学习和符号推理等。 这次关键性的会议不仅开启了之后数十年人工智能研究与创新的大门，也凝聚了一个充满信念的学术群体——他 们相信机器有能力复制人类的认知能力。达特茅斯会议的深远影响在于，它确立了人工智能作为一门学科和实践 领域的地位，推动人类迈向一个智能机器与人类协作的未来。 #2 感知机（1957年） 2025年，人工智能已经不再只是前沿科技圈的热词，而是真真正正地走进了我们的日常：生成图像、写代码、自 动驾驶、医疗诊断……几乎每个行业都在讨论 AI，拥抱 AI。 但今天的大模型奇点不是一夜之间到来的，它背后是一条充满突破、争议、冷寂与复兴交织的进化之路。从1956 年达特茅斯会议开始，到如今数千亿参数模型引发的全球技术竞赛，AI的发展史是一部关于人类如何模拟、拓 展，乃至重新定义智能的故事。 本文将带你一起回顾这条历程中的10个关键的历史性时刻，帮你理清人工智能是如何一步步从纸上设想，走向今 天这场 ...

图片来源：姚顺雨 个人博客 Z Highlights 姚顺 雨 是斯坦福大学自然语言处理研究员， OpenAI 员工 ，专注于强化学习与语言模型融合研究。本文译自姚顺 雨 于 2025 年 4 月 10 日发布的英文博客 《 The Second Half 》，内容整合了他在斯坦福课程 CS224N 及哥伦比亚大学前沿论坛的核心演讲观点。 我们正处于AI的中场阶段。几十年来，AI的核心一直是开发新的训练方法和模型。这条路奏效了：从在国际象棋和围棋上击败世界冠军，到在SAT和律师 资格考试中超越大多数人类，甚至赢得IMO和IOI金牌。 这些载入史册的里程碑——DeepBlue、AlphaGo、GPT-4，以及o系列模型——背后是AI方法的根 本性创新：搜索、深度强化学习、模型规模化，以及推理。 一切都在不断变得更好。 那现在究竟发生了什么变化？用三个词概括：强化学习终于奏效了。更准确地说：强化学习终于具备了泛化能力。经历了多次重大的绕行与一系列里程碑 之后，我们终于找到了一个通用的有效配方，能够利用语言和推理解决各种各样的强化学习任务。哪怕是在一年前，如果你告诉大多数AI研究人员，一个 统一的方法可以同时解决软 ...