作者 | 悦木Ivy 编辑 | 自动驾驶之心 原文链接： https://www.zhihu.com/question/637090848/answer/3363233578 点击下方 卡片 ，关注" 自动驾驶之心 "公众号 戳我-> 领取 自动驾驶近15个 方向 学习 路线 >>自动驾驶前沿信息获取 → 自动驾驶之心知识星球 本文只做学术分享，如有侵权，联系删文 先看下Chat和Agent的差别。 Chat（聊天）：纯粹的Chat，像是一个主要由"大脑和嘴"构成的智能体，专注于信息处理和语言交流。比如 ChatGPT这样的系统，它能够理解用户的查询，给出有用和连贯的回答，但它本身不直接执行任务。 Agent（代理）：像一个具有"手、脚"的智能体，它能够进行思考、决策，并且能执行具体的任务。 我们可以简单粗暴的理解为， Chat强调的是"说"，Agent强调的是"做"。 要你你选哪个？ 这么讲吧， 但凡人类对于躺平的美好追求还在，"让机器替我干活"的念头就不可能熄灭。 自ChatGPT发布后，从plugin的推出，到Function Calling再到Assistant API的面世，OpenAI这一系列动作 ...

wo[Table_Title] Research Report 13 May 2025 中国电子 China (Overseas) Technology 红杉 AI 峰会六大关键议题解读（3）：智能体觉醒，AI 从任务执行者迈向经济行为主体 Interpretation of the Six Key Topics of the Sequoia AI Summit (3): The Awakening of Intelligent Agents, AI's Transition from a Task Executor to an Economic Behavior Subject 姚书桥 Barney Yao 吴叡霖 Louis Ng barney.sq.yao@htisec.com louis.yl.ng@htisec.com [Table_yemei1] Flash Analysis [Table_summary] 事件： 在 2025 年红杉 AI 峰会上，"智能体经济"成为高频讨论话题。目前 AI 智能体正在从单一任务执行者演进为具备身 份、意图和协作能力的经济行为主体，开启人机共生的新阶段。 点评： ...

Core Insights - The emergence of the MCP (Model Context Protocol) is reshaping the AI industry, promoting a more egalitarian approach to technology and focusing on the effectiveness of AI products rather than the underlying models [1][3][10] - The global AI Agent market is projected to grow significantly, from $5.29 billion in 2024 to $216.8 billion by 2035, with a compound annual growth rate (CAGR) of 40.15% [3][10] - Major tech companies are increasingly adopting the MCP protocol, which aims to standardize interactions between AI models and external tools, akin to foundational internet protocols like HTTP [5][9] Industry Dynamics - The AI industry is experiencing a shift from traditional applications to AI Agents and terminal devices, driven by advancements in technologies such as natural language processing and machine learning [10] - The MCP protocol is seen as a solution to the complexities faced by developers in integrating various tools and models, highlighting a clear market demand for standardized protocols [8][9] - Companies like OpenAI, Tencent, and Alibaba are actively supporting the MCP protocol, indicating a collective movement towards a unified framework in the AI ecosystem [6][7][5] Competitive Landscape - The competition between MCP and Google's A2A (Agent2Agent) protocol illustrates the ongoing struggle for dominance in the AI Agent space, with both protocols seeking developer and enterprise support [7][9] - The industry is still in its early stages, with ongoing optimization of the MCP protocol and a focus on addressing the challenges of model consistency and interoperability [10][11] - The potential for collaboration between different protocols exists, particularly given the investment relationships among key players like Google and Anthropic [7][9] Future Outlook - The development of AI Agents is expected to lower the barriers for consumers in using software and smart hardware, with a focus on enhancing user experience through intuitive interactions [11] - The evolution of the MCP protocol is anticipated to address critical issues such as authentication and discovery mechanisms, which are essential for commercial applications [12] - As the market matures, the demand for effective applications rather than mere traffic aggregation will drive the future of the MCP marketplace [12]

不久前，我们发布了一篇文章《MCP，媲美TCP/IP？》，反响较好。其中，有一个读者的留言，启发了我们： MCP协议、Agent协作、能力注册，AI Agent大潮来袭，普通人如何"上车"？ 是啊，在这样一个快速变化的时代，普通人要怎么做，才能不被甩下时代的列车呢？ 大模型爆火之后，Prompt成了AI圈的"香饽饽"。 人人都在学提示词技巧，调教DeepSeek、ChatGPT、掌握LangChain、写出像样的函数调用结构，成了AI从业者的新标配。无论是运营、产品、工程师还 是自由职业者，都在追赶这波热潮。 但你是否也开始察觉——焦虑感，正在悄悄蔓延。 "我学了这么多Prompt技巧，怎么感觉用处越来越有限？" | : 突™ 山东 1小时间 | | --- | | " 普通AI行业从业者在这个变革中从何处入口以便后期能更加适配行业工作? 如果现在不学习 | | Prompt技巧，不明确相关任务流程，是否会正式来临面临落伍。 | | 现在学习的这些搭建技巧能否适配后期行业内容? | | 底层普通AI从业人员，好比现在的基础IT运维人员现在应该从哪些方面的技术去做学习铺垫? | | 作者赞过 | "LangCh ...

Core Insights - ECX reported a revenue of 1.22 billion RMB (approximately 170 million USD) for Q1 2025, marking a 30% year-over-year increase [1] - The company achieved a gross profit of 240 million RMB (approximately 33.5 million USD), with a stable gross margin of 20% [1] - ECX's shipment volume surged by 81% year-over-year, reaching 684,000 units, with over 8.7 million vehicles equipped with its technology globally [1] Business Structure - The collaboration with Geely Galaxy has significantly contributed to the growth in shipment volume, with Geely Galaxy's sales increasing by over 290% in March [2] - ECX secured new project orders for eight new Geely models, enhancing its position as a provider of intelligent solutions within the Geely Group [2] - The launch of the Antora® 1000SPB central computing platform and the development of high-level intelligent driving solutions based on NVIDIA DRIVE AGX platform are key innovations driving growth [2] International Expansion - ECX is preparing to establish a global supply chain and R&D center in Singapore to strengthen its delivery capabilities in Southeast Asia and international markets [3] - A strategic partnership with HERE Technologies aims to develop a next-generation global vehicle navigation system, enhancing localized adaptation for global OEM clients [3] - The company is focused on optimizing product cost structures and improving operational efficiency to support growth in the smart automotive sector [3] Long-term Strategy - The management anticipates continued growth in demand for intelligent cockpits and advanced driver assistance systems from global OEMs [3] - The integration of generative AI in vehicle human-machine interaction and software platform development is expected to bolster future business growth [3] - ECX aims to enhance the modularity and compatibility of its hardware and software platforms to accelerate adaptation processes for mainstream overseas automakers [3] Global Operations - Over the past eight years, ECX has established 12 global operation centers in countries including China, the UK, the US, Germany, Sweden, and Malaysia [4] - The company serves a diverse range of clients, including major automotive manufacturers such as Volkswagen, Volvo, Geely, Polestar, Hongqi, Dongfeng, and Changan [4] - ECX is focused on building core capabilities in key technology areas such as chip modules, intelligent cockpits, central computing, AI models, and software platforms to enhance its technological pricing power and platform leadership [4]

曲凯： 今天我们请来了国内强化学习 (RL) 领域的专家吴翼，吴翼目前是清华大学交叉信息研究院助理教授，他曾经在 OpenAI 工作过，算是国内最早研究强化学 习的人之一，我们今天就争取一起把 RL 这个话题给大家聊透。 首先吴翼能不能简单解释一下，到底什么是 RL？ 因此，RL 其实更通用一些，它的逻辑和我们在真实生活中解决问题的逻辑非常接近。比如我要去美国出差，只要最后能顺利往返，中间怎么去机场、选什么航 司、具体坐哪个航班都是开放的。 但 RL 很不一样。 RL 最早是用来打游戏的，而游戏的特点和分类问题有两大区别。 第一，游戏过程中有非常多的动作和决策。比如我们玩一个打乒乓球的游戏，发球、接球、回球，每一个动作都是非标的，而且不同的选择会直接影响最终的结 果。 第二，赢得一场游戏的方式可能有上万种，并没有唯一的标准答案。 所以 RL 是一套用于解决多步决策问题的算法框架。它要解决的问题没有标准答案，每一步的具体决策也不受约束，但当完成所有决策后，会有一个反馈机制来评 判它最终做得好还是不好。 吴翼： RL 是机器学习这个大概念下一类比较特殊的问题。 传统机器学习的本质是记住大量标注过正确答案的数据对。 ...

" 技术产品全球赋能810万辆车，亿咖通开启全球发展新阶段。 " 作者丨马广宇 编辑丨相辉 "2025年，我们的首要目标是实现持续盈利。" 在不久前结束的2025投资者日上，亿咖通科技联合创始人、董事长&CEO沈子瑜明确阐述了企业新一年的 发展目标，作为行业里少数具有汽车智能化全栈技术产品和全球交付能力的企业，亿咖通在2024年交出了 一份亮眼的成绩单。 3月中旬，亿咖通发布了2024年第四季度及全年业绩报告，宣布在2024年第四季度实现盈利。总出货量 同比增长33%，突破200万件，创下历史新高。全球范围内，搭载亿咖通技术产品的整车数量已达810万 辆，成绩可观。 值得注意的是，汽车供应链通常以 3-5 年为一代发展周期，亿咖通作为一家成立 8 年的企业，几乎是十 分迅速地达成了财务状况正循环的目标，而财务健康度作为考验企业经营状态乃至核心竞争力的关键指 标，考虑到亿咖通从创立之初就锚定全球化发展的特殊身位，实现这一成绩更为不易。 在多年的布局下，亿咖通的技术产品矩阵早已成为了能够满足全球市场需求的生态体系，也吸引了越来越 多像一汽红旗、大众集团这样的全球头部汽车品牌来合作，技术产品解决方案矩阵、全球化交付 ...

ECARX Holdings Inc. (NASDAQ:ECX) Q4 2024 Earnings Call March 11, 2025 8:00 AM ET Company Participants Rene Du - Head, Investor Relations Ziyu Shen - Chairman and Chief Executive Officer Peter Cirino - Chief Operating Officer Phil Zhou - Chief Financial Officer Conference Call Participants Daniel Wang - CICC Tony Shen - SPDBI Jenny Wang - UBS Wei Huang - Deutsche Bank Operator Good day and thank you for joining us. Welcome to ECARX’s Fourth Quarter and Full Year 2024 Earnings Conference Call. [Operator Instr ...