1月19日，特斯拉发文表示，特斯拉智能辅助驾驶硬件 AI 4 可以实现远超人类的自动驾驶安全水平，还 宣布特斯拉正计划重启 Dojo 3 芯片的研发工作，Dojo 3 将用于太空人工智能计算。 同时，文中同步发布"招聘启事"，邀请对"研发世界上产量最高芯片"感兴趣的工程师加入，并要求应征 者用三个要点简述其解决过的关键技术难题。 责任编辑：李昂 责任编辑：李昂 1月19日，特斯拉发文表示，特斯拉智能辅助驾驶硬件 AI 4 可以实现远超人类的自动驾驶安全水平，还 宣布特斯拉正计划重启 Dojo 3 芯片的研发工作，Dojo 3 将用于太空人工智能计算。 同时，文中同步发布"招聘启事"，邀请对"研发世界上产量最高芯片"感兴趣的工程师加入，并要求应征 者用三个要点简述其解决过的关键技术难题。 ...

Investment Rating - The report maintains a "Recommended" investment rating for the automotive industry, indicating that the industry index is expected to outperform the benchmark index by more than 10% [6]. Core Insights - The automotive industry is entering a phase of low growth, with a focus on high-quality development opportunities. Key areas of interest include high-growth automotive companies and structural opportunities within the supply chain, particularly in commercial vehicles and automotive technology [4][34]. - The China Association of Automobile Manufacturers (CAAM) forecasts that total vehicle sales in China will reach 34.75 million units in 2026, representing a year-on-year growth of 1%. Passenger vehicle sales are expected to grow by 0.5%, while commercial vehicle sales are projected to increase by 4.7% [3][35]. - New energy vehicles (NEVs) are anticipated to play a crucial role in driving industry growth, with expected sales of 19 million units in 2026, reflecting a significant year-on-year growth of 15.2% [3][35]. Summary by Sections 1. Weekly Market Review (January 10-16, 2026) - The automotive sector index increased by 0.49%, outperforming the Shanghai and Shenzhen 300 index by 1.06 percentage points. The automotive services sector saw the highest growth at 4.51% [12][15]. 2. Weekly Data Tracking (January 10-16, 2026) - From January 1-11, 2026, retail sales of passenger vehicles in China totaled 328,000 units, a 32% decrease year-on-year. Wholesale figures were 381,000 units, down 40% year-on-year [20][21]. 3. Industry News (January 10-16, 2026) - Significant developments include partnerships for advanced driving technologies and the introduction of new vehicle models by major manufacturers, indicating ongoing innovation in the sector [25][29][31]. 4. Key Manufacturer Sales Rankings (2025) - BYD led the passenger vehicle market with sales of 4.55 million units, followed by Geely and Chery. In the NEV segment, BYD also dominated with a market share of 29.7% [23][24]. 5. Future Outlook - The report emphasizes the importance of macroeconomic policies and industry governance in sustaining growth. The focus will be on maintaining competitive advantages in electric and intelligent vehicle technologies [34][36].

Investment Rating - The report maintains a "Recommended" investment rating for the automotive industry, indicating a positive outlook for certain segments within the sector [6]. Core Insights - The automotive industry is expected to experience low growth in 2026, with total vehicle sales projected to reach 34.75 million units, reflecting a year-on-year increase of 1%. Passenger vehicle sales are anticipated to grow by 0.5%, while commercial vehicle sales are expected to rise by 4.7%. The new energy vehicle (NEV) segment is projected to grow significantly, with sales expected to reach 19 million units, a 15.2% increase year-on-year, highlighting its role as a key growth driver [3][35]. - The report emphasizes the importance of macroeconomic policies and GDP growth targets in influencing vehicle sales, particularly in the passenger vehicle segment. The report suggests that sustained policy support is crucial for maintaining sales momentum [3][35]. Summary by Sections 1. Weekly Market Review (January 10-16, 2026) - The automotive sector index increased by 0.49%, outperforming the Shanghai Composite Index by 1.06 percentage points. The automotive services sector saw the highest gains, with a 4.51% increase [12][15]. 2. Weekly Data Tracking (January 10-16, 2026) - Retail sales of passenger vehicles from January 1-11 totaled 328,000 units, a 32% decrease year-on-year. Wholesale figures showed a similar trend, with 381,000 units sold, down 40% year-on-year. The NEV market also faced declines, with retail sales of 117,000 units, down 38% year-on-year [20][21]. 3. Industry News (January 10-16, 2026) - Key developments include partnerships for advanced driving technologies and initiatives to promote autonomous vehicle testing in regions like Hong Kong. Additionally, the report highlights the ongoing negotiations between China and the EU regarding electric vehicle trade, which could stabilize market conditions [25][28][30]. 4. Investment Recommendations - The report suggests focusing on structural opportunities within the automotive sector, particularly high-growth companies and regions, as well as the recovery of commercial vehicles and advancements in automotive technology [4].

Core Viewpoint - Chery Automobile is fully embracing AI, with Chairman Yin Tongyue stating that AI is as transformative as the steam engine and electricity, and aims to rapidly catch up with Tesla's Full Self-Driving (FSD) capabilities [6][7]. Group 1: AI Strategy and Goals - Chery aims to become one of the most committed and in-depth technology companies in China embracing AI, with a focus on three strategic directions: empowering products, intelligence, and industry to achieve product equality, emotional equality, and industrial equality [6][7]. - The company is benchmarking against Tesla, sending teams to the U.S. to experience Tesla's FSD and Grok model combinations to identify gaps and accelerate catch-up efforts [6][7]. Group 2: AI Empowerment in Products - Chery is focusing on its core business of smart vehicles, enhancing smart cockpits, intelligent driving, powertrains, and chassis through AI [8]. - The company plans to implement L2++ level driver assistance across its entire lineup domestically and aims for mass production of L3/L4 level autonomous vehicles and Robotaxi [9]. Group 3: International Development and Innovation - Internationally, Chery will develop VPD (Valet Parking Driver) technology to ensure all users can equally enjoy the safety, security, and convenience brought by AI [10]. - The company has established an open innovation system focusing on nine forward-looking areas, including 2C technologies (batteries, chips), 2R technologies (humanoid robots, Robotaxi), and 3G technologies (green energy, materials, computing) [10]. Group 4: Challenges in AI Integration - Chery acknowledges three major challenges in the integration of AI and the automotive industry: safety challenges due to the complexity of training data and the risks of AI decision-making [12]. - Legal and ethical challenges arise from data privacy, algorithm fairness, and defining responsibilities in accidents, necessitating a governance framework aligned with human values [13]. - Quality management challenges are highlighted, as the introduction of extensive software may lead to numerous bugs, requiring a new quality management logic suitable for AI development [14].

1月18日，特斯拉首席执行长伊隆·马斯克（Elon Musk）宣布一项雄心勃勃的人工智能（AI）芯片路线图，计划每九个月推出新一 代AI 处理器，这个速度将超越竞争对手英伟达（Nvidia）和AMD 的年度发布节奏。 马斯克表示，这个策略旨在使特斯拉能生产全球最产量的AI芯片，并在自动驾驶和机器人技术领域实现更快的创新。 马斯克透露，其新一代的自动驾驶芯片AI5的设计已接近完成，AI6芯片正在早期开发中，并将通过合作伙伴三星电子和台积电进 行生产，分别使用2nm和3nm制程。 根据预计， AI5芯片将提供比现有AI4高达50倍的性能，并在每美元性能上提升10倍，功耗效率提高3倍，专为全自动驾驶 （FSD）、Robotaxi和Optimus机器人量身打造。AI6的目标是实现约2倍于AI5的性能，并计划在2028年进入量产。 特斯拉已与三星签署了一项价值165亿美元的协议，以支持AI6的开发和量产。未来的AI7、AI8和AI9芯片也将遵循这一九个月的 设计周期，旨在快速迭代AI技术，并在竞争中获得优势。 事实上，特斯拉在数月前才刚宣布解散Dojo 超级电脑团队。此前报导指出，Dojo 计划负责人Peter Ban ...

Core Insights - Changan Automobile is transitioning from a traditional car manufacturer to a smart low-carbon mobility technology company, showcasing its commitment and capabilities through the "Beidou Tianshu" strategy [1][3] Group 1: Strategic Development - Since the launch of the "Beidou Tianshu" strategy in 2018, Changan has completed the 1.0 phase, achieving significant breakthroughs in intelligent technology and improving market performance, with vehicle sales increasing from 1.76 million in 2019 to 2.91 million projected for 2025, marking a continuous growth over seven years [3] - The company has established a central integrated software and hardware architecture (SDA platform) and seven core technology systems, focusing on intelligent driving, intelligent cockpit, and intelligent chassis, utilizing over 2,500 key technologies to implement more than 500 functional scenarios, including 48 industry-first features [3] Group 2: Technological Achievements - Changan has made significant advancements in intelligent driving, achieving 17 milestone results, including the first domestic 2,000-kilometer intelligent driving test and the continuous iteration of APA 5.0-7.0, establishing a technological advantage [3] - The company received a level 3 autonomous driving permit from the Ministry of Industry and Information Technology and the first official license plate for level 3 autonomous driving in China, marking a transition from technical validation to mass production and compliant commercial use [4] Group 3: Future Plans - The "Beidou Tianshu 2.0" strategy, announced in February 2025, aims to build a smart architecture centered around the Tianshu model, enhancing intelligent driving from "rule-based" to "flexible response," and creating a cockpit that provides emotional security [5] - Changan plans to explore level 3 autonomous driving for consumers in 2026, scale up level 3 and demonstrate level 4 Robotaxi in 2027, achieve partial level 4 applications in 2028, and reach complex scenario level 4 goals by 2030 [5] Group 4: Industry Perspective - The company emphasizes that the scaling of autonomous driving is a test of ecological collaboration, requiring the establishment of standards, data sharing, and risk-sharing to transform China's scene advantages into industrial strengths [6] - Changan is constructing a future mobility paradigm of "1+N+X," which includes one core brain, multiple diverse carriers (cars, humanoid robots, flying cars), and various service agents, promoting the evolution of mobility towards safety, intelligence, three-dimensionality, and sharing [6]

自动驾驶要解决的是场景的泛化性，具身要解决的是行为的泛化性。 点击下方 卡片 ，关注" 第一具身范式 "公众号 第一时间获取具身智能 干货 编辑 | 第一具身范式 原文链接： http://xhslink.com/o/9dt9pYOtnY 谷歌waymo最近在一次采访中提到:自动驾驶是最简单的机器人，是最复杂的社交游戏。这启发了我去认真思 考一下自动驾驶和具身机器人的关系。 正好最近抽空把physical intelligence的pi系列论文好好看了下，一直觉得自动驾驶和机器人在技术栈上很类 似，甚至一度觉得自驾其实是机器人的子集，现在看完的想法是两者在量产角度上差异远比想象的大。具身 智能的发展周期可能也和自动驾驶不太一样。 泛化性: 自动驾驶一直想解决的是场景的泛化问题。简单来说，就是自动驾驶其实想要的是对当前场景能够有一个考 虑尽可能全面的理解，然后做出对应的决策。举个具体的例子自动驾驶需要在前面有个锥桶的时候知道刹车 但也需要当前面有个载着锥桶的工程车的时候知道不用刹车。体现的是对场景的认知泛化能力。 比如多段式模块规则时代，对运动信息有预测模块，对语义信息有融合模块，在规划层做场景信息的整合与 理 ...

Group 1 - New Chip Journey (新芯航途) was established on December 19, 2023, in Suzhou, focusing on integrated circuit chip design, particularly in the autonomous driving sector [1][2] - The company has a competitive advantage in the niche market due to its professional technical team and innovative R&D capabilities [1] - New Chip Journey has completed angel round financing in August 2025 and has optimized its equity structure through capital adjustments [2] Group 2 - The company has made investments in two enterprises and holds 35 trademark registrations and four administrative licenses, indicating a growing intellectual property portfolio [2] - The importance of chips as core components in the rapidly developing autonomous driving industry is increasing, positioning New Chip Journey to seize market opportunities [2] - The addition of shareholders like Chery Automobile (奇瑞汽车) will enhance the company's resource integration and R&D capabilities, facilitating faster growth in the autonomous driving chip sector [2][3]

政策与技术双驱 智驾 L3 与 L4 的变局 20260118 摘要 英伟达开源 Alpaca 模型及相关生态，降低 Robotaxi 规模化落地和 L2+技术普及的技术门槛，将行业竞争焦点从算法转向工程和运营能力， 有利于跨区域复制和提升监管效率。 北美自动驾驶政策趋于宽松，简化审批流程、放宽安全要求，旨在促进 技术创新和商业化，为本土及全球车企提供更高效灵活的研发环境，加 速自动驾驶技术进步。 自动驾驶技术的核心短板包括极端天气适配、施工路段识别和用户信任 体系构建。解决方案包括模态增强、视觉与语言理解结合、线控系统冗 余设计以及提供可解释性决策。 国内自动驾驶产业发展需关注技术科学性和长期复现能力，补充可解释 性，重视仿真和物理 AI 应用，优化样本分布。商业化方面，可考虑一次 性放开更大规模 OTA 指标，实施跨城运营。 2026 年中国智能驾驶领域关键进展包括 L3 级别带条件商业化、 Robotaxi 规模扩展（数万台规模）以及自动驾驶企业在模型和方法论 上的迭代优化。 Q&A 英伟达模型开源对 Robotaxi 规模化落地和 L2+级技术普及有哪些核心影响？ 以及对自动驾驶政策中安全可解释性要求 ...

Core Insights - The article highlights the successful road testing of autonomous logistics vehicles in Liaoyuan City, showcasing advancements in smart logistics [1] Group 1: Vehicle Specifications and Performance - Two new energy autonomous vehicles were tested, each measuring 3.5 meters in length, with a cargo volume of 6 cubic meters and a load capacity of 870 kilograms [1] - The vehicles have a full electric range of 200 kilometers and a maximum speed of 40 kilometers per hour, equipped with an L4 autonomous driving system, one radar, and twelve high-definition cameras for 360° environmental perception [1] - The testing route included a total of 68 kilometers, covering one primary highway, one rural road, five urban main roads, and roads in a sock industry park, with approximately 1100 kilometers of road testing completed so far [1] Group 2: Efficiency and Cost Savings - The autonomous delivery vehicles have demonstrated a 30% increase in delivery efficiency compared to traditional delivery methods [1] - Each vehicle reportedly saves over 50% in operational costs per month [1] Group 3: Future Developments and Industry Impact - The initiative is expected to reduce accident risks and alleviate regulatory pressures while creating new job roles such as autonomous driving operators and remote dispatchers [1] - Liaoyuan Jiaofa Group plans to establish a unified regulatory platform, gradually expanding application scenarios and integrating smart warehousing and data platforms to promote the transformation and upgrading of the logistics industry [1]