整理 | 美股研究社 在这个快速变化的时代，人工智能技术正以前所未有的速度发展，带来了广泛的机会 。 《AI日 报 》致力于挖掘和分析最新的AI概念股公司和市场趋势，为您提供深度的行 业 洞察和价 值 分 析。 A I 快 报 1. CoreWeave宣布，首次从市场获得英伟达新款高端AI芯片GB300 NVL72，供应方为戴尔科 技 。 通过英特尔的AI服务器系统，公司可以支持（OpenAI等）客户开发和部署规模更大、更加复杂 的AI模型。 2. 据报道，英特尔(INTC.US)新任首席执行官陈立武正考虑对其晶圆厂代工业务进行重大改革， 直接以14A工艺对阵竞争对手，以吸引苹果(AAPL.US)、英伟达(NVDA.US)等大客户。 业内消息称，陈立武根据业内反馈判断，英特尔18A工艺在吸引新客户方面正面临困难。因此， 他考虑将更多资源投入到更有可能与台积电竞争的14A工艺节点上。若最终决定停止进一步推广 18A工艺，英特尔或将面临数亿美元至数十亿美元的资产减值。这一决策复杂且涉及大量资金， 预计董事会可能要到秋季会议才能做出最终定夺。但需要强调的是，这一潜在调整不会影响英特 尔自身产品部门，以及微软(MSF ...

以下文章来源于红色星际 ，作者红色星际科技 红色星际 . 让更多人，更深入地了解自动驾驶行业！ 今年这家新势力将智驾投入的重心从端到端转移到了VLA，试图在端到端上取得拔尖成绩之后继续乘胜突击。这位负责人作为"舵手"可谓核心关键人物，而就在这样的节 骨眼上，如此核心的人物休假，对于这家新势力在前沿技术的研发上会产生什么样的影响呢？ 本文只做学术分享，如有侵权，联系删文 今年这家新势力的智驾核心团队并不稳定。智驾最核心的三个部门：端到端、世界模型、量产，端到端负责人已经离职，世界模型负责人带病休假，就只有量产的负责人 还在。 智驾一号位似乎对此并不担忧，还是信心满满，认为等新模型量产上车会在业界取得断代式领先。不过，端到端和世界模型是最重要的两个技术，核心人物不稳定，并不 是一个好现象。 这几年新势力在智驾自研上都出现一个严重的现象，人才流动和动荡成为常态。而且，越是某一阶段做的拔尖的新势力，人才越频繁的动荡。 作者 | 红色星际科技 来源 | 红色星际 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 据悉 ...

哈啰、蚂蚁金服与宁德时代，加快脚步，做Robotaxi（无人出租车）业务。 6月底，其选定的业务带头人——Yolanda亮相，接受《21CBR》等媒体群访。 四天前，这三家公司，通过旗下投资主体，共同发起成立"上海造父科技"，三方首期出资超30亿元，专 注于L4级自动驾驶技术研发、应用和商业化落地。 Yolanda的职位，就是造父科技首席科学家。其拥有加州大学伯克利分校博士学位，师从美国工程院院 士，有超10年智能驾驶领域工作经验，曾参与构建Tesla Autopilot系统。 她透露，造父科技已经组建约百人团队，其中70%为技术岗，超半数成员拥有博士学位。 这意味着，高层排兵布阵，已经搭建好了核心人才矩阵。 "Robotaxi正处于行业爆发前期。"Yolanda向记者介绍，未来将持续吸纳全球优秀人才。 以下是Yolanda的口述分享，内容有编辑。 三家合力 车端，我们可以使用宁德时代的长续航、快充动力电池，还能够运用其换电站，提升运营效率。滑动底 盘是新产品，天然是为 Robotaxi打造的，哈啰会与宁德时代共同探索该领域。 团队计划，分两步走。 第一步，我们将走差异化路线，搭建运营平台，兼容不同车企与Ro ...

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 对于26届毕业的算法小伙伴，这个时候应该找到实习了。 如果不是特别满意的实习岗位，往往很容易顾虑。一方面没有实习怕影响秋招，另一方面可能去了也学 不到东西。。。 我们在求职星球里回答很多类似的问题，对于没有实习经历的小伙伴来说，大多数我都建议一定要去实 习，即使方向不对口，也是比没有实习经历强一些的。 以下的一些小伙伴在 AutoRobo知识星球 的星友咨询的问题： 星主您好～冒昧打扰您啦。我是双非本，九硕，电子信息类。想来咨询一下关于实习选择的问题。在实 验室做过一些3D感知、目标检测、分割相关的工作，只能说跑通过开源算法，没有深入的去优化过。平 时老板也是比较放养的状态，加上自己也比较松散，现在研二准备实习有些慌。匆匆忙忙投递实习简 历，最近拿到了一些主机厂的实力，是后处理和数据岗位，怕过去就是打打杂，想问问您的建议~ 星主答：建议还是去实习实习，同时还要补补前沿自动驾驶算法的内容，比如大模型、世界模型、扩散 模型、强化学习等等。既然已经意识到自己的短版，就不能在不拼一把。挤一挤时间这几个月冲刺一下 总比 ...

Core Viewpoint - The rivalry between Pony.ai and WeRide has intensified, particularly following comments made by Pony.ai's CTO regarding their competitive positioning in the autonomous driving sector, leading to a public rebuttal from WeRide's CFO [2][3]. Financial Performance - From 2022 to Q1 2025, Pony.ai's revenue showed a consistent increase, with figures of 489 million, 515 million, 538 million, and 100 million yuan, while WeRide's revenue declined from 526 million to 72 million yuan during the same period [3][5]. - In terms of net profit, Pony.ai reported losses of 1.06 billion, 895 million, 1.96 billion, and 308 million yuan, totaling over 4.22 billion yuan since 2022. WeRide's losses were higher, totaling over 6.15 billion yuan during the same timeframe [5][9]. - WeRide maintained a relatively high gross margin, with figures of 232 million, 183 million, and 15.17 million yuan, and gross margins of 44.1%, 45.7%, and 30.7% respectively [6]. Market Capitalization - As of July 1, Pony.ai's market capitalization was approximately $4.69 billion, which is about 2.15 times that of WeRide at $2.17 billion. WeRide's stock price has dropped over 50% since its IPO, while Pony.ai's stock has remained relatively stable [8][11]. Business Strategy - Pony.ai focuses on a dual strategy of "Robotaxi + Robotruck" and aims to provide technology solutions to automotive companies, while WeRide employs a diversified approach with a product matrix that includes Robotaxi, Robobus, Robovan, and Robosweeper [12][13]. - WeRide has over 1,200 autonomous vehicles in operation and has secured over 10,000 orders for customized L4 autonomous vehicles, whereas Pony.ai has around 300 Robotaxis and 190 Robotrucks [12][13]. Industry Outlook - The Robotaxi market is projected to potentially generate up to $34 trillion by 2030, with significant competition emerging from various players, including Tesla and established companies like Waymo and Cruise [14]. - The ability to generate revenue and establish operational capabilities is becoming increasingly critical for companies in the Robotaxi sector, as reliance on financing is diminishing [14][15].

关注 「电动车公社」 和我们一起重新思考汽车 就在一周之前， 特斯拉的无人驾驶出租车 （robotaxi）在美国正式开启试运营。 十几 辆特斯拉 Model Y将面向公众提供打车服务， 每单收费 4.2 美元。 到了 6 月 28 日， 特斯拉又开始炫技：一辆Model Y在无人驾驶的状态下， 从特斯拉得州超级工厂直奔用户家中，总用时30分钟， 实现了全球首次无人驾驶 交付！ 这也就意味着， 特斯拉离实现自动驾驶的目标又近了一步。 结合萝卜快跑去年在武汉开启试运营的消息，以及其他几家头部自动驾驶公司的最新进展， robotaxi 似乎很快就要来到大家面前了。 众所周知，风口上的生意抢的就是速度，来得早的头部企业会吃掉大部分甚至绝大部分市场份额，来得晚的可能连口汤都喝不到。 而就在 robotaxi 第一梯队眼看要吃到肉的时候，居然还有新玩家刚刚准备进场，且看样子颇为自信。 这个新玩家，名字也很有意思： 造父科 技 ！ 01. 师出三家，融资超30亿 造父科技，全名 上海造父智能科技有限公司 ， 6 月 23 日刚刚成立。算起来，跟特斯拉 robotaxi正式 开启试运营是在同一天。 | | | 產公司 查老板 ...

Core Insights - The 17th International Transportation Technology and Equipment Exhibition showcased significant advancements in autonomous driving technology, highlighting its transformative impact on modern transportation and logistics [1][2]. Group 1: Autonomous Vehicles - The exhibition featured a new generation of autonomous vehicles, including the seventh-generation Robotaxi from Pony.ai and the first L4 mass-produced autonomous vehicle from Didi, attracting considerable attention from attendees [1]. - Autonomous trucks are currently operational on several domestic routes, such as the Beijing-Tianjin route, where they can reduce logistics costs by approximately 20% compared to traditional trucks [2]. Group 2: Cost Efficiency and Technology - The implementation of autonomous trucks not only saves labor costs but also reduces energy consumption through precise speed control and road prediction capabilities [2]. - The navigation systems supporting these vehicles are evolving to enhance safety and efficiency, incorporating features like lane-level navigation and vehicle-to-vehicle communication for accident prevention [3]. Group 3: Future of Transportation - The advancements in autonomous driving technology are expected to redefine transportation experiences, positioning them as a new driving force in the industry [3].

Core Viewpoint - The article discusses the challenges and opportunities faced by Hello Bike as it ventures into the Robotaxi market, highlighting the significant financial investments required and the uncertain path to commercialization [1][2][3]. Group 1: Robotaxi Business Development - Hello Bike has announced its entry into the Robotaxi sector, securing over 3 billion yuan in funding from Ant Group and CATL [2]. - The market for Robotaxi is projected to grow significantly, with Goldman Sachs predicting 500,000 autonomous taxis in China by 2030 and a market size of $47 billion by 2035 [2]. - Despite the optimistic outlook, the high costs associated with autonomous driving technology pose a significant challenge, with the cost of a single autonomous vehicle being around 480,000 yuan, nearly double that of a regular passenger car [2][3]. Group 2: Financial Viability and Investment - Hello Bike has raised over 10 rounds of financing, totaling more than 10 billion yuan, but the latest funding round was in 2021, indicating potential financial strain [4]. - The company has faced difficulties in monetizing its bike-sharing business, with significant losses reported in previous years, totaling over 4.8 billion yuan from 2018 to 2020 [5][6]. - The decline in investment from Ant Group, from 4 billion yuan in 2019 to 500 million yuan in 2023, reflects a shift in focus towards more promising sectors [6]. Group 3: Marketing and Regulatory Challenges - Hello Bike has attempted to increase revenue through partnerships and promotional activities, such as collaborations with luxury brands and coffee chains [7]. - However, regulatory challenges exist, as many cities have restrictions on advertising on shared bicycles, which could hinder Hello Bike's marketing strategies [8][9]. - The perception of shared bicycles as mobile advertisements has raised concerns among users, potentially affecting the brand's image and user experience [9][10]. Group 4: Future Outlook - The company aims to achieve commercialization of its Robotaxi service within three years, focusing on both domestic and international markets [3]. - The transition from bike-sharing to autonomous driving represents a significant shift in business strategy, with the need for substantial investment and innovation to succeed [10].

Core Insights - The article discusses advancements in autonomous driving frameworks, specifically highlighting the World4Drive, SafeMap, TopoStreamer, and BEV-VAE models, which improve various aspects of autonomous driving technology. Group 1: World4Drive Framework - The World4Drive framework, developed by CASIA and Li Auto, integrates spatial semantic priors and multimodal driving intention modeling, achieving an 18.1% reduction in L2 error (0.61m to 0.50m) and a 46.7% decrease in collision rate (0.30% to 0.16%) [2][3] - It introduces an intention-aware latent world model that simulates the evolution of the physical world under different driving intentions, closely aligning with human decision-making logic [3] - The framework demonstrates state-of-the-art planning performance without the need for perception annotations, with a training convergence speed improvement of 3.75 times [3] Group 2: SafeMap Framework - The SafeMap framework, proposed by Tsinghua University and others, utilizes dynamic Gaussian sampling and panoramic feature distillation to construct robust high-definition maps from incomplete observations, achieving an 11.1% improvement in mAP when key views are missing [9][10] - It features two innovative modules: G-PVR for perspective view reconstruction and D-BEVC for correcting bird's-eye view features, ensuring high accuracy in map construction even with missing camera views [10] - Experimental results show SafeMap significantly outperforms existing methods, providing a plug-and-play solution for enhancing map robustness [10] Group 3: TopoStreamer Model - The TopoStreamer model, developed by CUHK and Tencent, addresses the temporal consistency challenges in lane topology reasoning, achieving a 3.4% improvement in lane segment perception mAP (reaching 36.6%) and a 2.1% increase in centerline perception OLS (reaching 44.4%) [18][21] - It introduces three innovative modules to ensure temporal consistency in lane attributes and improve feature representation learning [21] - TopoStreamer achieves state-of-the-art performance in lane segment topology reasoning on the OpenLane-V2 benchmark dataset [21] Group 4: BEV-VAE Framework - The BEV-VAE framework, proposed by Shanghai QiZhi Institute and Tsinghua University, constructs a bird's-eye view latent space for multi-view image generation and precise 3D layout control, achieving a spatial consistency metric (MVSC) of 0.9505 on the Argoverse 2 dataset [29][31] - It supports new view synthesis by adjusting camera poses and demonstrates strong cross-view consistency [34] - The framework allows for controllable synthesis based on 3D object layouts, enhancing the capabilities of autonomous driving scene understanding [34]

1.《 AI技术赋能传统智能终端，推动新型智能硬件的需求增长 》摘要 ■ 预计 2025年上半年 ，消费电子行业将呈现"政策与AI双轮驱动"特征，中国的"国补"政策推动中端 消费市场复苏，AI技术深入应用于PC和手机等终端设备，推动智能化转型。 ■ 半导体行业仍处于高景气周期，AI算力需求推动海外算力提升，国内加速自主研发，尤其是 电子级 超纯水 材料市场的国产化进程也在加速。 ■ AI技术推动可穿戴产品需求释放，智能硬件的商业化进程加速，尤其是AI眼镜等新形态产品的商业 化，成为下半年消费电子行业的重要驱动力。 ■ 消费电子 （尤其是AI赋能的智能硬件和可穿戴设备）依然是热点，短期内政策支持和技术创新将带 动需求增长。 2、《 透露IPO进展，荣耀新资本故事跳出"手机"》摘要 ■ 荣耀 已于2025年6月26日获得 深圳证监局 的上市辅导备案受理，目前正处于辅导阶段，预计至少 需要三个月才能进入下一步。 ■ 荣耀的 阿尔法战略 于2025年3月发布，计划将公司从传统智能手机制造商转型为全球领先的 AI终 端生态公司 。 ■ 这一战略分为三步走：从智能手机入手，拓展AI智能体时代的技术边界，最终进入通用人工智 ...