具身智能要想往大规模应用走，像英特尔这样的头部芯片公司必须突破算力架构。 当俄罗斯首个 AI 人形机器人「艾多尔」伴着电影《洛奇》的主题曲蹒跚登场时，所有人都以为某个高光时刻 即将来临。 没想到，「帅不过三秒」 —— 向观众挥手后，它迅速失去平衡、倒地抽搐，被工作人员匆忙拖走。 近期翻车的机器人可不止这一家。 9 月，特斯拉 Optimus 因反应迟缓被吐槽； 1X 预售款 的「惊艳演示」因系远程遥控，被舆论 diss 到起 飞。 业内人士对此并不意外。很多演示高度依赖人工操控，大量机器人连「站稳完成操作」都难，在工厂里「插个 dongle 、贴个膜」，堪比「登月」。 英特尔在与数十家具身智能团队沟通过程中也发现，机器人「能跑会跳」和「能在产线干活」之间，还存在巨 大鸿沟。 到底是什么原因阻挡它们踏入生产一线呢？ ▍ 困在算力平台里的具身智能 11 月 19 日，重庆 ·2025 英特尔技术创新与产业生态大会的圆桌现场，训练数据、应用、「大脑 / 小脑分 家」的架构问题都被摆上台面。但有一个答案被反复提及，算力平台正成为横在具身智能落地面前的最大门槛 之一。 目前业内已量产、相对成熟的人形机器人，大多采用「 ...

机器之心原创 作者:Sia 具身智能要想往大规模应用走，像英特尔这样的头部芯片公司必须突破算力架构。 当俄罗斯首个 AI 人形机器人「艾多尔」伴着电影《洛奇》的主题曲蹒跚登场时，所有人都以为某个高 光时刻即将来临。 没想到，「帅不过三秒」 —— 向观众挥手后，它迅速失去平衡、倒地抽搐，被工作人员匆忙拖走。 近期翻车的机器人可不止这一家。 9 月，特斯拉 Optimus 因反应迟缓被吐槽； 1X 预售款 的「惊艳演示」因系远程遥控，被舆论 diss 到起飞。 业内人士对此并不意外。很多演示高度依赖人工操控，大量机器人连「站稳完成操作」都难，在工厂里 「插个 dongle 、贴个膜」，堪比「登月」。 英特尔在与数十家具身智能团队沟通过程中也发现，机器人「能跑会跳」和「能在产线干活」之间，还 存在巨大鸿沟。 到底是什么原因阻挡它们踏入生产一线呢？ 困在算力平台里的具身智能 11 月 19 日，重庆 ·2025 英特尔技术创新与产业生态大会的圆桌现场，训练数据、应用、「大脑 / 小脑分家」的架构问题都被摆上台面。但有一个答案被反复提及，算力平台正成为横在具身智能落地面 前的最大门槛之一。 目前业内已量产、相对成熟的人 ...

Core Viewpoint - NVIDIA has launched the Jetson AGX Thor developer kit and production modules, aimed at enhancing robotics across various industries, including manufacturing, logistics, transportation, healthcare, agriculture, and retail [1][15]. Product Overview - Jetson Thor is designed to power millions of robots, featuring an NVIDIA Blackwell GPU and 128GB of memory, delivering up to 2,070 FP4 teraflops of AI compute within a 130-watt power envelope [3][16]. - Compared to its predecessor, Jetson Orin, Jetson Thor offers up to 7.5 times higher AI compute and 3.5 times greater energy efficiency, enabling the execution of complex generative AI models [4][16]. Industry Adoption - Early adopters of Jetson Thor include prominent companies such as Agility Robotics, Amazon Robotics, Boston Dynamics, Caterpillar, Figure, Hexagon, Medtronic, and Meta, with others like John Deere and OpenAI evaluating its capabilities [2][8]. - The platform has attracted over 2 million developers and a growing ecosystem of over 150 hardware, software, and sensor partners since its inception in 2014 [7][16]. Technological Impact - Jetson Thor addresses significant challenges in robotics by enabling real-time, intelligent interactions between robots and their environments, which is critical for applications in humanoid robotics, agriculture, and surgical assistance [5][16]. - The system supports any popular AI framework and generative AI model, fully compatible with NVIDIA's software stack from cloud to edge, enhancing its utility in various applications [6][16]. Market Availability - The NVIDIA Jetson AGX Thor developer kit is currently available for purchase starting at $3,499, with production modules accessible through worldwide distribution partners [10][16].

Financial Data and Key Metrics Changes - The company reported a GAAP net loss of $9,300,000 or $0.48 per share in Q2 2025, an increase from a net loss of $8,000,000 or $0.46 per share in Q2 2024 [22] - Non-GAAP net loss was $6,700,000 or $0.35 per share in Q2 2025, compared to a non-GAAP net loss of $5,500,000 or $0.31 per share in the prior quarter [22] - Cash burn decreased to $7,100,000 in Q2 2025 from $8,100,000 in Q1 2025, despite one-time expenses [20][22] - The company ended the quarter with cash, cash equivalents, and marketable securities of $19,200,000, which has since more than tripled [22][23] Business Line Data and Key Metrics Changes - The company has signed six revenue-generating contracts in Q2 2025, tripling the number of contract wins from two in the previous quarter [19][40] - The sales funnel has grown significantly, leading to 30 new potentially high-value customer engagements [7][19] - The launch of Optus, a next-generation platform, has been deployed to multiple customers, enhancing the company's ability to scale efficiently [7][12] Market Data and Key Metrics Changes - The company is seeing strong traction in various sectors including defense, smart infrastructure, rail, trucking, aviation, and security, indicating a diverse market presence [16][40] - The integration with NVIDIA's DRIVE AGX ecosystem is expected to accelerate OEM collaborations and expand market reach [17][30] Company Strategy and Development Direction - The company aims to transition from product development to active sales and deployment, focusing on delivering differentiated products that gain market traction [25] - The capital light financial strategy allows the company to maintain low operating costs while driving significant scale [13][20] - The focus on physical AI and the integration of third-party software solutions through Optus is expected to enhance market opportunities [12][70] Management's Comments on Operating Environment and Future Outlook - Management expressed confidence in the pipeline, with over 100 potential customers actively engaged and 30 in advanced negotiations [19][60] - The company anticipates modest top-line revenue growth for the remainder of the year but emphasizes the importance of accelerating customer engagements [19] - Management highlighted the importance of maintaining a disciplined approach to capital allocation while scaling operations [23][86] Other Important Information - The company has secured a $30,000,000 opportunity with a top global transportation OEM, expected to contribute to revenue this year [6][48] - The total potential liquidity, including cash and credit facilities, is approximately $126,000,000, providing a strong foundation for future growth [23][84] Q&A Session Summary Question: Can you provide more details on the Navidion integration? - The integration with NVIDIA has positioned the company at the top of performance benchmarks, simplifying conversations with OEMs and enhancing credibility [29][30] Question: Can you elaborate on Optus and its role in the broader strategy? - Optus combines sensing and analytics, allowing for tailored AI solutions and rapid deployment across various markets, filling the gap while automotive ramps up [34][35] Question: What is the status of the customer pipeline? - The company has over 100 engaged customers, with 30 in advanced negotiations, and is seeing traction across diverse industries [40][60] Question: Are there additional deliverables for the $30,000,000 opportunity? - The company is on the customer's timeline for integration and deployment, actively working on the project [48][49] Question: Can you provide insight into the sales and marketing expenses? - The increase in sales and marketing expenses is primarily due to reallocating funds from G&A and R&D, rather than new incremental spending [53] Question: What does "physical AI" mean in the context of the company's strategy? - Physical AI refers to the interaction of AI and sensing with the real world, extending beyond automotive applications [70] Question: Is the company looking to partner with defense contractors? - The company is open to partnerships with defense contractors and is actively pursuing opportunities in that sector [76][81]

Group 1: Core Innovations - YUAN is redefining real-time video analytics and autonomous decision-making with its Pandora NX Super and AIR NX Super platforms, built on the NVIDIA Jetson Orin platform [1] - The Smart Sea Patrol solution enhances maritime safety through real-time threat detection and 360-degree monitoring, integrating multi-sensor data for rapid hazard identification [3] - Smart Farming solutions utilize NVIDIA Isaac Sim for precise crop monitoring and resource optimization, enabling rapid pest detection and targeted spraying [4] Group 2: Advanced Applications - The Smart Drone, powered by Jetson Orin NX, provides high-resolution aerial inspections and excels in infrastructure monitoring and emergency response [5] - YUAN's solutions enable real-time edge AI and multi-sensor fusion, with the upcoming ARC AI Platform aimed at enhancing humanoid robotics [6] - YUAN will showcase its next-gen AI robotics at COMPUTEX 2025, highlighting innovations for land, sea, and air [6]