Core Insights - The core perspective presented by Shen Xiangyang emphasizes that opportunities arise not only from technology itself but also from understanding how technology transforms carriers, interactions, computing, and data, thereby redistributing capabilities, resources, and value [19]. Algorithm Paradigm - AI algorithms have evolved from supervised learning, which focuses on building expression and generation capabilities, to reinforcement learning that incorporates causality and execution [3]. - The next phase of exploration is moving towards autonomous learning at a higher cognitive level [4]. Intelligent Carrier - Current focus has shifted from language and multimodal models to world models and embodied models, indicating a transition from abstract symbolic spaces to physical spaces [7][8]. Human-Machine Interaction - Shen highlights the importance of innovation in human-machine interaction, which has undergone several paradigm shifts over the past 70 years, reflecting changes in underlying technological waves [11]. - Developers today must recognize the interaction opportunities brought by intelligent features, with natural interaction evolving from passive responses to proactive modes [12]. Computing Architecture - The trend in computing architecture is moving from general-purpose to specialized, and from singular to diverse solutions, with a focus on balancing performance, cost, and energy efficiency [14][15]. Data Utilization - Data has transitioned from being static textbooks in the simulation world to dynamic feedback in the exploration world, and finally to evidence for hypothesis validation in the induction world [17]. Future of Individuals and Organizations - As AI amplifies individual capabilities, tasks that previously required multi-step collaboration can now be completed by single individuals end-to-end [21]. - The focus of organizations may shift from control to defining problems and integrating individual collaboration [22]. Innovation Projects - The IDEA Research Institute announced the establishment of two innovation platforms and a series of innovative projects, including the DINO-X Grasp model aimed at enhancing embodied intelligence [24][25]. - DINO-X Grasp can accurately identify objects and predict grasp points, demonstrating versatility across various devices and scenarios [29][41]. MoonBit Programming Language - The MoonBit programming language, developed by Zhang Hongbo's team, has gained significant traction, with user numbers increasing from 26,000 to over 100,000 in a year and a half [37]. - MoonBit is designed for the AI era, featuring a native AI assistant and outperforming Java in performance benchmarks by nearly ten times [38]. Smart Renderer - The Smart renderer, developed by Wang Jiaping's team, offers real-time rendering capabilities that significantly reduce the time required for scene rendering in the film industry [46][48]. - Major companies in the film industry, such as MoreVFX and Light Chaser Animation, have begun utilizing the Smart renderer [49].