以下文章来源于底线思维 ，作者白玉京 底线思维 . 观察者网时评专栏 本文来自微信公众号： 底线思维 ，作者：白玉京 当马斯克推动SpaceX收购xAI，并提出"太空可能成为生成式AI算力成本最低的地方"时，很多人还 没意识到新一轮太空竞赛的发令枪已经打响。 几乎在同一时间，SpaceX向美国联邦通信委员会申请部署数量高达100万颗的低轨卫星网络，这一动 作更像是在为某种超出现有通信需求的体系提前占位。两件事放在一起看，很难再把它们理解为彼此 孤立的商业决策。 过去几年，人工智能竞争常被简化为芯片性能和模型规模的比拼，但随着训练和推理规模持续放大， 电力消耗、散热能力以及数据中心扩容，正逐渐成为算力增长的现实瓶颈。 在这样的背景下，把算力与通信系统一并推向太空，不再只是科幻式设想，而是一种围绕能源条件、 系统结构和长期成本的现实选择。太空算力真正要回答的，远不只是芯片性能问题，而是能源从哪里 来、系统能否闭环，以及是否具备工程可行性。 如果说在低轨巨型互联网星座上，中国更多是在追赶别人已经跑通的路径，那么在"天算"这个尚未定 型的新赛道上，至少仍然存在提前进入、参与定义的空间。"天算"未必会变成第二个星链，但 ...

【文/观察者网专栏作者 白玉京】 当马斯克推动SpaceX收购xAI，并提出"太空可能成为生成式AI算力成本最低的地方"时，很多人还没意 识到新一轮太空竞赛的发令枪已经打响。 几乎在同一时间，SpaceX向美国联邦通信委员会申请部署数量高达100万颗的低轨卫星网络，这一动作 更像是在为某种超出现有通信需求的体系提前占位。两件事放在一起看，很难再把它们理解为彼此孤立 的商业决策。 过去几年，人工智能竞争常被简化为芯片性能和模型规模的比拼，但随着训练和推理规模持续放大，电 力消耗、散热能力以及数据中心扩容，正逐渐成为算力增长的现实瓶颈。 在这样的背景下，把算力与通信系统一并推向太空，不再只是科幻式设想，而是一种围绕能源条件、系 统结构和长期成本的现实选择。太空算力真正要回答的，远不只是芯片性能问题，而是能源从哪里来、 系统能否闭环，以及是否具备工程可行性。 如果说在低轨巨型互联网星座上，中国更多是在追赶别人已经跑通的路径，那么在"天算"这个尚未定型 的新赛道上，至少仍然存在提前进入、参与定义的空间。"天算"未必会变成第二个星链，但一旦系统形 态和运行规则被率先确立，后来者再想进场，成本就会明显提高。这也正是当前重 ...

Core Viewpoint - The government of Yantai has emphasized the importance of strengthening technological leadership and accelerating the construction of a high-level innovative city, alongside deepening the integration of education and industry, and enhancing cooperation between Yantai University and top institutions like Peking University and Tsinghua University [1][3]. Group 1: Achievements and Current Status - Yantai University has seen significant technology transfer success, with over 90 million yuan in technology transaction revenue expected in 2024 [3]. - The university has dispatched 109 technology specialists to various enterprises, with 22.4% of students choosing to stay in Yantai for employment or further studies [3]. - Notable achievements include a domestic market share of over 95% for the university's "carbon four separation technology" and participation in the launch of the "Tiansuan Constellation" satellite [3]. Group 2: Challenges and Bottlenecks - There is a mismatch between educational supply and industrial demand, particularly in emerging fields like green chemicals and renewable energy, which require interdisciplinary talent [4]. - The integration of technological innovation and industrial upgrading is insufficient, with existing research platforms not fully utilized [5]. - The talent mobility and resource-sharing mechanisms are not well-established, leading to challenges in attracting high-end talent [5]. - The collaboration between academia and local government lacks a robust policy framework and communication channels [5][6]. Group 3: Recommendations for Improvement - A proposed annual forum, "Peking University, Tsinghua University, and Yantai Cooperation Development Forum," aims to enhance collaboration and address local development needs [7]. - The establishment of specialized industry colleges in collaboration with Peking University and Tsinghua University is suggested to align educational programs with local industrial demands [7]. - The creation of a high-end innovation platform to tackle key technological challenges and the establishment of a four-party collaboration system for technology transfer are recommended [8]. - A talent-sharing initiative is proposed to attract top talent in green technology and marine science, along with a mutual hiring mechanism for faculty [8]. Group 4: Vision for the Future - Yantai University aims to deepen its collaboration with Peking University and Tsinghua University to contribute to the city's development as a model for green, low-carbon, and high-quality growth [9]. - The ongoing partnership, which began 40 years ago, is expected to enhance Yantai's intellectual advantages in regional competition [9].

Core Insights - The successful launch of the "Zhuque-2" rocket on May 17 marks the initiation of the second phase of the "Tiansuan Constellation" project, with the "Beiyou-2" and "Beiyou-3" satellites entering their designated orbits [1] Group 1: Project Overview - The "Tiansuan Constellation" is an open-source space computing experimental platform, with the second phase planning to launch a total of 24 satellites focusing on advanced fields such as space computing, 6G networks, and intelligent remote sensing [1] - The first batch of satellites in this phase, "Beiyou-2" and "Beiyou-3," were developed by Beijing University of Posts and Telecommunications in collaboration with Changsha Tianyi Research Institute [1] Group 2: Technological Innovations - The development team overcame key technological challenges related to the reliability of space servers, real-time capabilities of hyperspectral cameras, and the stability of laser communication systems in orbit [2] - Significant technological breakthroughs have enabled the satellites to possess deep payload integration and dynamic optimization capabilities, addressing issues such as unreliable computational supply, poor timeliness of remote sensing data, and limited communication bandwidth [2] Group 3: Future Developments - Following the launch, the satellites will conduct a series of frontier technology experiments in satellite internet, including inter-satellite high-capacity laser communication and dynamic tuning of satellite laser communication payload rates [2]