二〇二五年四月 杨柯巍 中国电子信息产业发展研究院 军民所【商业航天研究中心（筹）】 所长 1 1 2 3 我国商业航天 发展所处阶段 3 ( 一 ) 起步期（ 2015-2020年 ) ：政策破冰与市场萌芽 商业航天 成长期 成熟期 起步期 转型期 技术触发阶段 ( 2019—2020年 ) 缓慢起步阶段 ( 2015—2018年 ) ● 星际荣耀首次入轨发射、星河动力"谷神星一号"成功 官飞： 《关于促进商业运载火箭规范有序发展的通知》出台。 . 2 2023年中央经济工作会议，2024年、2025年政府工作报告都 点名"商业航天"，全产业链实现快速发展，星箭场用企业 陆续挺过"从0到1"的过程，以高水平安全保障高质量发展逐 步成为共识。 《江山链生本· 加强 指式与 ● 《国家民用空间基础设施中长期发展规划(2015—2025 年）》出台; ● 蓝箭航天、星际荣耀等民营航天企业相继成立。 ● 2020年卫星互联网被纳入国家"新基建"; ● 2021年中国星网成立，《关于促进微小卫星有序发展和 加强安全管理的通知》印发； ● 2022年力箭一号、捷龙三号等火箭成功首飞。 (二)成长期(2020年至今):产 ...

Investment Rating - The report does not explicitly provide an investment rating for the aerospace industry. Core Insights - The commercial aerospace sector is rapidly developing, driving innovation across high-end manufacturing, information communication, new materials, and new energy sectors, while integrating into various economic and social fields [3][10]. - The establishment of a comprehensive aerospace economic measurement system is deemed feasible, important, and urgent, reflecting the contributions of aerospace to the economy and measuring the scale of the commercial aerospace industry [6][9]. Summary by Sections Section 1: Importance of Aerospace Economic Measurement - The concept of "aerospace economy" describes the economic implications of aerospace technologies and products, with the commercial aerospace industry scale estimated based on the aerospace economic scale, which in the U.S. accounts for approximately 80% [10]. - The aerospace industry has evolved through three stages: military and research aerospace, civilian aerospace, and the rapid development of commercial aerospace, with significant applications in environmental monitoring, urban planning, and disaster prevention [11]. Section 2: Measurement as a Response to Trends - The measurement of aerospace economics is a necessary choice to align with the development trends and statistical needs of the aerospace sector, especially as the demand for accurate statistics has increased due to the rapid growth of the commercial aerospace industry [15][18]. - The government has recognized commercial aerospace as a new growth engine, emphasizing the need for data-driven decision-making in aerospace governance [17]. Section 3: Current Measurement Challenges - There are significant discrepancies in the measurement of aerospace economics in China, including unclear definitions, diverse statistical standards, and inconsistent reporting, which hinder effective domestic and international comparisons [20][21]. - The report highlights the need for a unified measurement framework to support accurate analysis and decision-making in the aerospace sector [22][26]. Section 4: Proposed Measurement Framework - The report suggests the establishment of a "China Aerospace Economy and Core Industry Directory" to categorize and identify relevant industries based on international standards [49][51]. - The proposed framework includes the design of a satellite account for aerospace economics, which would encompass supply and use tables, balance sheets, and input-output tables to meet the statistical needs of various stakeholders [55][60].

Investment Rating - The report does not explicitly state an investment rating for the new generation of smart terminals. Core Insights - The new generation of smart terminals is defined as products that leverage advanced technologies such as artificial intelligence, advanced computing, virtual reality, the Internet of Things, and 5G/6G to achieve autonomous decision-making, efficient computing power utilization, multimodal data processing, interactive experiences, and personalized services [12][15]. Summary by Sections 1. Overview of New Generation Smart Terminals - New generation smart terminals can be categorized into two types: upgraded traditional smart terminals that integrate AI and entirely new forms with unique functionalities [15]. - They are further divided into consumer-oriented and industry-oriented smart terminals, with examples including AI smartphones, AI PCs, and smart industrial terminals [15]. 2. Key Technology System of New Generation Smart Terminals - The technology system is centered around AI, integrating storage-computing, communication, audio-video, sensing, interaction, display, battery, and software/operating systems [41]. 3. Development Status of Key Links in New Generation Smart Terminals - The report highlights the importance of SoC chips, which need to integrate multiple heterogeneous computing units and move towards green and low-carbon development [48]. - AI will increase the demand for memory capacity and bandwidth, pushing for advancements in LPDDR5 memory technology [49]. 4. Relevant Policies for New Generation Smart Terminals - Since the release of the "New Generation Artificial Intelligence Development Plan" in 2017, various policies have been issued to support the development of AI terminals, focusing on innovation, integration, and application scenarios [70]. 5. Challenges Facing the Development of New Generation Smart Terminals - The industry faces hardware performance bottlenecks, security risks, and a lack of standards, which complicate the promotion and integration of AI capabilities in terminals [80]. 6. Recommendations for the Development of New Generation Smart Terminals - The report suggests accelerating core technology research, building an AI application ecosystem, and establishing industry standards to enhance safety and regulatory frameworks [85][88].

Investment Rating - The report does not explicitly provide an investment rating for the industry Core Insights - The report emphasizes the urgent need to optimize and improve the governance path for artificial intelligence (AI) ethics in China, as the technology rapidly permeates various sectors, raising ethical concerns globally [4][6] - It identifies five major ethical risks associated with AI, including infringement risks, discrimination risks, social risks, accountability risks, and uncontrollability risks [14] - The report outlines the current governance practices in international contexts, particularly in the EU and the US, and compares them with China's evolving governance framework [26][30][31] Summary by Sections Governance Status - The report discusses the current state of AI ethical risk governance, highlighting the need for a comprehensive legal framework and the establishment of regulatory bodies to address ethical concerns [31][32] - It notes that China has begun to explore specific governance practices, including the introduction of laws and regulations aimed at managing AI-related ethical risks [32][33] Ethical Risks - The report categorizes AI ethical risks into five dimensions: infringement risks (direct violations of personal rights), discrimination risks (unjust outcomes for specific groups), social risks (negative impacts on societal order), accountability risks (difficulties in defining responsibility), and uncontrollability risks (loss of human control over AI) [14] International Practices - The report reviews international governance practices, noting that the EU has a more rigid and systematic approach, while the US emphasizes flexibility and voluntary compliance [26][30] - It highlights various international agreements and initiatives aimed at establishing ethical standards for AI, such as the "Bletchley Declaration" and the "Seoul Declaration" [18][19] Optimization Path Suggestions - The report suggests several paths for optimizing AI ethical governance, including enhancing the legal framework, strengthening multi-party collaborative governance mechanisms, and deepening international cooperation [11][55] - It emphasizes the importance of balancing technological advancement with ethical safety, addressing the complex relationship between technology development and ethical governance [46][49]

中国低空经济应用场景研究报告 （2025） 低空经济应用场景发展现状和特点 低空经济应用场景商业化路线图 低空经济应用场景发展挑战 中国电子信息产业发展研究院 产业政策研究所（先进制造业研究中心） 下一步工作建议 二〇二五年四月 / 003 / 009 / 013 / 015 1 3 3 二 一 低空经济应用场景发展现状和特点 p 按用途将低空经济应用场景分为生产作业、交通运输、文旅体验、安防安保等四大领域 | | 生产作业 | | | 交通运输 | | | 文体活动 | | 安防安保 | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | 公共服务 | 应急保障 | 科学探测 | 商业生产 | 载物运输 | 载人交通 | 文旅应用 | 体育应用 | 教育培训 | 安防巡逻 | 要地防护 | | 违建巡查 | 灾害救援 | 地质与海洋 勘探 | 农林植保 | 特种物资运 输 | 城市空中交 通（UAM） | 无人机表演 | 航空运动 | 科普体验 | 城市安全巡 逻 | 低空监测识 别体系 | | 河道巡检 | 紧急医 ...

Investment Rating - The report does not explicitly provide an investment rating for the manufacturing industry innovation index [1]. Core Insights - Innovation is identified as the primary driving force for development, especially in the context of a new round of technological revolution and industrial transformation [7]. - The manufacturing industry in China is at a strategic opportunity period for high-quality development and transformation, making the assessment of innovation levels crucial for implementing a strong manufacturing strategy [7]. Summary by Sections Manufacturing Innovation Index System - The manufacturing innovation index system includes various components such as innovation resources, innovation collaboration, innovation output, innovation environment, and innovation performance [10][12]. - Key metrics include R&D expenditure intensity, the proportion of R&D expenditure in total social R&D investment, and the number of effective invention patents among large-scale industrial enterprises [12]. National Manufacturing Innovation Index - The national manufacturing innovation index has shown a steady increase, with an average annual growth rate of 11.6% from 2012 to 2022 [22]. - The growth rates of various sub-indices are as follows: innovation resources (8.1%), innovation output (12.0%), and innovation collaboration (19.2%) [22]. - The index values have increased from 100 in 2012 to 299 in 2022, indicating a significant upward trend [17]. Provincial Manufacturing Innovation Index - The report highlights that Guangdong, Jiangsu, and Zhejiang lead in innovation indices, with scores above 200, while 19 regions fall below the national average [57]. - The effective invention patent numbers and patent applications in Guangdong are significantly higher than the national average, indicating strong innovation output [76][77]. - The report also notes that the innovation performance index for Guangdong is 4.5 times the national average, showcasing its leading position [91].

中国电子信息产业发展研究院 工业经济研究所 二O二五年四月 中国制造业国际化：趋势、风险 及应对 思路架构二 整体篇 历史演变：制造业国际化的四个阶 段 内外双驱：日益复杂的外部环境与 自身发展的内驱动力 风险挑战：地缘变局、蝴蝶效应、 市场风险 政策篇 增强前瞻性、扩大融合性、提高抗 压性、优化系统性 产业篇 新能源汽车：产业生态国际化 关键矿产：竞争愈发激烈 动力电池：国际化投资热情高涨 光伏：全球造，全球卖 经验篇 日本：先产品后产能、先大企业后 中小企业跟随、先劳动密集型后资 本技术密集型 德国：以坚实的工业基础与细分制 造领域的持续深耕，推动"德国制 造"国际化 韩国：依托龙头企业发展，建立和 完善支持制造业国际化的相关制度 纺织：全球需求不足影响国际化 我国制造业国际化已进入转型升级新阶段 我国制造业国际化发展的阶段划分 当前我国制造业国际化的新格局 我国制造业国际化分为起步期、快速发展期、均衡扩张期、转型升级 期四个阶段 | | 第一阶段 | | | | 第二阶段 | | | | 第三阶段 | | | | 第四阶段 | | | | | | | | | | | | --- | --- | --- ...

2025 年 3 月 17 日 第 12 期 总第 893 期 我国生产性服务业较快发展 背后仍需关注三大问题 生产性服务业是现代化产业体系的重要组成部分，是推动产 业结构升级、提升创新力和竞争力、促进经济高质量发展的重要 力量。基于 OECD 投入产出表分析发现，我国生产性服务业呈现 规模增长较快、占 GDP 比重稳步提升、结构持续优化三大亮点， 同时也存在总体发展不充分、知识密集型服务业比重偏低、对制 造业带动作用较弱等三大问题。建议从供需两侧同时发力，强化 生产性服务业要素供给保障，加快制造业产业链价值链升级，促 进生产性服务业与制造业深度融合。 - 1 - 一、我国生产性服务业发展呈现三大亮点 国家统计局并没有公开发布生产性服务业的统计数据，为进 行国际对比研究，本文将 OECD 投入产出表中的批发零售业、交 通运输业、信息与通信服务、金融保险业、专业与科技服务、行 政与辅助服务等六个行业界定为生产性服务业。 表 1 《生产性服务业统计分类（2019）》行业大类与 OECD 对照 | 国家统计局《生产性服务业统计分类（2019）》 | OECD 对照行业（国际标准 | 行业类型 | | --- | - ...

Investment Rating - The report suggests a positive outlook for the artificial intelligence industry, particularly highlighting the innovative breakthroughs of DeepSeek as a significant opportunity for investment and growth in the sector [1]. Core Insights - DeepSeek's innovative model architecture and training methods have achieved a low-cost and high-performance balance, potentially leading to a leap in China's AI industry and smart upgrades across various sectors [1][3]. - The company's open-source strategy has accelerated the dissemination and innovation of AI technology, fostering a more inclusive technological ecosystem [5][7]. - DeepSeek's rise has significantly boosted confidence in China's AI industry and its international influence, challenging existing beliefs about the necessity of powerful hardware for AI development [7][8]. Summary by Sections DeepSeek's Innovative Breakthroughs - DeepSeek has achieved a dual breakthrough in cost and performance through innovative model architecture, reducing development costs significantly. The V3 version's training cost was approximately $557.6 million, which is about 10% of Meta Llama 3.1 and 6% of OpenAI GPT-4o [3]. - The use of reinforcement learning techniques has enhanced model performance, with DeepSeek R1 showing exceptional results in complex tasks, rivaling top models like OpenAI-o1-1217 [3]. Impact on China's AI Development - DeepSeek's advancements have invigorated the AI sector in China, encouraging more companies to invest in cutting-edge technology and accelerating innovation [7]. - The international recognition of DeepSeek has elevated China's standing in the global AI landscape, with major tech companies like Microsoft and Amazon integrating DeepSeek into their platforms [7][8]. Opportunities for Industry Development - DeepSeek's efficient architecture reduces reliance on high-end computing facilities, providing opportunities for domestic chip manufacturers to validate and commercialize their technologies [8]. - The open-source nature of DeepSeek's offerings has created significant commercial value for cloud service providers, with major players like Alibaba Cloud and Huawei Cloud integrating DeepSeek into their services [8][9]. - The low pricing of DeepSeek's API services, significantly lower than market rates, lowers the barriers for AI application integration, stimulating market demand [9]. Recommendations for Further Development - The report emphasizes the need to strengthen foundational research and innovation capabilities in AI, encouraging interdisciplinary collaboration to explore new theoretical paradigms [12]. - It suggests enhancing public platforms and industry ecosystems to support small and medium enterprises in adopting AI technologies [14]. - The report advocates for optimized policy support and risk management mechanisms to foster innovation and ensure the sustainable development of the AI industry [15].

2025 年 2 月 10 日 第 4 期 总第 679 期 给CISA 主任的报告：开源安全 - 1 - 【译者按】2024 年 10 月，美国网络安全和基础设施安全局（CISA） 网络安全咨询委员会（CSAC）下设的技术咨询委员会（TAC）小组委员 会发布《给 CISA 主任的报告：开源安全》。报告认为，开源软件已在政 府部门、关键基础设施和公共生活中得到广泛应用，其漏洞将对下游的行 业领域产生广泛影响，各国政府均已高度重视开源安全。报告围绕开源软 件安全消费规范、安全消费规范鼓励措施、引导商业公司承担开源安全责 任以及维护人工智能开源安全四个问题，分别开展研究并提出建议。赛迪 智库信息化与软件产业研究所对报告进行了编译，期望对我国有关部门有 所帮助。 【关键词】开源软件 开源安全 安全消费规范 商业公司 人工智能 一、前言 美国网络安全和基础设施安全局（CISA）网络安全咨询委员 会（CSAC）于 2021 年 6 月正式成立。该委员会是一个独立的咨 询机构，围绕 CISA 网络安全使命相关的政策、计划、规划和培 训的制定、完善和实施，定期向 CISA 提供独立的、战略性的、 可操作的建议、咨询及报告。 ...