Group 1 - The company's ability to inject computing power industry business depends on the success of bankruptcy reorganization, compliance with relevant laws and regulations, and obtaining necessary approvals [1] - There is significant uncertainty regarding the timing, scale, and future profitability of the asset injection, with potential risks including intensified market competition leading to overall industry profit margin decline and extended cost recovery periods [1] - The proposed donation of the Halinger Intelligent Computing Center by industry investors has not yet been completed, posing risks of insufficient future funding and extended construction timelines [1] Group 2 - Increased market competition may result in underutilization of cabinet space at the Halinger Intelligent Computing Center, potentially leading to lower-than-expected profitability [1] - The income and profit scale of industry investors are relatively small, and there are risks associated with the company's lack of qualifications to undertake new orders from the same controlling entity [1] - Although the reorganization investment agreement has been signed, there are still risks related to the inability of the reorganization investors to fulfill their investment obligations due to insufficient funding [1]

Core Viewpoint - The ability of *ST Dongyi to inject computing power industry business depends on the success of bankruptcy reorganization, compliance with relevant laws and regulations, and obtaining necessary approvals [1] Summary by Relevant Sections - **Asset Injection Uncertainty** - The timing, scale, and future profitability of the asset injection are highly uncertain [1] - Potential risks include intensified market competition leading to overall industry profit margin decline and extended cost recovery periods [1] - **Investment Risks** - The proposed donation of the Halinger Intelligent Computing Center by industry investors has not yet been completed, posing risks of insufficient future funding and extended construction timelines [1] - Increased market competition may result in underutilization of cabinet space at the Halinger center, potentially leading to lower-than-expected profitability [1] - **Investor and Order Risks** - The revenue and profit scale of industry investors are relatively small, and the company may lack the qualifications to undertake new orders from the same controlling entity, Chang Kong Construction [1] - Risks include failure to introduce orders according to the operational plan and overall decline in order scale and profit levels due to intensified market competition [1] - **Reorganization Agreement Risks** - Although the investment-related agreement for the reorganization has been signed, there are risks that the investors may not fulfill their funding obligations as per the agreement [1] - There is a possibility that the agreement could be terminated, revoked, deemed ineffective, or unfulfilled [1]

Core Viewpoint - The Ministry of Industry and Information Technology (MIIT) is soliciting opinions on the draft of the "Guidelines for the Construction of Computing Power Standard System (2025 Edition)", aiming to establish over 50 standards by 2027 to enhance the computing power standard system [1] Group 1: Standards Development - The guidelines focus on various aspects including basic general standards, computing power facilities, computing power equipment, computing network integration, computing power interconnection, computing power platforms, computing power applications, computing power security, and green low-carbon initiatives [1] - The initiative aims to promote the development of a standard system that supports innovation and development in enterprises, with over 500 companies expected to participate in the promotion and implementation of these standards [1] Group 2: Collaboration and Innovation - The guidelines emphasize the importance of collaboration among research institutions, industry associations, and industrial alliances to develop standards, fostering communication and cooperation in technology innovation, industrial promotion, infrastructure construction, and talent cultivation [1] - The initiative is designed to effectively support the construction of a national integrated computing power network, enhancing the overall capabilities of the industry [1]

Core Viewpoint - The digital economy theme index and related ETFs are showing positive performance, driven by advancements in domestic AI and computing power chips, indicating a strong growth trajectory for the sector [1][2]. Group 1: Market Performance - As of October 21, 2025, the CSI Digital Economy Theme Index (931582) increased by 0.94%, with notable gains from companies such as Jingsheng Electronics (4.67%) and Zhaoyi Innovation (4.62%) [1]. - The Digital Economy ETF (560800) rose by 0.70%, with a trading volume of 2.4424 million yuan and a turnover rate of 0.36% [1]. - The average daily trading volume for the Digital Economy ETF over the past month was 28.5447 million yuan, with the latest share count reaching 684 million [1]. Group 2: Industry Insights - CITIC Securities highlights that domestic large models are rapidly iterating, and domestic computing power chips are achieving seamless adaptation, fostering a closed-loop ecosystem that supports the continuous development of domestic AI [1]. - Zheshang Securities projects that by 2025, domestic computing power will rise, driven by high capital expenditures from major companies and a growing consensus on computing power localization [2]. - The CSI Digital Economy Theme Index includes companies involved in digital economy infrastructure and high digitalization applications, reflecting the overall performance of listed companies in this sector [2]. Group 3: Key Constituents - As of September 30, 2025, the top ten weighted stocks in the CSI Digital Economy Theme Index accounted for 54.31% of the index, including companies like Dongfang Wealth (10.51%) and SMIC (6.34%) [2][4].

Group 1 - The report highlights opportunities in long-term urban investment bonds, particularly in regions supported by new policy financial tools, which are expected to enhance local economic development and project financing [2][9][12] - Specific regions such as Wenzhou Yongjia County, Nanning, and Changji Prefecture are identified as key areas for investment due to their potential to stabilize employment and attract bank loans and social capital [2][23] - The report emphasizes the importance of addressing hidden debts and overdue payments in local governments, particularly in cities like Jilin City, which are expected to receive special bonds for project construction [2][24] Group 2 - The report notes that during the 14th Five-Year Plan period, certain areas are expected to become focal points for national strategic industries, including logistics hubs and computing power centers, which will receive significant government support [3][10][19] - Cities such as Xining, Qingyang, and Karamay are highlighted for their potential to form industrial clusters and attract investment in long-term bonds due to their strategic importance in future industries [3][25] - The widening credit spreads for urban investment bonds with maturities over five years are noted, suggesting a potential investment opportunity in specific bonds from regions mentioned [3][25] Group 3 - The report discusses the recent recovery in the bond market, with a general decline in yields, particularly in credit bonds, which have seen a more significant drop compared to government bonds [4][5] - It suggests that short- to medium-term credit bonds may offer better value as safe-haven assets in the current uncertain market environment [4][5] - The report recommends focusing on high-grade urban investment bonds as core assets, particularly those with a maturity of 2 years or less, while also considering opportunities in the primary market [5][4]

Core Viewpoint - OpenAI and Broadcom have announced a strategic partnership to develop and deploy custom AI chips and computing systems with a total power consumption of 10GW over the next four years, addressing the growing demand for large-scale computing [1] Group 1: AI Chip Development - OpenAI aims to advance its first self-developed AI chip to meet increasing large-scale computing needs [1] - The collaboration is expected to encourage other internet giants to follow suit in investing in computing power [1] Group 2: National Initiatives - On October 14, the Ministry of Industry and Information Technology launched a special initiative for "millisecond computing" focusing on the development of computing networks [1] - The initiative aims to create a high-speed, high-capacity, low-latency metropolitan network, providing millisecond-level computing resource access across various applications [1] Group 3: Market Outlook - CITIC Construction Investment Securities believes that the demand for computing power driven by AI remains strong and the industry outlook is positive [1] - Despite potential short-term market fluctuations and adjustments, the long-term perspective remains optimistic, recommending a focus on leading companies in the sector [1]

Core Viewpoint - The demand for computing power driven by AI is strong and the industry remains optimistic despite potential short-term fluctuations due to market volatility, tariff issues, chip structure, and valuation adjustments [1] Group 1 - The computing power industry is experiencing robust demand and sustained prosperity due to AI [1] - Short-term market fluctuations may lead to some adjustments in the sector [1] - The recommendation is to adopt a medium to long-term perspective and favor leading companies in the industry [1]

Core Insights - Since the "14th Five-Year Plan," China has initiated a new round of digital infrastructure development, establishing a national integrated computing network that connects various regions, positioning itself as the second-largest computing power globally, which supports the growth of the digital economy and creates new opportunities for growth [1][8][16] Group 1: Digital Infrastructure Development - The total computing power in China has seen an annual growth rate of 30% since the "14th Five-Year Plan," with over 70% of the new computing power coming from the western regions [8] - The "East Data West Computing" project has established eight national computing hub nodes and ten data center clusters, with over 60% of cities able to access a computing cluster within 5 milliseconds [9][11] - The construction of the largest and most extensive network infrastructure globally has made China one of the countries with the richest data application scenarios [16] Group 2: Economic Impact and Growth - In Gansu's Qingyang, the digital economy has transformed, with new enterprises generating nearly 10 billion yuan in revenue, showcasing the rapid growth of the digital economy in previously underdeveloped areas [13] - The emergence of over 30 "computing new cities" across China during the "14th Five-Year Plan" period has become a critical support for the artificial intelligence era, facilitating various applications from animation rendering to intelligent driving model training [18] - The digital revolution is redefining agriculture and manufacturing, with automated systems and real-time data collection enhancing productivity and efficiency [23][26]

Core Insights - The Chinese computing power industry is receiving significant policy support, marking a transition from the first phase of scale expansion to a second phase focused on deep integration with the real economy and value creation [1] - The national strategy emphasizes the construction of new information infrastructure and the application of computing power in various industries, with a focus on security and data protection [1][10] Policy and Infrastructure Development - On October 11, a joint implementation plan was released by multiple government departments to promote service-oriented manufacturing innovation, setting a new direction for the computing power industry [1] - Shanghai has responded with its own action plan for high-quality development of the intelligent terminal industry, aiming to create a cluster for intelligent computing servers and promote the large-scale application of core components [1] - The establishment of a national computing power network is underway, with various cities forming alliances to meet regional computing demands [3][4] Industry Growth and Company Initiatives - Companies like Dongfang Guoxin and Qidi Design are actively investing in intelligent computing center construction, with significant project investments reported [5][6] - The computing power infrastructure is expanding rapidly, with cities like Ceyuan and Langfang leading in data center operations and service provision [4][5] Technological Advancements - The domestic computing power industry is experiencing a historic shift from "usable" to "well-used" due to policy guidance and technological breakthroughs [7] - Major companies like Huawei are advancing their AI chip development, with plans for continuous product iterations and high-performance cluster solutions [7][8] AI Commercialization and Industry Integration - The integration of computing power with various industries is accelerating, with a focus on creating high-quality industry data sets and promoting the commercialization of AI applications [10][11] - Companies are forming strategic partnerships to enhance AI capabilities in sectors such as finance, education, and security, indicating a trend towards deeper industry integration [11][12] Future Outlook - The demand for computing power is expected to grow significantly as AI applications become more widespread, with emerging sectors like low-altitude economy poised for rapid development [13]

Industry Insights - The Ministry of Industry and Information Technology (MIIT) has launched a special initiative called "Millisecond Computing" aimed at developing a high-capacity, low-latency metropolitan network that provides millisecond-level computing resources. The goal is to achieve interconnectivity of computing centers within 1 millisecond, resource access in key locations within 1 millisecond, and application accessibility with network latency under 10 milliseconds by 2027 [1] - Dongwu Securities highlights that the AI computing power sector is experiencing rapid iterations in hardware and software collaboration, which is expected to lower the cost of computing power significantly within 12-18 months. This shift is anticipated to enable developers to utilize more complex models and systems, thereby enhancing the parameters, context, and parallelism of AI applications [1] Company Highlights - Yuanjie Technology focuses on the research, design, production, and sales of optical chips, primarily serving the telecommunications and data center markets [2] - Cambridge Technology specializes in the development, production, and sales of terminal devices and high-speed optical modules for telecommunications, data communications, enterprise networks, and home networks [2]