Investment Rating - Rating: Outperform [1] - Target Price: RMB 114.94 [1] Core Insights - The company focuses on AI and cloud computing, with both sectors contributing nearly half of its revenue. The launch of the AICP platform and products like XaaS and security GPT are key developments [1][5] - The company is expected to maintain growth driven by increased demand for cloud services and a turning point in cybersecurity, with projected EPS for 2025-2027 at RMB 0.77, RMB 1.18, and RMB 1.54 respectively [5][15] - The report highlights a recovery in growth for Q1 2025, with significant year-on-year revenue increases and a focus on enhancing product quality and innovation [5][16] Financial Summary - Revenue projections show a slight decline in 2024 to RMB 7.52 billion, followed by growth to RMB 8.08 billion in 2025, and reaching RMB 10.07 billion by 2027 [3][6] - Net profit attributable to shareholders is forecasted to increase significantly from RMB 197 million in 2024 to RMB 649 million by 2027, reflecting a strong growth trajectory [3][6] - The company’s gross profit margins for security and cloud services are 79.31% and 44.24% respectively, indicating robust profitability in these segments [5][17] Strategic Focus - The company plans to enhance its product offerings in XaaS and AI, while also expanding its market reach to key industry clients and overseas markets [5][19] - Future strategies include focusing on key production lines, promoting innovation in AI and cybersecurity, and strengthening organizational capabilities [5][20] - The company aims to deepen its channel system and improve management efficiency to boost profitability over the next 3-5 years [5][20]

特别提示： | 证券代码：300454 | 证券简称：深信服 | 公告编号：2025-046 | | --- | --- | --- | | 债券代码：123210 | 债券简称：信服转债 | | 深信服科技股份有限公司 关于信服转债恢复转股的提示性公告 本公司及董事会全体成员保证信息披露的内容真实、准确、完整，没有虚假 记载、误导性陈述或重大遗漏。 董事会 二〇二五年六月十二日 1 / 1 1、债券代码：123210 债券简称：信服转债； 2、转股起止日期：2024 年 2 月 2 日至 2029 年 7 月 26 日； 3、暂停转股日期：2025 年 6 月 5 日至 2025 年 6 月 12 日； 4、恢复转股日期：2025 年 6 月 13 日。 深信服科技股份有限公司（以下简称"公司"）因实施 2024 年度权益分派， 根据《深信服科技股份有限公司向不特定对象发行可转换公司债券募集说明书》 及相关规定，"信服转债"自 2025 年 6 月 5 日至本次权益分派股权登记日止暂 停转股，具体内容详见公司于 2025 年 6 月 4 日 在 巨 潮 资 讯 网 （http://www.cninfo.co ...

Industry Rating - The industry rating is maintained at "Outperform" [1] Core Insights - The report suggests that the years 2025-2026 are expected to mark the commercialization of AI Agents, driven by advancements in large model capabilities and the introduction of key players like OpenAI, Anthropic, Microsoft, and Google [2][3] - The total addressable market (TAM) for AI Agents is estimated at approximately 3.61 trillion yuan, with significant opportunities in vertical industries such as IT and finance [2][3] - AI Agents are projected to replace specific standardized roles in various sectors, with a focus on enhancing productivity and ROI for businesses [3][48] Summary by Sections Investment Rating - The report maintains a rating of "Outperform" for the industry [1] Market Potential - The total addressable market (TAM) for AI Agents is estimated at 3.61 trillion yuan, calculated based on total salary levels, replacement rates, and efficiency optimization [50][54] - The report highlights six core scenarios where AI Agents are expected to penetrate first, including coding, banking, human resources, legal, cybersecurity, and customer service [61][62] Key Industry Insights - AI Agents are anticipated to significantly impact the banking sector, with 54% of jobs having a high potential for automation, leading to an estimated TAM of 1774.32 billion yuan in this field [71][77] - In the coding sector, AI is expected to automate up to 99% of coding tasks by the end of 2025, with a TAM of approximately 4357.84 billion yuan [63][68] - The report emphasizes that AI Agents will first replace roles in high-salary, white-collar industries such as finance, IT, and law, where the ROI for AI implementation is more favorable [55][57] Company Recommendations - The report suggests monitoring companies involved in general office applications and enterprise SaaS, as well as those in vertical markets like banking, human resources, and cybersecurity [3][4]

Core Viewpoint - The digital economy theme index has experienced a decline, with significant drops in key component stocks, indicating a challenging market environment for the sector [1][2]. Group 1: Market Performance - As of June 10, 2025, the CSI Digital Economy Theme Index (931582) fell by 1.99%, with major declines in stocks such as Deepin Technology (300454) down 7.14% and Yonyou Network (600588) down 6.22% [1]. - The Digital Economy ETF (560800) also decreased by 1.98%, with a latest price of 0.74 yuan and a trading volume of 21.44 million yuan [1]. - The Digital Economy ETF has reached a recent high in scale at 788 million yuan and a recent high in shares at 1.042 billion, ranking in the top half of comparable funds [2]. Group 2: Investment Trends - Shenwan Hongyuan suggests that a structural bull market in A-shares may require significant technological catalysts, particularly in AI, embodied intelligence, and defense industries [1]. - The report highlights key products to watch in the second half of the year, including DeepSeekR2, GPT5, and the mass production version of Tesla's Optimus [1]. - There is a noted increase in leveraged funds, with the Digital Economy ETF seeing net inflows of 13.6 million yuan and a total of 12.12 million yuan over the last five trading days [2]. Group 3: Key Holdings - As of May 30, 2025, the top ten weighted stocks in the CSI Digital Economy Theme Index accounted for 50.98% of the index, with notable companies including Dongfang Wealth (300059) and SMIC (688981) [3][5]. - The performance of these key stocks varied, with Dongfang Wealth down 2.47% and Hikvision (002415) down 1.55%, while Haiguang Information (688041) saw an increase of 2.86% [5].

Group 1: Global AI Outlook - The report highlights a new paradigm in AI development characterized by new scaling, new architecture, and new total addressable market (TAM) opportunities [1] - The demand for computing power is expected to rise due to advancements in both training and inference processes, potentially unlocking new TAMs [1][3] - The report maintains a positive outlook on AI industry investments, anticipating that global AI applications will enter a performance harvesting phase [1] Group 2: Model Development - The pre-training scaling law is anticipated to open a new starting point for model development, with significant innovations in architecture being explored [2][23] - The report notes that the classic transformer architecture has reached a parameter scale bottleneck, with existing public data nearly exhausted [2][20] - Major tech companies are experimenting with new architectures, such as Tencent's Hunyuan TurboS and Google's Gemini Diffusion, which may accelerate scaling law advancements [23][24] Group 3: Computing Power Demand - The report identifies a clear long-term upward trend in computing power demand, driven by both training and inference needs [3][32] - New scaling paths are emerging in the post-training phase, with ongoing exploration of new architectures that may reignite pre-training demand narratives [3][33] - The deployment of large-scale computing clusters, such as OpenAI's StarGate, is expected to support the exploration of pre-training [38] Group 4: Application Development - The report indicates that the rapid advancement of agent applications is leading to a performance harvesting phase for global AI applications [4][67] - The commercialization of agent products is accelerating, with domestic AI applications quickly iterating and entering the market [4][67] - The report emphasizes that agent applications are evolving from simple tools to complex solutions, with significant growth expected in various sectors [5][68] Group 5: Business Model Transformation - The shift from traditional software delivery to outcome-based delivery is highlighted as a key trend, with quantifiable ROI accelerating the adoption of agent applications [5] - Specific sectors such as consumer-facing scenarios (advertising, e-commerce) and AI in marketing/sales are expected to lead in commercialization due to their inherent advantages [5][67] - The report notes that AI applications in HR are transitioning from efficiency tools to strategic hubs, indicating a broader transformation in business models [5][67]

Group 1 - The market experienced fluctuations on June 6, with mixed performance across the three major indices [1] - The Computer ETF (159998) saw a slight decline of 0.11% at the close, with a premium rate of 0.88% and a trading volume of nearly 100 million yuan [1] - The Computer ETF has attracted significant capital inflow, with over 54 million yuan net inflow on June 5, marking the first position among similar products [1] Group 2 - The Computer ETF tracks the CSI Computer Theme Index, which includes stocks from companies involved in information technology services, application software, system software, and computer hardware [1] - As of March 31, the top ten holdings of the ETF included leading companies such as Hikvision, Zhongke Shuguang, and Keda Xunfei, with Zhongke Shuguang accounting for 6.86% of the fund's net value [1] - According to a report from Shenwan Hongyuan Securities, the computer industry is currently at a historical low in terms of valuation and profitability, similar to the telecommunications sector before its 2019 rally [2] Group 3 - The computer industry's return on equity (ROE) has been in a downtrend since 2015, with a TTM ROE of 1.7% as of Q1 2025, reaching a historical low [2] - The PB ratio stands at 2.6, which is at the 40th percentile since 2010, indicating potential for recovery as trends in artificial intelligence and domestic innovation progress [2] - The outlook for the computer and media sectors suggests a potential rebound in ROE as AI penetration increases and content industries are restructured [2]

Group 1 - The core viewpoint of the news highlights the positive performance of the artificial intelligence sector, with the China Securities Artificial Intelligence Industry Index showing an increase of 0.14% and several constituent stocks experiencing significant gains [1] - The artificial intelligence ETF (515980) has seen a trading volume of 72.55 million yuan and a recent average daily trading volume of 125 million yuan over the past month, indicating strong market interest [3] - The total scale of the artificial intelligence ETF has reached 3.538 billion yuan, with a total inflow of 53.59 million yuan over the last 22 trading days, suggesting robust investor confidence [3] Group 2 - The upcoming Global Artificial Intelligence Technology Conference (GAITC 2025) in Hangzhou will feature prominent figures from academia and industry, focusing on AI technology breakthroughs and industry trends [3] - The computer sector has experienced a new wave of growth driven by AI and self-controllable themes, despite facing fundamental pressures due to limited downstream spending [4] - UBS notes that international market interest in Chinese assets is increasing, particularly in new productivity sectors like artificial intelligence and high-end manufacturing, which are reshaping the narrative of the Chinese stock market [4] Group 3 - The artificial intelligence ETF is currently the only product tracking the artificial intelligence industry index, providing investors with exposure to high-growth potential companies in the sector [5] - Investors can also consider the Hua Fu Artificial Intelligence ETF linked funds to gain access to the artificial intelligence sector [5]

| 证券代码：300454 | 证券简称：深信服 | 公告编号：2025-045 | | --- | --- | --- | | 债券代码：123210 | 债券简称：信服转债 | | 深信服科技股份有限公司 关于信服转债转股价格调整的公告 本公司及董事会全体成员保证信息披露的内容真实、准确、完整，没有虚假 记载、误导性陈述或重大遗漏。 重要提示： 1.债券代码：123210 债券简称：信服转债 2.调整前转股价格：110.97 元/股 3.调整后转股价格：110.91 元/股 4.转股价格调整生效日期：2025 年 6 月 13 日 一、 可转债的基本情况 (一)可转债发行上市情况 （3）因实施 2023 年年度权益分派，"信服转债"的转股价格由 111.31 元/股 调整为 111.26 元/股，调整后的转股价格自 2024 年 6 月 5 日起生效。 （4）公司 2022 年度限制性股票激励计划预留授予部分（第二批）的限制性 股票于 2024 年 11 月 6 日上市流通，公司总股本由 419,848,962 股增加至 419,970,712 股。"信服转债"的转股价格由 111.26 元/股调整为 1 ...

| 证券代码：300454 | 证券简称：深信服 公告编号：2025-044 | | --- | --- | | 债券代码：123210 | 债券简称：信服转债 | 深信服科技股份有限公司 2024 年年度权益分派实施公告 本公司及董事会全体成员保证信息披露的内容真实、准确、完整，没有虚假 记载、误导性陈述或重大遗漏。 特别提示： 1、公司回购专用证券账户中的股份（5,148,347 股）不享有利润分配权利， 不参与本次权益分派。本次权益分派将以公司现有总股本（421,881,295 股）剔除 已回购股份后的 416,732,948 股为基数。本次现金分红总额（税前）=实际参与现 金分红的股本×每 10 股分红金额÷10 股=416,732,948 股×0.63 元÷10 股= 26,254,175.72 元。 2、本次权益分派实施后，根据股票市值不变原则，实施权益分派前后公司 总股本保持不变，现金分红总额分摊到每一股的比例将减小。因此，本次权益分 派实施后除权除息价格计算时，按总股本折算每 10 股现金红利=现金分红总额÷ 总股本（含已回购股份）×10 股=26,254,175.72 元÷421,881,2 ...

Group 1 - The Industrial Internet ETF (159778.SZ) increased by 0.61%, with its associated index, Industrial Internet (931495.CSI), rising by 0.42% [1] - Key constituent stocks such as Industrial Fulian rose by 2.99%, GoerTek by 2.13%, Inspur Information by 1.78%, State Grid NARI by 0.91%, and Deepin Technology by 3.06% [1] - CITIC Securities research report indicates that the commercialization of L4 level autonomous driving is accelerating, with leading Robotaxi companies experiencing a 400% year-on-year increase in weekly orders [1] Group 2 - Changjiang Securities highlights that distributed low-orbit constellations built using commercial aerospace capabilities possess strong anti-jamming and anti-destruction capabilities, which are fundamental drivers of defense demand in commercial aerospace [2] - The satellite wireless interconnection market is projected to have significant potential, with estimates that by 2030, the domestic "Thousand Sails" and GW constellation networking will create a market space exceeding 270 billion yuan for satellite manufacturing and 80 billion yuan for rocket launch services [2] - Shenwan Hongyuan research shows that the midstream structural components/subsystem segment of the military industry chain saw a 12.09% year-on-year increase in advance payments and contract liabilities in Q1 2025, while the downstream assembly segment's inventory growth rate reached 17.73%, indicating a sequential transmission of industry chain prosperity [2]