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]

Market Overview - The global market has experienced significant fluctuations due to trade tensions and the impact of Trump's tariffs, leading to a notable depreciation of the US dollar and high volatility in US Treasury yields [10][11] - As of June 1, 2025, the tactical asset allocation strategy by Shenwan Hongyuan has yielded a 4.8% increase year-to-date, outperforming the strategic allocation by 0.8%, primarily through over-allocating to Hong Kong stocks and gold while under-allocating to oil and the US dollar [10][11] Investment Strategy - The report emphasizes a PB-ROE dual low reversal strategy, highlighting that since April 2025, global trade tensions have led to significant market volatility, with Japanese, German, and US markets rising by 5-10% compared to pre-tariff levels, while Chinese asset prices have slightly declined [12] - The report suggests that the risk appetite in the A+H markets has decreased, favoring dividend assets, with consumer, pharmaceutical, and banking sectors showing strong performance, while the TMT sector lags behind [12] Sector Analysis - The report identifies several sectors with potential investment opportunities, including consumer goods, pharmaceuticals, and banking, which have shown positive growth, while the TMT sector remains under pressure [12][15] - The analysis indicates that sectors like healthcare and certain consumer products have stabilized after a prolonged downturn, while real estate and liquor sectors have yet to show significant recovery [14][15] Future Outlook - The report anticipates that the restructuring of the global financial order will provide opportunities for the internationalization of the Renminbi and the revaluation of Renminbi assets, particularly in the context of the Belt and Road Initiative [10][11] - It also highlights the importance of strategic asset allocation, suggesting that non-US assets may offer better long-term returns as the dollar's credibility declines [10][11] Technological Innovations - The report discusses the potential of new hardware innovations, particularly in the context of AI and semiconductor advancements, which are expected to drive significant changes in the technology landscape [22] - Key areas of focus include GPU+HBM, optical devices, silicon photonics, and autonomous vehicle technologies, which are anticipated to present substantial investment opportunities in the near to medium term [22]

资金流向方面，今日发生大宗交易的科创板股中，有5股获主力资金净流入，净流入资金居前的有爱科 赛博、华特气体、皓元医药等，净流入资金为2321.59万元、1292.52万元、872.28万元，净流出资金居 前的有骄成超声、中控技术、虹软科技等，净流出资金为1782.55万元、1403.24万元、969.20万元。 （数据宝） | 代码 | 简称 | 成交笔 | 大宗交易数量 | 平均成交价格 | 相对收盘价折溢价 大宗交易金额 | | --- | --- | --- | --- | --- | --- | | | | 数 | （万股） | （元） | （%） （万元） | | 688268 | 华特气 | 23 | 233.00 | 44.81 | -9.33 10440.78 | | | 体 | | | | | | 688069 | 德林海 | 4 | 226.00 | 14.50 | -23.24 3277.00 | | 688131 | 皓元医 | 4 | 49.90 | 41.00 | -11.22 2045.90 | | | 药 | | | | | | 688458 | 美芯晟 | 2 | 30.27 ...

消息面上，继深圳市设立两只人工智能产业基金之后，山东省印发《关于支持人工智能全产业链创新发 展的若干政策措施》，要求发挥省新旧动能转换基金引领带动作用，吸引社会资本共同构建人工智能产 业基金群，以"全生命周期基金"体系，拓展人工智能企业市场化融资渠道。山东人工智能产业保持年均 30%以上高速增长，集聚骨干企业近1000家，带动相关产业规模超千亿元。 相关产品：科创板人工智能ETF（588930） 6月9日，A股市场震荡上行，人工智能板块走势积极。科创板人工智能指数成份股中，海天瑞声涨超 9%，虹软科技、乐鑫科技、金山办公涨超3%，中邮科技、中科星图、道通科技、有方科技涨超2%。科 创板人工智能ETF（588930）现涨0.52%，实时成交额突破5000万元，过去20个交易日资金净流入超 7000万元，市场热度较高。 （文章来源：每日经济新闻） 科创板人工智能ETF（588930）跟踪的科创板人工智能指数布局30只科创板人工智能龙头，覆盖AI产业 链上游算力芯片、中游大模型云计算、下游机器人等各类创新应用，聚焦电子、计算机、机械设备、家 电、通信五大行业，前五大成份股合计权重47%，或具有较高的AI主题纯度和更 ...

Group 1 - The AI sector is experiencing a rise, with the Huaxia Sci-Tech AI ETF (589010) increasing by 0.29% as of 10:05 AM, led by Hai Tian Rui Sheng's 7.18% gain [1] - Over the past five days, the Huaxia Sci-Tech AI ETF has risen by 4.14%, aiming for a five-day consecutive increase [1] - Meta is considering a significant investment in the AI data labeling startup Scale AI, potentially exceeding $10 billion, which could become one of the largest private financing rounds in history [1] Group 2 - Huatai Securities emphasizes that Agent products must deeply integrate with vertical industry and scenario data for successful commercialization [2] - The AI application in marketing and sales has transitioned from exploration to large-scale implementation, making it one of the fastest-growing areas for AI model deployment [2] - The latest TTM price-to-earnings ratio for the Huaxia Sci-Tech AI ETF is 119.73, positioned at the 22.71% percentile, indicating a low valuation compared to the past year [2]

Core Viewpoint - The company is entering a phase of implementation for its research outcomes, with an anticipated turning point in performance driven by advancements in video codec technology and strategic partnerships [1][2]. Group 1: Video Codec Technology - The company's video codec technology serves as a competitive moat, with leadership in industry standard formulation and compatibility with heterogeneous computing architectures [1]. - The integration of DeepSeek enhances the understanding capabilities of the company's BlackEye model, leading to the launch of multiple DeepSeek integrated machines that contribute to new growth [1]. - The company is progressing well with key research projects, including in-vehicle video processing technology and AI-generated video content, which are now in the product testing phase [1]. Group 2: Automotive Sector - In 2024, the revenue share from in-vehicle smart cockpit products is expected to rise to 15%, driven by consumer interest in smart cockpit features [2]. - The company collaborates with major automotive manufacturers to develop integrated content and middleware solutions, enhancing the visual experience of smart cockpits [2]. - Partnerships with content providers like Zhejiang Radio and Television Group and iQIYI are expected to diversify revenue streams through new models such as subscription and advertising [2]. Group 3: Industrial and Satellite Applications - The demand for high-definition video transmission and storage is projected to accelerate in the AI era, with over 74% of China's internet traffic being video-related [3]. - The company is one of the few in China with capabilities in ultra-low latency video encoding and AI recognition, serving various sectors including smart transportation and energy [3]. - The integration of advanced models like DeepSeek significantly enhances the company's product capabilities, with faster market deployment anticipated [3]. Group 4: Media and Culture - The company's primary revenue source, expected to account for approximately 61% of total revenue in 2024, benefits from the increase in high-definition channel content [4]. - The company is deeply involved in the formulation of ultra-high-definition industry standards and has supported major events like the Beijing Winter Olympics and the Qatar World Cup [4]. - AI technology is being applied in video production processes, enhancing content quality and expanding business opportunities [4]. Group 5: Financial Projections - Revenue forecasts for the company from 2025 to 2027 are projected at 439 million, 587 million, and 735 million yuan, reflecting year-on-year growth rates of 43.01%, 33.93%, and 25.14% respectively [5]. - The net profit attributable to the parent company is expected to reach 12 million, 34 million, and 59 million yuan during the same period, with significant growth rates of 111.59%, 187.25%, and 73.53% [5]. - The company is valued at a price-to-sales ratio (PS) of 8, 6, and 5 for the years 2025 to 2027, indicating a favorable position compared to peer companies [5].

Investment Rating - The report maintains a positive outlook on the computer sector, indicating a recovery in industry prosperity driven by AI applications and technological independence [3][4]. Core Insights - 2025 is projected to be a pivotal year for AI applications, with significant advancements expected following the release of ChatGPT in late 2022. The report emphasizes that AI will become a key driver of business growth, enhancing efficiency and creating new demand [3][4]. - The report highlights the urgency for technological self-reliance in China, particularly in light of international trade tensions and tariffs that could impact the global technology supply chain [3][4]. - The computer sector is currently in its fourth bullish phase over the past decade, with a notable recovery in sentiment and performance expected [8][18]. Summary by Sections 1. Historical Overview of the Computer Sector - The computer sector has experienced four major bullish phases since 2013, with the current phase starting in September 2024 [8][18]. 2. Current Position of the Computer Sector - Institutional holdings in the computer sector remain low, with levels fluctuating around 2-6% over recent years. As of Q1 2025, holdings have slightly increased to 2.9% but are still considered low historically [15][18]. - The valuation levels of the SW computer index are at historical highs, indicating potential for further growth [18][21]. 3. Promising Sub-sectors and Logic for 2025 - The report identifies AI, technological independence, and the impact of tariffs as key areas of focus for investment in 2025. It notes that AI is transitioning from a follower to a leader in the global landscape, with domestic models closing the gap with international counterparts [38][43]. - The report anticipates that 2025 will mark the commercial rollout of AI agents, with significant advancements in AI capabilities expected [43][46]. - The report discusses the potential for mergers and acquisitions in the computer sector, driven by regulatory changes that may stimulate market activity similar to the 2013-2015 period [21][22]. 4. Company Performance Predictions and Valuations - The computer sector is expected to see a revenue increase in 2025, with a projected total revenue of 12,693.99 billion yuan in 2024, reflecting a year-on-year growth of 4.78%. However, net profits are expected to decline significantly [25][29]. - The report highlights that large-cap companies are showing more stable performance compared to small-cap companies, with significant growth in revenue and profits anticipated for large-cap firms in Q1 2025 [29][34].

Core Viewpoint - The launch of the real-time video call feature by Doubao marks a significant milestone in the commercialization of multi-modal perception technology, which is expected to drive demand for upstream computing chips and AI terminal SoC chips [3]. Group 1: Market Performance - As of May 28, 2025, the Shanghai Stock Exchange Sci-Tech Innovation Board Artificial Intelligence Index (950180) increased by 0.41%, with notable gains from companies such as Chipone Technology (688521) up 2.87% and Stone Technology (688169) up 2.14% [2]. - The Sci-Tech AI ETF (588790) rose by 0.71%, with a latest price of 0.57 yuan and a turnover rate of 3.68%, totaling a transaction volume of 113 million yuan [2]. - Over the past two weeks, the Sci-Tech AI ETF saw a significant scale increase of 85.04 million yuan, ranking second among comparable funds [3]. Group 2: Fund Flows and Performance - The latest net inflow for the Sci-Tech AI ETF was 18.72 million yuan, with a total of 67.48 million yuan net inflow over the last five trading days, averaging 13.50 million yuan per day [3]. - The fund has experienced continuous net purchases from leveraged funds for five consecutive days, with the highest single-day net purchase reaching 12.64 million yuan [3]. Group 3: Fund Characteristics - Since its inception, the Sci-Tech AI ETF has achieved a maximum monthly return of 15.59%, with an average monthly return of 15.59% and a winning probability of 60.53% [4]. - The fund's management fee is 0.50% and the custody fee is 0.10%, which are among the lowest in comparable funds [4]. - The latest price-to-earnings ratio (PE-TTM) of the index tracked by the Sci-Tech AI ETF is 116.77, indicating a valuation lower than 83.1% of the historical data over the past year [4]. Group 4: Top Holdings - As of April 30, 2025, the top ten weighted stocks in the Shanghai Stock Exchange Sci-Tech Innovation Board Artificial Intelligence Index include companies like Lattice Semiconductor (688008) and Cambricon Technologies (688256), with the top ten stocks accounting for a significant portion of the index [5][8].

Investment Rating - The report does not provide a specific investment rating for the computer industry, but it indicates a neutral rating for the computer software sub-industry [3]. Core Insights - The report highlights the rapid progress in the Robotaxi industry, driven by improved policies and regulations, decreasing costs, and accelerated commercialization efforts by major players [5][6][7][24]. - The overall performance of the computer industry index has declined by 3.02% in the week of May 19-23, 2025, ranking last among 31 primary industries [33]. Summary by Sections 1. Robotaxi Industry Developments - Policies and regulations related to Robotaxi have been rapidly established in China, including guidelines for autonomous vehicle operation and local trials in cities like Beijing and Wuhan [5][12]. - The total lifecycle operating costs of Robotaxi are decreasing, with significant reductions in manufacturing costs due to mass production of components like chips and sensors, exemplified by a 70% cost reduction in the latest Robotaxi models [6][18]. - Major Robotaxi companies are accelerating their commercialization efforts, with Waymo reporting a service volume of 250,000 rides per week and Baidu's Robotaxi achieving 100% unmanned operation [7][27]. 2. Market Performance - The computer industry index has seen a slight decline of 0.11% year-to-date, outperforming the CSI 300 index by 1.22 percentage points [33]. - The report lists the top five gainers and losers in the computer industry for the week, with significant fluctuations in stock prices [35]. 3. Investment Recommendations - The report recommends focusing on companies such as Wanma Technology, Hongsoft Technology, and Ruiming Technology due to the rapid advancements in the Robotaxi sector [36].

Group 1 - The A-share market saw a slight decline in the three major indices, with the Shanghai Stock Exchange Sci-Tech Innovation Board Artificial Intelligence Index experiencing a minor pullback [1] - The Sci-Tech Innovation Board Artificial Intelligence ETF (588930) opened lower but slightly rebounded, with a narrowing decline of 1.04% and a trading volume exceeding 10 million yuan [1] - The index comprises 30 large-cap listed companies that provide foundational resources, technology, and application support for artificial intelligence, reflecting the overall performance of representative AI industry stocks in the Sci-Tech Innovation Board market [1] Group 2 - China Galaxy Securities indicated that the computer industry may have seen bottom signals in Q1 2025, with revenue growth maintaining an upward trend and significant acceleration, suggesting a gradual recovery in market demand [2] - Shanxi Securities noted that AI has positively impacted Tencent's advertising and gaming businesses, with Tencent increasing investments in AI applications, while Alibaba's AI-related product revenue has achieved triple-digit growth for seven consecutive quarters [2] - The sustained capital expenditure of 380 billion yuan by domestic and international tech giants in AI validates the long-term demand for computing power despite macroeconomic disturbances [2]