Core Insights - The report from GF Securities highlights the increasing importance of AI memory as a foundational capability for context continuity, personalization, and historical information reuse, which is expected to accelerate the deployment of applications like AI Agents [1] Group 1: AI Agent and CPU Demand - The demand for CPUs is expected to increase due to AI Agents driven by three main factors: (1) Amplified application call volume as users can simultaneously engage multiple Agents, leading to higher system requests and increased CPU load [1] (2) Orchestration bottlenecks as the decision-making process involves multiple external tools, which increases CPU utilization [1] (3) Sandbox isolation raises overhead costs, necessitating additional CPU, memory, and storage configurations [1] Group 2: CPU Ratio Trends - According to a report by Semianalysis, the CPU ratio per GPU megawatt (MW) is currently below 10% but is projected to rise to 15% by Q2 2026, indicating a significant increase in CPU demand alongside GPU deployment [2] Group 3: Memory and Interface Chip Opportunities - The demand for memory modules is expected to grow as AI servers shift towards full insertion (2DPC) configurations to meet higher capacity and bandwidth requirements, leading to a doubling of DIMM demand relative to CPU growth [3] - The transition from traditional RDIMM to MRDIMM configurations will require more complex interface chips, which will drive up the average selling price (ASP) of supporting chips, thus expanding the market for memory modules and interface chips [3]

[Table_Page] 行业专题研究|电子 2026 年 1 月 25 日 证券研究报告 [Table_Title] AI 的 Memory 时刻 4 | | | AI DRAM | | 需求旺盛，行业供需紧张有望延续 | | --- | --- | --- | --- | --- | | [Tabl 分析师： | 王亮 | 分析师： 耿正 | | 分析师： 焦鼎 | | e_Author] | SAC 执证号：S0260519060001 | | SAC 执证号：S0260520090002 | SAC 执证号：S0260522120003 | | | SFC CE.no: BFS478 | | | | | | 021-38003658 | | 021-38003660 | 021-38003658 | | | gfwangliang@gf.com.cn | | gengzheng@gf.com.cn | jiaoding@gf.com.cn | | 分析师： | 张大伟 | | | | | | SAC 执证号：S0260523050001 | | | | | | 021-38003657 | | | | ...

广发证券研报指出，AI的Memory时刻，AI记忆成为支撑上下文连续性、个性化与历史信息复用的底层 能力，持续扩展模型能力边界，有望促进AI Agent等应用加速落地。AI记忆的价值正从"费用项"转变 为"资产项"，相关上游基础设施价值量、重要性将不断提升。建议关注产业链核心受益标的。 ...

Investment Rating - The report assigns a "Buy" rating for the industry, indicating an expectation that stock prices will outperform the market by more than 10% over the next 12 months [26]. Core Insights - The evolution from Attention to Memory in AI highlights the increasing importance of memory systems, which are essential for context continuity and personalized memory in AI applications. This shift is crucial for enhancing the capabilities of AI models and agents [2][11]. - AI memory is expected to accelerate the deployment of AI agents, as it is not merely a single module but a sustainable, evolving system capability. The memory system is structured into three layers: form, function, and dynamics, which significantly expands the usability of large models [2][3]. - The value of AI memory is transitioning from a cost item to an asset item, allowing for the accumulation of reusable experiences that enhance user engagement and operational efficiency. This transformation is expected to improve retention rates and average revenue per user (ARPU) [2][3]. Summary by Sections From Attention to Memory - AI memory is becoming increasingly vital, with philosophical questions about identity and origin reflecting memory issues. The ability to store and retrieve information over time is essential for AI's evolution from single-instance reasoning to long-term operational agents [11][20]. Investment Recommendations - The report suggests focusing on core beneficiaries within the industry chain, as AI memory is set to expand the boundaries of model capabilities and promote the rapid deployment of AI agents. The increasing importance of upstream infrastructure related to AI memory is highlighted [20][21]. Key Company Valuations and Financial Analysis - The report includes a detailed valuation and financial analysis of key companies in the sector, with several companies rated as "Buy" based on their expected performance and reasonable value estimates [4].

Core Insights - The focus of AI competition is shifting from scaling to memory, with OpenAI leading the charge and Google following closely behind in continuous learning [1][5][19] - The latest global AI report indicates that Google is making significant strides in overtaking competitors [1] - Altman believes that memory capacity is crucial for AI's intelligence level, and the potential for AI memory is nearly limitless [10][26] Group 1: Memory as a Key Focus - Altman emphasizes that memory will be the core issue for AI companies in 2026, as it is essential for achieving general artificial intelligence (AGI) [19][22] - Current AI memory capabilities are still primitive and at an early stage, comparable to the GPT-2 era [11][12] - Enhancing AI memory is seen as a critical technological challenge that must be overcome to establish deeper relationships between AI and users [12][25] Group 2: Implications for AI Development - The ability of AI to remember details about users' lives is viewed as a key step towards achieving superintelligent AI (ASI) [9][10] - Experts, including notable figures like Yoshua Bengio and Eric Schmidt, are advocating for a rigorous evaluation of memory capabilities in AI [13] - The introduction of memory mechanisms in ChatGPT is seen as a significant advancement, with other models expected to follow suit [22][24] Group 3: Future Predictions - Andrew Pignanelli predicts that once ChatGPT achieves breakthroughs in memory, all model providers will enhance their applications with memory capabilities [24] - Despite advancements, the industry is still far from perfecting long-term memory systems, and current solutions are considered temporary fixes [25] - The first true AGI will require a powerful memory system alongside a strong processing unit, indicating that memory is essential for a "digital self" [26][27]

Core Viewpoint - Recently, AI memory company Suisheng Technology announced the completion of a multi-million dollar angel round financing, led by Sequoia China Seed Fund, with participation from Anker Innovation co-founder Gao Tao, and Gao Hu Capital serving as the exclusive financial advisor. The company's first product is expected to be launched in the European and American markets in 2026 [1]. Group 1: Company Overview - Suisheng Technology was founded in August 2025 by Wang Shiyuan, the former global CMO and China president of Anker Innovation. Wang has been instrumental in Anker's global brand and marketing efforts since joining in 2015 [3]. - The core team consists of members from leading tech companies such as Anker Innovation, Huawei, and Tencent, possessing extensive experience in smart hardware, AI algorithms, and AI Agent product development and commercialization [3]. Group 2: Strategic Focus - Suisheng Technology has chosen "AI Memory" as its strategic mainstay, differentiating itself from the increasingly homogenized AI hardware for meeting recording and translation in China. The company believes that "memory" will play a crucial role in the future AI ecosystem [4]. - The company aims to create a complete closed loop of memory management and action execution through the synergy of software and hardware, enhancing efficiency in both work and life [4]. Group 3: Market Opportunity - The AI memory sector in Europe and America is still in its early stages, with a scarcity of players proficient in both software and hardware. This presents a structural opportunity for global teams [5]. - Compared to the intense competition in China, the AI memory track in Europe and America is still developing. Some overseas teams excel in software experience but lack hardware delivery capabilities, while Chinese teams have advantages in supply chain and product iteration speed but are relatively weak in brand building and channel experience in the Western markets [6]. - Suisheng Technology sees this "capability mismatch" as a strategic window, with a significant innovation space available due to the lack of balanced capabilities in teams. The company plans to establish a growth cycle of "product sales - data accumulation - product iteration" to build user awareness in the European and American markets [6].

硬氪独家获悉，前安克创新全球CMO、中国区总裁王时远离职后创业，成立"穗升科技"；公司聚焦 AI录音硬件赛道，依托软硬件结合方案实现记忆管理与行动的闭环。 "我们认为，'记忆'会在未来AI生态中发挥关键作用。基于AI生态构建结构化、可调用的用户个人记 忆，能作为上下文（context）为AI提供支撑，让Agent更精准地服务用户。"王时远告诉硬氪，硬件 只是收集、储存用户记忆的入口和载体，"其中，声音仅作为短期的数据输入源之一；从中长期来 看，多模态信息的输入方式将会日趋成熟并逐步普及。" 与国内相比，AI记忆赛道在欧美呈现出不同的竞争格局。王时远向硬氪分析指出，目前该领域在海外 仍处早期发展阶段，真正具备软硬件协同能力的成熟玩家十分稀缺。尽管部分海外企业擅长软件体验 与垂直场景，国内团队则在市场响应速度上更具优势，但能同时在硬件迭代与软件体验上保持领先的 参与者并不多见。 这正是王时远和团队眼中的战略机会。无论是技术架构还是应用场景均未固化，这也为硬件创新、软 件迭代及生态构建留下充分的想象和探索空间。 王时远表示，团队熟悉欧美市场的营销与销售模式，计划以此为基础，构建起"产品销售—数据积累 —产品迭代"的 ...

Core Viewpoint - Suisheng Technology, founded by former Anker Innovation CMO Wang Shiyuan, focuses on the AI recording hardware sector, aiming to enhance memory management and actions through integrated software and hardware solutions. The first product is expected to launch in the European and American markets in 2026 [4][6]. Group 1: Company Overview - Suisheng Technology has completed a multi-million dollar angel round of financing, led by Sequoia China Seed Fund, with participation from Anker co-founder Gao Tao. Gao Hu Capital served as the exclusive financial advisor [6]. - The company was established in August 2025, with a founding team that has extensive experience in AI research and development, having previously worked at leading tech companies such as Anker, Huawei, and Tencent [6]. Group 2: Product Strategy - The company aims to differentiate itself in the AI memory sector, which is currently facing homogenization in the domestic market. Existing products are primarily limited to basic functions like meeting minutes and translation [6]. - Wang Shiyuan believes that "memory" will play a crucial role in the future AI ecosystem, providing structured and callable personal memory to support AI context, allowing agents to serve users more accurately [7]. Group 3: Market Positioning - Compared to the domestic market, the AI memory sector in Europe and America is still in its early development stage, with few mature players capable of integrating software and hardware [7]. - The strategic opportunity lies in the lack of established technology frameworks and application scenarios, allowing for innovation in hardware and software iterations [7]. Group 4: Marketing and Sales Strategy - The team plans to leverage their familiarity with European and American marketing and sales models to create a virtuous cycle of "product sales - data accumulation - product iteration" [8]. - Suisheng Technology will utilize its long-term experience in cross-border marketing and channel operations to accelerate its commercialization process by integrating local channel resources [9].

Core Insights - The article discusses the establishment of "SuiSheng Technology" by Wang Shiyuan, former CMO of Anker Innovations, focusing on AI recording hardware and memory management solutions [1][2] - SuiSheng Technology has completed a multi-million dollar angel round financing led by Sequoia China Seed Fund, with participation from Anker's co-founder [1] - The first product is expected to launch in the European and American markets in 2026 [1] Company Overview - SuiSheng Technology was founded in August 2025, with a team that includes members from Anker Innovations, Huawei, and Tencent, possessing extensive experience in AI and smart hardware [1][2] - The company aims to differentiate itself in the AI memory sector, which is currently facing homogenization in the domestic market [1][2] Product Strategy - The company focuses on creating a seamless hardware device combined with AI memory functions to enhance user efficiency in work and life [2] - Wang Shiyuan emphasizes that "memory" will play a crucial role in the future AI ecosystem, providing context for AI to better serve users [2] Market Positioning - The AI memory sector in Europe and America is still in its early stages, with few mature players capable of integrating hardware and software [2] - SuiSheng Technology sees this as a strategic opportunity, as the technology architecture and application scenarios are not yet fixed, allowing for innovation and exploration [2] Commercialization Strategy - The company plans to leverage its team's expertise in cross-border marketing and channel operations to accelerate its commercialization process [3]

Core Insights - The article emphasizes that memory is becoming a critical breakthrough in the evolution of AI, transitioning from "instant answer tools" to "personalized super assistants" [1][4] - A new machine learning paradigm called "Nested Learning" has been proposed, allowing large language models to learn new skills without forgetting old ones, marking significant progress towards AI that mimics human memory [3][4] Group 1: Shifts in AI Landscape - The focus of large models is shifting from size and speed to memory capabilities and understanding user needs, indicating a new competitive landscape in AI [4][5] - Current large models struggle with long-term memory due to inherent limitations in their architecture, leading to issues like forgetting critical user information during interactions [6][7] Group 2: Memory Mechanisms - Existing models typically have context windows of 8k-32k tokens, which can lead to early information being "pushed out" during long conversations, causing loss of context [6] - The lack of a shared memory mechanism among multiple agents results in "memory islands," where users must repeatedly provide information, diminishing the user experience [7] Group 3: Innovations in Memory - Companies like Google, OpenAI, and Anthropic are focusing on enhancing memory capabilities in AI models, responding to industry demands for long-term, stable, and evolving memory systems [7][10] - Red Bear AI has developed "Memory Bear," a product that addresses the memory limitations of traditional models by implementing a human-like memory architecture [10][11] Group 4: Memory Bear's Architecture - "Memory Bear" utilizes a hierarchical, dynamic memory structure inspired by the human brain's hippocampus and cortex, allowing for efficient memory management [11][13] - The system distinguishes between explicit memory (easily codified information) and implicit memory (subjective understanding), enhancing its ability to recall and utilize user-specific data [15][16] Group 5: Practical Applications and Impact - "Memory Bear" has shown significant improvements in various applications, such as AI customer service, where it creates dynamic memory maps for users, enhancing interaction quality and reducing the need for repetitive information sharing [20][21] - In marketing, "Memory Bear" tracks user behavior to create personalized marketing strategies, moving beyond traditional recommendation systems [22] - The technology has also improved knowledge acquisition efficiency in organizations and personalized education experiences, demonstrating its versatility across sectors [23][24] Group 6: Industry Consensus and Future Directions - The consensus in the industry is that memory capabilities are essential for advancing AI technology and applications, with increasing investments and explorations into human-like memory systems [24]