每经AI快讯，有投资者在投资者互动平台提问：贵公司的电子后视镜已经有车型使用吗？ （记者 张明双） 协创数据（300857.SZ）1月9日在投资者互动平台表示，公司电子后视镜相关产品目前通过终端客户的 渠道已实现市场化应用，覆盖日常行车辅助等需求。 ...

Core Viewpoint - The article discusses the severe memory shortage affecting technology companies, particularly in the context of the CES trade show, highlighting the impact on product pricing and availability due to the shift in focus from DRAM to high-bandwidth memory for AI data centers [1][2][3]. Group 1: Memory Shortage Impact - The shortage of DRAM, essential for laptops and smartphones, is exacerbated by major manufacturers prioritizing high-bandwidth memory for AI applications, leading to significant price increases [2][3]. - Dell's COO Jeff Clarke stated that the current shortage is the worst he has seen, with demand far exceeding supply, and prices in the spot market have surged fivefold since September [3]. - Analysts predict that DRAM prices could rise by approximately 40% in Q4 2025, with potential increases of up to 60% in Q1 2026, indicating a prolonged shortage [3]. Group 2: Company Responses and Innovations - Phison Electronics has introduced aiDAPTIV, a product designed to enhance PC GPU memory bandwidth, allowing manufacturers to reduce DRAM capacity without sacrificing performance [5]. - Ventiva has developed a fanless cooling solution that frees up space in laptops for additional memory, addressing the physical limitations imposed by traditional cooling methods [6]. - Both companies aim to reduce reliance on cloud computing by enhancing AI capabilities on personal computers, which could shift demand dynamics in the memory market [7][8]. Group 3: Market Dynamics and Future Outlook - The article emphasizes the need for a coordinated effort among PC manufacturers, memory producers, and technology companies to address the memory crisis and avoid significant price hikes that could impact sales [8]. - There is a growing sentiment that if personal computers can meet the AI processing needs, the focus on data centers may need to shift, potentially leading to a reevaluation of memory production strategies [9].

Core Insights - The discussion at CES 2026 highlights a shift in the tech industry from viewing AI as a standalone technology to exploring its practical applications in various scenarios [2] - Lenovo's strategy in the AI era is outlined, emphasizing the transformation of the global AI industry from public services to personalized and enterprise-level applications [2] AI Computing Power and Model-Driven Terminal Ecosystem Reconstruction - The global AI computing market is projected to reach $115.2 billion by 2026, growing at a rate of 42.8%, significantly outpacing traditional computing markets [3] - The rapid development of model miniaturization technology is challenging the notion that performance is solely determined by parameter scale, enabling smaller models to achieve comparable capabilities to larger ones [3][4] Integration of AI and Terminals - The integration of AI with terminals is seen as a necessary solution to the core issues of public AI, such as insufficient personalization and the inability to process private data [4] - Future terminal ecosystems are expected to evolve into three main forms: upgraded existing terminals, new perception-focused terminals like AI glasses, and edge computing terminals for secure, private calculations [5][19] Rise of Personal AI - The emergence of personal AI signifies a paradigm shift from platform-centric to user-centric AI services [6] - Personal AI is characterized by four key features: synchronized perception, trusted computation, exclusive service connections, and continuous evolution [8][22] Challenges in Personal AI Implementation - Personal AI faces four major technical challenges: building heterogeneous computing platforms, managing multiple models and agent scheduling, long-term memory management, and core experience innovation [9][24] - Lenovo's "teammate" personal AI aims to enhance interaction logic through situational awareness, proactive service, and direct execution of tasks [9][37] Enterprise-Level AI Implementation Challenges - Successful enterprise-level AI deployment requires upgrading digital infrastructure, restructuring business processes, and cultivating AI talent [10][45] - Talent development is identified as the most critical challenge, with a focus on training middle management to lead AI integration efforts [10][46] Future Competitive Landscape in AI - The core competitiveness in the AI era will hinge on "integration and implementation," with a shift in focus from technical parameters to scenario value [11] - Companies that effectively grasp trends and deepen implementation will emerge as winners in the intelligent era [11]

Group 1 - The company plans to participate as a limited partner in the establishment of the Shanghai Tongge Phase II Venture Capital Partnership (Limited Partnership), referred to as "Tongge Phase II Fund" [1] - The total scale of the Tongge Phase II Fund is approximately 697 million yuan, with the company and the general partner, Shanghai Tongge Xingchen Management Consulting Partnership (Limited Partnership), contributing 99 million yuan and 1 million yuan respectively to complete the fund establishment [1] - After the fund is established, the company and Shanghai Tongge Xingchen will jointly contribute about 597 million yuan with other limited partners, with the company's total contribution not exceeding 230 million yuan, accounting for no more than 33% of the fund's total scale [1] Group 2 - The investment focus of the fund includes artificial intelligence, XR and spatial computing, new materials and advanced manufacturing, aerial mobility, and semiconductors [1] - The company's participation in the establishment of the Tongge Phase II Fund is beneficial for broadening its investment channels [1]

Core Insights - Lenovo held the 2026 Lenovo TechWorld conference, showcasing its strategic achievements in AI, emphasizing its role in the global AI industry chain [1] Group 1: Strategic Partnerships and Product Launches - Lenovo gathered CEOs from major chip companies like NVIDIA, AMD, Intel, and Qualcomm, highlighting its significant position in the AI industry [1] - The company launched a new AI inference server, ThinkSystem SR675i, in collaboration with AMD, optimized for large-scale data center inference scenarios [2] - Lenovo announced a partnership with NVIDIA to establish an AI cloud super factory, supported by NVIDIA's accelerated computing platform [2] Group 2: Consumer AI Developments - Lenovo introduced the overseas version of its personal AI super agent, Lenovo Qira, marking a shift from application-level to system-level AI strategy [3] - Qira integrates Lenovo's PC, tablet, and Motorola phone ecosystems, allowing seamless task switching across devices with core capabilities of responsiveness, execution, and situational awareness [3] - The company expects to enhance user engagement and create a differentiated software ecosystem in the competitive AI PC market with Qira [3] Group 3: Financial Projections - Lenovo maintains its FY26/27/28 Non-HKFRS net profit estimates at $1.66 billion, $1.85 billion, and $2.08 billion, with year-on-year growth rates of 15.5%, 11.4%, and 12.4% respectively [3] - Corresponding Non-HKFRS EPS projections are $0.12, $0.14, and $0.15 for FY26, FY27, and FY28 [3]

Group 1 - The State Intellectual Property Office of China has granted a patent to BOE Technology Group Co., Ltd. and Chengdu BOE Display Technology Co., Ltd. for "Array Substrate and Its Preparation Method, Display Panel," with the authorization announcement number CN116794894B, applied on June 2023 [1] - BOE Technology Group Co., Ltd., established in 1993 and located in Beijing, primarily engages in the manufacturing of computers, communications, and other electronic devices, with a registered capital of 37,413.88 million RMB [1] - The company has invested in 74 enterprises, participated in 299 bidding projects, holds 775 trademark records, and possesses 5,000 patent records, along with 47 administrative licenses [1] Group 2 - Chengdu BOE Display Technology Co., Ltd., established in 2015 and located in Chengdu, focuses on software and information technology services, with a registered capital of 2,155 million RMB [1] - The company has participated in 609 bidding projects, holds 558 patent records, and possesses 841 administrative licenses [1]

每经AI快讯，冠捷科技1月9日晚间发布公告称，公司第十一届第七次董事会临时会议于2026年1月9日 以通讯方式召开。会议审议了《关于聘任公司高级管理人员的议案》等文件。 每经头条（nbdtoutiao）——独家对话特斯拉FSD跨美第一人：4400公里"零接管"，手没碰过方向盘！作 为激光雷达销售员，他为何站队马斯克的"纯视觉"？ （记者 曾健辉） ...

Group 1 - The company announced that its wholly-owned subsidiary, Automotive Electronics, agreed to publicly transfer 20% equity of Shanghai Rijing through the Shanghai United Assets and Equity Exchange [1] - The initial transfer price was set at 103.63154 million yuan, and the buyer is Japan Precision Co., Ltd. [1] - The contract was signed on December 2, 2025, and the transaction certificate was issued by the exchange on December 10, 2025 [1] Group 2 - The equity transfer registration was completed on January 8, 2026, and Automotive Electronics no longer holds shares in Shanghai Rijing [1] - After Shanghai Rijing completes the FDI registration, Automotive Electronics will proceed with foreign settlement and receive payment [1]

Core Viewpoint - The company, Xinyin Electronics, is actively engaged in the production of high-speed connectors, specifically USB4 connectors, and is in the process of developing several other products [2] Product Development - The company is currently producing USB4 connectors [2] - The company has DDR5 SO-DIMM memory card slot connectors in small batch production [2] - DDR memory card slot connectors are also in small batch production [2] - Mini FAKRA automotive interfaces are in the certification stage [2]

天眼查资料显示，广东致能半导体有限公司，成立于2018年，位于深圳市，是一家以从事科技推广和应 用服务业为主的企业。企业注册资本2123.0355万人民币。通过天眼查大数据分析，广东致能半导体有 限公司共对外投资了4家企业，参与招投标项目7次，财产线索方面有商标信息9条，专利信息120条，此 外企业还拥有行政许可21个。 国家知识产权局信息显示，广东致能半导体有限公司;徐州致能半导体有限公司取得一项名为"一种半导 体级联器件及其封装方法"的专利，授权公告号CN120709163B，申请日期为2025年8月。 声明：市场有风险，投资需谨慎。本文为AI基于第三方数据生成，仅供参考，不构成个人投资建议。 来源：市场资讯 徐州致能半导体有限公司，成立于2020年，位于徐州市，是一家以从事计算机、通信和其他电子设备制 造业为主的企业。企业注册资本32000万人民币。通过天眼查大数据分析，徐州致能半导体有限公司参 与招投标项目11次，专利信息39条，此外企业还拥有行政许可5个。 ...