支撑这些机器人精准完成动作、适配复杂环境的核心，是一颗具备自主决策、路径规划与数据处理能力 的"机器人大脑"。这家专注于四足机器人研发、生产与场景应用的高新技术企业，正以技术创新推动行 业变革。 1月13日，山东星界创新机器人有限公司的智能实训场内，多款功能细分的机器人正井然有序地在各自 专属的训练区域开展训练。 在导航技术上，公司近期实现关键突破。位于山东省滨州市沾化区的研发团队成功攻克SLAM激光雷达 导航技术在复杂环境中的应用难题。"我们通过提取激光雷达运行时创造的点云数据，构建高精度三维 地图，为机器人植入'智能大脑'，使其在昏暗、复杂甚至跨楼层场景中也能稳定导航。"公司总经理李 德政说。 这项技术已应用于最新款服务机器人"小星"，使其在接收到语音指令后，能自动规划最优路径、灵活避 障，流畅完成巡检、导引等任务。 目前，公司机器人已成功应用于山东达民化工、中国移动（600941）呼和浩特数据中心、集地广场、青 岛地铁等多个项目，覆盖化工巡检、安防监控、机房运维、公共交通管理等场景。 客户反馈显示，机器人可替代80%以上重复巡检工作，巡检覆盖率从60%—70%提升至100%，实现24小 时不间断作业，隐 ...

Core Insights - The first "National Green Computing Power Facility" promotion event was held in Shanghai, announcing the list of facilities for 2025, with over 12 million standard racks expected in use, positioning China as the second-largest in global computing power scale [1][2] - The demand for computing power is growing exponentially, driven by applications such as AI model training and industrial internet empowerment, leading to a significant increase in energy consumption [1] - The International Energy Agency reported that global data center electricity consumption has been growing at an annual rate of approximately 12% since 2017, more than four times the global total electricity consumption growth rate [1] Group 1: National Green Computing Power Facilities - A total of 60 data centers were included in the 2025 National Green Computing Power Facility list, covering various sectors such as industrial, information communication, energy, internet, finance, and public institutions [2] - The Ministry of Industry and Information Technology has cultivated over 300 national green computing power facilities, achieving positive results in energy efficiency and collaborative development [2] Group 2: Technological Advancements and Innovations - Shanghai is supporting the implementation of key green low-carbon technologies in computing power centers, with advanced technologies like liquid cooling and AI intelligent operation being scaled up [3] - The city has established a public computing power service platform and a diverse supply structure involving major telecom operators and market participants [3] Group 3: Industry Performance and Market Outlook - Companies in the green computing power supply chain are experiencing robust production and sales, with some firms reporting high capacity utilization and a strong order backlog [4] - The company Shunling Environment has developed a modular liquid cooling system to meet the growing demand for high-density cooling in data centers, serving major clients like Huawei and ByteDance [4] - Kstar, a leading UPS manufacturer, is benefiting from the global data center infrastructure boom, with increasing orders and exports to over 180 countries [5]

Group 1 - The first "National Green Computing Power Facility" promotion and exchange event was held in Shanghai, announcing the list of national green computing power facilities for 2025, with over 12 million standard racks expected to be in use in China, making it the second-largest globally in terms of computing power scale [1] - The demand for computing power driven by AI model training and industrial internet applications is growing exponentially, leading to a significant increase in energy demand, with global data center electricity consumption growing at an average annual rate of about 12% since 2017, more than four times the global total electricity consumption growth rate [1] - The event featured 60 data centers selected for the national green computing power facility list, including facilities from various sectors such as industrial, information communication, energy, internet, finance, and public institutions [1] Group 2 - The Ministry of Industry and Information Technology emphasizes the importance of green and low-carbon development of computing power facilities, having cultivated over 300 national green computing power facilities across various sectors, achieving positive results in energy efficiency and collaborative computing power [2] - Shanghai currently has over 160 data centers with approximately 250,000 operational racks, accounting for nearly 3% of the city's total electricity consumption, and supports the implementation of green low-carbon technologies in various centers [2] - Shanghai's intelligent computing power deployment has reached 120,000 P (FP16), forming a diverse supply structure with a public computing power service platform and collaboration among major operators and market entities [2] Group 3 - The core goal of "green computing power" is to reduce the energy consumption of computing infrastructure, with PUE (Power Usage Effectiveness) being a key metric for measuring this goal, where a lower PUE indicates higher operational efficiency and reduced carbon emissions [3] - Several companies in the green computing power supply chain showcased new products and systems at the event, with some expressing optimism about industry growth, highlighting high capacity utilization and sufficient orders [3] - Companies like Shenling Environment are expanding their production capabilities for liquid cooling systems, which are essential for high-density heat dissipation in data centers, and have established partnerships with major clients such as Huawei and ByteDance [4] Group 4 - UPS (Uninterruptible Power Supply) systems are essential for data center construction, significantly impacting computing power output, with companies like Kstar showcasing their latest integrated system models at the event [4] - Kstar's data center business is benefiting from a global infrastructure boom, with increasing orders and shipments contributing to rapid overall performance growth, and their products meeting national green energy efficiency standards [4] - The company serves major internet, finance, and communication clients while expanding its overseas business, exporting products to over 180 countries and regions [4]

Group 1 - The core viewpoint emphasizes the importance of future industries in driving high-quality economic development and optimizing industrial structure, with various regions actively promoting innovation and application of future industries [1][4] - Multiple local governments are focusing on supporting enterprise growth and increasing financial support as key measures to foster innovation in future industries, recognizing that enterprises are the core carriers for technology exploration and commercialization [2][3] - The future industries are characterized by high investment, high risk, long cycles, and multiple technology iterations, necessitating a balanced mechanism between encouraging bold investments and ensuring accurate investments to avoid resource misallocation [3] Group 2 - The Ministry of Industry and Information Technology has outlined nine key future industries, including metaverse, brain-computer interfaces, quantum information, humanoid robots, generative artificial intelligence, biomanufacturing, future displays, future networks, and new energy storage, indicating a structured approach to future industry development [4] - Significant progress has been made in future industries, such as continuous upgrades in brain-computer interface technology and the expansion of humanoid robot industry scale, while challenges remain in the collaborative development of innovation chains, industrial chains, and ecological chains [4][5] - The government is committed to implementing strategic technology projects and establishing future industry pilot zones, aiming to create a competitive environment for future industries across the country [5][6]

AI基金大摩优享六个月持有期混合A（012368）披露2025年四季报，第四季度基金利润942.15万元，加权平均基金份额本期利润0.0276元。报告期内，基金 净值增长率为2.98%，截至四季度末，基金规模为2.97亿元。 该基金属于偏股混合型基金。截至1月21日，单位净值为0.91元。基金经理是余斌和缪东航。 基金管理人在四季报中表示，本基金长期看好低波红利板块，目前中国的经济发展日趋成熟，经济增速将出现一定程度的放缓，因此未来中国企业的利润增 长速度也将跟随经济放缓。企业产能扩张的需求减少，企业将减少资本开支，自由现金流将大幅改善，企业将提升分红以回报投资者。 十年期国债收益率目前仍然处于低位，低波红利指数的股息率平均而言显著高于十年期国债收益率，这将有利于红利板块的估值提升。即便红利板块的估值 由于市场风格原因未能提升，我们认为持有高红利股票获得稳定分红，也能让基金获得显著高于国债收益率的回报。 截至1月21日，大摩优享六个月持有期混合A近三个月复权单位净值增长率为0.93%，位于同类可比基金557/621；近半年复权单位净值增长率为1.05%，位于 同类可比基金598/621；近一年复权单位净值增长 ...

电信运营商也要跨界卖保险？ 近日，"移动医保保""移动家保保"等保险产品宣传图在社交平台流传，引发市场广泛关注。 1月22日，《国际金融报》记者致电中国移动官方客服、走访上海多家营业厅了解到，中国移动已在上海、山东等地推出"移动医保保""移动家保保"等保险 产品。 王莹/摄 在业内人士看来，运营商将业务触角延伸至保险领域，主要是为了应对传统通信市场饱和，开辟第二增长曲线，实现业务多元化。一方面，对公司而言，这 能利用庞大的用户基础和线下渠道变现，推动向"生态运营商"转型；另一方面，对消费者来说，运营商渠道提供了更便捷、低门槛的保险服务，尤其惠及下 沉市场。 两款保险产品上架 在中国移动上海潍坊路营业厅内，记者注意到，"移动医保保""移动家保保"两款产品的宣传展板已摆放在大厅醒目位置。工作人员介绍称，产品上线以来， 已陆续有市民到店咨询详情。 "运营商跨界卖保险的核心逻辑，是依托自身核心优势挖掘存量价值、拓展服务生态。"苏商银行特约研究员付一夫在接受记者采访时指出，从行业趋势看， 通信主业增长放缓，运营商亟需寻找新的利润增长点，而保险作为高频刚需的金融服务，与通信场景天然契合，成为生态延伸的重要方向。此举可盘活 ...

智通财经1月22日讯（编辑 冯轶）据Wind数据显示，南向资金今日成交约931.59亿港元，较前一日再度缩量约158亿；约占今日恒指成交总额的39.68%，占 比跌破四成。 趋势上，南向资金已连续5个交易日保持流入，期间累计买入252.17亿港元。 个股方面，交易所数据显示，今日南向资金 港股今日延续震荡走势，三大指数涨跌不一。南向资金净流入约52.39亿港元。其中，沪港股通净流入约27.75亿港元，深港股通净流入约24.64亿港元。 中国移动今日跌0.25%，资金前5日减持4235万股，短线延续流出趋势。 腾讯控股今日跌0.83%，资金前5日加仓470万股，短线流入趋缓。 大幅净买入：阿里巴巴-W(09988.HK)14.13亿港元；中国海洋石油(00883.HK)5.92亿港元；快手-W(01024.HK)4.79亿港元；中国人寿 (02628.HK)4.45亿港元；泡泡玛特(09992.HK)3.74亿港元；小米集团-W(01810.HK)4.47亿港元。 大幅净流出：中国移动(00941.HK)8.74亿港元；腾讯控股(0700.HK)6.69亿港元；地平线机器人-W(09660.HK)1.21亿港元 ...

Investment Rating - The report does not provide a specific investment rating for the industry. Core Insights - The document outlines the importance of training data cleansing as a critical step in ensuring the quality of AI model training, emphasizing the need to eliminate content that violates core socialist values, discriminatory content, and any illegal or infringing information [8][11]. - It presents safety principles for training data cleansing, including safety control, diversity of data sources, transparency, and continuous iteration [17]. - The report identifies various risks associated with training data, such as data quality risks, risks of violating socialist core values, discriminatory content risks, commercial legal risks, and risks of infringing on others' legal rights [18][20][21][23][24]. Summary by Sections 1. Scope - The document provides guidelines applicable to all entities involved in training data cleansing, including AI model developers and data suppliers [11]. 2. Terminology Definitions - Key terms are defined, including "training data," "training data cleansing activities," and "security of training data cleansing process," which are essential for understanding the context of the guidelines [12][13]. 3. Safety Principles for Training Data Cleansing - The principles include: - Safety and control: Ensuring data legality and content safety [17]. - Diversity: Considering various data sources and scenarios [17]. - Transparency: Documenting rules and decisions for auditability [17]. - Continuous iteration: Adapting strategies based on feedback and changes [17]. 4. Risk Identification Dimensions - The report categorizes risks into several dimensions, including: - Data quality risks: Issues like incompleteness, inaccuracies, and outdated data [18]. - Violations of socialist core values: Content that could negatively impact society [20]. - Discriminatory content risks: Systematic biases against specific groups [21]. - Commercial legal risks: Risks associated with intellectual property violations [23]. - Infringement of legal rights: Risks related to personal information and other rights [24]. 5. Cleansing Methods - Various methods for data cleansing are outlined, including: - Data quality processing: Addressing completeness, accuracy, timeliness, usability, and redundancy [31]. - Source control: Ensuring data comes from reliable and verified sources [33]. - Content safety review: Implementing filtering rules and models for harmful content [35]. - Personal information protection: Techniques for anonymization and de-identification [36]. - Bias mitigation: Strategies to balance representation in training data [38]. 6. Implementation Process - The report details a structured process for implementing data cleansing, including: - Data collection and source review: Defining the scope and purpose of data collection [41][42]. - Initial data quality processing: Checking for completeness, accuracy, and relevance [43][44]. - Risk identification and cleansing strategy formulation: Using statistical and technical methods for risk assessment [48][49]. - Data cleansing execution: Following established methods to cleanse data [51]. - Secondary risk assessment: Evaluating the effectiveness of cleansing methods [52][53][54][55][56]. - Continuous monitoring and iterative optimization: Establishing feedback mechanisms and regular reviews [62][64].

新华社香港1月22日电 香港恒生指数22日涨44.9点，涨幅0.17%，收报26629.96点。全日主板成交2348.6 亿港元。 石油石化股方面，中国石油化工股份涨3.41%，收报5.15港元；中国石油股份涨2.38%，收报8.6港元； 中国海洋石油涨2.52%，收报22.74港元。 香港本地股方面，长实集团涨1.76%，收报44.04港元；新鸿基地产涨3.63%，收报114.2港元；恒基地产 跌0.2%，收报30.66港元。 中资金融股方面，中国银行涨0.22%，收报4.47港元；建设银行跌0.78%，收报7.61港元；工商银行跌 0.64%，收报6.17港元；中国平安跌2.34%，收报66.9港元；中国人寿跌3.82%，收报31.72港元。 【纠错】 【责任编辑:赵阳】 国企指数跌8.65点，收报9114.3点，跌幅0.09%。恒生科技指数涨16.14点，收报5762.44点，涨幅 0.28%。 蓝筹股方面，腾讯控股跌0.83%，收报597.5港元；香港交易所跌0.23%，收报426.4港元；中国移动跌 0.25%，收报79港元；汇丰控股涨1.17%，收报129.7港元。 ...

来源：环球网 近日，广西多地遭遇强寒潮侵袭，雨雪冰冻天气给通信网络带来严峻挑战。中国移动广西公司闻令而 动，以"人民邮电为人民"的使命担当，迅速启动应急预案，统筹调度、科技赋能、人力攻坚，在冰雪一 线筑起一道坚不可摧的通信屏障，全力保障人民群众生产生活秩序，于极寒中传递央企的温暖力量。 人员徒步进入故障点排障。侯明璇摄 踏雪巡 检，负重抢修，保障山区网络"不掉线" 林峰摄 挺进"云端"，专业护航：极寒环境下的网络韧劲 面对贺州姑婆山、大容山国家森林公园等海拔较高区域的雨雪冰冻，中国移动广西公司组织技术骨干迎 雪巡检、冒寒检测。保障人员运用专业仪器，细致排查光缆线路与基站运行状态，在"云端"之上筑牢网 络稳定运行的根基，为山区居民、游客及冬季旅游经济提供坚实可靠的通信支撑。 移动工作 人员全力筑牢寒潮通信"生命线"。 邓文涛摄 迎寒而上，使命必达：冰雪深山中的通信守护 在融水苗族自治县杆洞乡、环江毛南族自治县等高寒山区，道路凝冰、车辆难行。中国移动广西公司抢 修队伍肩扛油机、手拎设备，踏冰雪、攀险径，徒步挺进基站故障点。他们在刺骨寒风中调试设备、抢 通线路，以"信号满格"守护着偏远地区少数民族群众的通信生命线， ...