智通财经APP获悉，华鑫证券发布研报称，全球大模型迭代仍在提速而未现放缓迹象。AI模型调用量预 计明年仍会持续高增，且增长并非线性连续，一旦手机助手、多模态等应用出现爆款，可能会带动AI 模型调用量和算力需求出现跃迁式增长。建议重点关注海内外AI产业发展趋势，关注AI算力、国产AI 硬件、AI应用等方面的投资机会。 火山引擎还发布了视频生成模型Seedance 1.5 pro，支持音画同步输出、多人多语言对白配音，具备更强 大的影视级叙事张力，满足影视、漫剧、电商、广告的高阶创作需求。 为便利企业的AI应用和Agent部署，火山引擎全面升级企业级AIAgent平台AgentKit，发布企业版AI原生 编程工具TRAE，并推出HiAgent"1+N+X"智能体工作站。在模型定价上，火山引擎进一步创新模型定 价模式，推出"AI节省计划"，是业内首个大模型节省计划，覆盖火山引擎上所有的按量后付大模型产 品，包括豆包大模型家族、各类三方开源模型。 字节围绕AI大模型打造战略生态，上修未来收入增长目标 字节火山引擎当前已经打造了从模型到MaaS、Agent开发、Agent运营的一整套AI云原生架构。目前豆 包大模型日均 ...

Group 1: Major Corporate Developments - WeChat's installation package has increased several hundred times over the past decade due to the addition of more features, but the company assures that it will not continue to grow indefinitely and is working on optimizations [2] - Meta announced the acquisition of Manus for a deal worth up to several billion dollars, with Manus continuing operations in Singapore and its CEO becoming Meta's VP [3] - JD.com and Yushutech launched their first offline store in Beijing, marking a significant step in their partnership and the company's investment in the smart robotics sector [3] Group 2: Industry News and Trends - Tencent's AI Lab deputy director, Yu Dong, has left the company for personal development reasons, marking a notable change in the AI division [5] - Fire Mountain Engine has become the exclusive AI cloud partner for the 2026 Spring Festival Gala, indicating a growing trend of AI integration in major events [7] - Waymo has been spotted testing its autonomous taxis in London, although the vehicle was not operating in autonomous mode during the observation [8] Group 3: Market Movements and Financial Updates - Zhihui plans to globally issue 37,419,500 H-shares at a price of HKD 116.20 per share, with trading expected to begin on January 8, 2026 [13] - The price of certain storage cards will be increased starting January 5, 2026, due to rising costs from global supply chain issues [12]

Core Viewpoint - The implementation of the "Artificial Intelligence Large Model" national standards marks a new phase for China's large model industry, emphasizing scientific authority and unified norms [1] Group 1: National Standards Implementation - The series of standards is the first to focus on general large models, filling a gap in the technical evaluation system [1] - The standards clarify performance, safety, and service capability requirements, with the evaluation capabilities recognized by the China National Accreditation Service for Conformity Assessment [1] Group 2: Evaluation Benchmark - The "Qiusuo"-LMBench evaluation benchmark integrates methods, datasets, and an automated platform, becoming an industry-recognized "capability metric" [1] - The evaluation system has established a whitelist of mainstream large models, providing a basis for selection by state-owned enterprises and supporting companies like Sinopec and Southern Power Grid in building industry-specific large models [1] Group 3: Assessment Achievements - Over a thousand assessment tasks have been completed, utilizing large models more than 950,000 times, accurately identifying common issues such as hallucination control and content safety [1] - The efforts have aided nearly 30 manufacturers in technology iteration, promoting a closed loop of "R&D—Assessment—Application—Upgrade" [1] Group 4: Internationalization of Standards - The internationalization of standards is underway, with English, Lao, and Khmer versions being prepared for ASEAN, filling a regional gap [1] - The Chinese open-source model architecture has been adopted by projects in countries like Singapore, showcasing the global influence under the guidance of these standards [1]

Group 1 - The article discusses the announcement by Spiber, a biotech unicorn in Japan, regarding the involvement of Masaya Kawanami, the daughter of SoftBank's founder Masayoshi Son, in the company's operations starting in the first half of 2026. This marks her first public acknowledgment of her family background in a business context, raising speculation about potential succession plans for SoftBank [2] - Realme, a smartphone brand under OPPO, has responded to rumors of large-scale layoffs, stating that business operations are normal and that the rumored layoffs are part of regular personnel changes at year-end. The company is set to launch new products in January [4] - Bawang Tea's stock price fell nearly 15% due to concerns over caffeine content in their products, prompting the company to issue a statement clarifying that their tea's caffeine levels are comparable to that of a latte and significantly lower than that of an Americano [4] - Chow Tai Fook faced backlash over a gold pendant design, which was criticized on social media. The company responded that the product is selling well online and that customer feedback will be considered for future designs [6] - Dong'an Group confirmed the suspension of three Porsche dealerships due to operational difficulties, with plans to address employee salary issues within 60 days. The company denied rumors of investor flight [9] - Midea's Building Technology division announced a 5% price increase across all product lines due to rising raw material costs, effective from December 15, 2025 [11][12] - The first national standards for AI large models have been implemented, focusing on technical evaluation and aiding over 30 companies in technology iteration [13] Group 2 - Lenovo plans to unveil an AI super-intelligent agent at CES, which is expected to compete with existing AI assistants by offering more comprehensive features and cross-device connectivity [17] - Douyin has launched a long-form content feature allowing users to publish articles up to 8,000 words, enhancing its content creation capabilities [17] - The UK used car market saw a 2.8% growth in Q3 2025, with over 2 million transactions, marking the strongest performance since 2021 [22] - Apple's smartphone sales in China doubled in November, with a 128.4% year-on-year increase, while the overall smartphone market grew less than 2% [22]

Core Viewpoint - The "Ningbo Data Application Promotion Regulation" aims to cultivate a data factor market, promote the legal and efficient application of data, and drive the development of the digital economy in Ningbo [1][2] Group 1: Regulation Features - The regulation emphasizes three main features: strengthening data application and liquidity, deepening industrial cultivation with a focus on uniqueness, and improving institutional guarantees with an emphasis on innovation [1] - The regulation's goal is to ensure data is "available, flowing, and effectively used," focusing on the entire lifecycle management of data application, including sharing, circulation, and utilization [1] Group 2: Public Data Management - Public data is to be shared as a principle, with non-sharing being the exception; data authorities at municipal and district levels are required to manage public data uniformly and create a public data directory [1] - The regulation includes provisions for protecting rights related to data processing and derived data, establishing a foundation for promoting data flow and value realization [1] Group 3: Industrial Cultivation - The regulation highlights the importance of industrial cultivation, detailing policies for nurturing enterprises, establishing an enterprise cultivation database, and promoting multi-dimensional data integration across industries [2] - It supports the construction of high-quality data sets and corpora, as well as the development and training of artificial intelligence large models [2] Group 4: Trusted Data Space - The regulation mandates the establishment of a trusted data space that is manageable, interconnected, and co-creates value, encouraging enterprises to participate in its construction [2] - It aims to leverage public data to facilitate compliant use, secure development, and trustworthy delivery of data, thereby maximizing data value [2]

Group 1 - The modernization of the industrial system is the material and technical foundation of Chinese-style modernization, with advancements in industrial, agricultural, service sectors, and infrastructure development [1] - The Central Economic Work Conference emphasized the continuous promotion of the modernization of the industrial system by 2025, focusing on innovation-driven growth and nurturing new momentum [1][3] - The current focus is on leading new quality productivity development through technological innovation, reinforcing the real economy to gain strategic advantages in future development and international competition [1][2] Group 2 - The domestic and international environments are complex, necessitating a focus on the real economy for modern industrial system construction, supported by policies aimed at stabilizing growth in key industries [2] - The agricultural sector has shown resilience, achieving a stable grain output of over 1.4 trillion jin, despite natural disasters [3] - In November, 30 out of 41 major industrial sectors reported year-on-year growth in added value, indicating steady progress in the industrial economy [3] Group 3 - The integration of technological and industrial innovation is advancing, with more research outcomes transitioning from laboratories to production lines, supported by the establishment of over 2,400 pilot platforms nationwide [7][8] - Key industrial enterprises have improved their CNC rate to 68.5%, and the comprehensive utilization rate of industrial solid waste has reached 57% [8] - The number of high-tech enterprises in the country has reached 504,000, with over 140,000 specialized and innovative small and medium-sized enterprises cultivated, enhancing the innovation capacity of enterprises [9] Group 4 - The integration of industries is a crucial path for building a modern industrial system, with a focus on enhancing economic efficiency and creating new industries and business models [10] - The acceleration of the integration between the real economy and the digital economy is evident, with over 7,000 advanced intelligent factories established [11][12] - The government is promoting the deep integration of manufacturing and service industries, with policies aimed at enhancing the development of modern services alongside advanced manufacturing [12][13]

本报北京12月26日电 记者李春剑从市场监管总局了解到，GB/T 45288《人工智能 大模型》系列国家标 准近日正式实施，标志着我国大模型产业进入"科学权威、统一规范"新阶段。作为首部聚焦通用大模型 的国家标准，该系列填补了技术评价体系空白，明确性能、安全与服务能力要求，配套评测能力已获中 国合格评定国家认可委员会认可。 市场监管总局相关负责人表示，依托这一系列标准构建的"求索"-LMBench评测基准，整合了测评 方法、数据集与自动化平台，已成为行业公认的"能力标尺"，为全链条技术升级提供坚实支撑。 业内专家表示，大模型国家标准的落地实施，不仅解决了产业发展的"度量衡"问题，更通过强链固 链提升了产业链整体韧性。 （来源：光明日报） 转自：光明日报 ...

Group 1 - The modern industrial system is the material and technical foundation of Chinese-style modernization, with advancements in industry, agriculture, services, and infrastructure development [1] - The Central Economic Work Conference emphasized the continuous promotion of the modern industrial system by 2025, focusing on innovation-driven growth and nurturing new momentum [1][2] - The integration of technological and industrial innovation is deepening, with a focus on advanced manufacturing as the backbone of the modern industrial system [1][3] Group 2 - The domestic and international environment is complex, necessitating a focus on the real economy for the construction of a modern industrial system [2] - The State Council has issued guidelines to enhance financial services for the real economy, with increased support for key industries such as steel and automotive [2] - The resilience of the real economy is highlighted by the stable grain production exceeding 1.4 trillion jin and growth in 30 out of 41 industrial sectors in November [3] Group 3 - The integration of new technologies is crucial for enhancing industrial competitiveness, addressing issues of "large but not strong" and "complete but not refined" in certain sectors [6] - The number of high-tech enterprises has reached 504,000, with over 140,000 specialized and innovative small and medium-sized enterprises cultivated [9] - The construction of over 2,400 pilot platforms supports the seamless connection between innovation and industry [7] Group 4 - The integration of digital and physical economies is accelerating, with over 7,000 advanced intelligent factories established and more than 100 million devices connected to industrial internet platforms [11] - The focus on service-oriented manufacturing is growing, with policies promoting the deep integration of manufacturing and service industries [12] - The future outlook for economic development is optimistic, driven by active factor flows and innovation, with a commitment to intelligent, green, and integrated development [12]

Group 1 - The core viewpoint of the article is that China's National Market Supervision Administration has implemented a series of national standards for artificial intelligence large models, marking a new phase of "scientific authority and unified norms" in the large model industry [2] - The series of standards fills a gap in the technical evaluation system, clarifying requirements for performance, safety, and service capabilities, with the evaluation capabilities recognized by the China National Accreditation Service for Conformity Assessment [2] - The "Qiusuo"-LMBench evaluation benchmark has been established, integrating methods, datasets, and an automated platform, becoming an industry-recognized "capability metric" [2] Group 2 - The evaluation system has formed a whitelist of mainstream large models, providing a basis for selection by state-owned enterprises and supporting companies like Sinopec and Southern Power Grid in building industry-specific large models [2] - Over 1,000 evaluation tasks have been completed, with large models called upon more than 950,000 times, accurately identifying common issues such as hallucination control and content safety, aiding nearly 30 manufacturers in technology iteration [2] - The internationalization of the standards is also progressing, with English, Lao, and Khmer versions prepared for ASEAN, filling regional gaps, and the Chinese open-source model architecture has been adopted by projects in countries like Singapore, showcasing the global influence under the standards [2]

2025年12月，安永企业家奖20周年特别访谈现场，云知声智能科技股份有限公司创始人兼首席执行官黄 伟博士下了这个定义。坐在他身旁的东芯半导体股份有限公司董事长蒋学明表示认可，他创业的时间比 20年长了一倍。 黄伟在2012年创立云知声，带着科学家背景闯入人工智能领域，他的公司率先将深度学习应用于产业化 实践。而蒋学明自上世纪80年代起就投身实业，从纺织、水泥再到半导体，他的轨迹几乎是中国产业变 迁的缩影。 两人代表着不同代际的中国硬科技企业家：黄伟是新一代科技创业者，在技术快速迭代中寻找颠覆式创 新的机会；蒋学明是传统产业转型代表，凭借对产业趋势的敏锐洞察不断跨界重生。 而他们也同时身处一个重要的时间节点。2025年，人工智能大模型已从实验室走向产业落地，全球半导 体产业链在地缘政治的撕扯中加速重构。技术热潮如潮水般涌来，也裹挟着泡沫、焦虑与不确定性。 在安永企业家奖20周年的特别访谈中，我们试图厘清几个关键问题：在技术爆炸与市场重构的今天，企 业如何判断趋势与泡沫？出海成为必选项的背景下，如何应对日益复杂的地缘风险？而穿越多个周期的 企业家，又如何理解"坚守"与"变革"之间的平衡？ "20年对一个企业来讲 ...