Investment Rating - The report emphasizes that the highest return on investment in 2026 will come from modern data infrastructure rather than the latest AI models [11][14]. Core Insights - The gap between AI ambitions and actual operations is widening across industries, with organizations needing to focus on foundational work to achieve transformative wins [6][5]. - Companies are shifting from broad experimentation to specific, high-value use cases, moving AI from proof-of-concept to enterprise-level deployment [19][15]. - Successful organizations are prioritizing data lifecycle management, modernization, and human capabilities to shape their AI-driven transformation strategies [48][59]. Summary by Sections Overview - The report highlights a significant disconnect between organizational expectations of AI and the foundational work required for successful implementation [6][5]. - Many companies are still relying on outdated systems and manual processes, which hinders their ability to leverage AI effectively [6][9]. 2026 AI Trends - AI success in 2026 will be driven by data infrastructure rather than new models, with a focus on creating accessible and real-time data management systems [11][12]. - Organizations are expected to transition from broad AI experiments to targeted applications that deliver measurable business value [15][17]. Industry-Specific Trends - The report outlines predictions for various sectors, including: - Airlines will require rapid experimentation due to competitive pressures [65]. - Retail will see AI operating behind the scenes, influencing pricing and supply chain decisions [66]. - Healthcare will experience regulatory advancements that promote AI-driven innovations [69]. Preparing for 2026 - Companies need to invest in data management and governance to support AI initiatives effectively [48][51]. - A cultural shift is necessary for organizations to embrace AI as a core component of their business strategy rather than a standalone project [30][60]. Conclusion - The foundation laid in data infrastructure and governance will determine the success of AI initiatives in 2026, with companies that prioritize these areas likely to thrive [87][89].

Core Viewpoint - Asterinas has launched Asterinas NixOS, the first distribution featuring a pure Rust operating system kernel, marking a significant step towards native memory safety in the OS industry [1][2]. Group 1: Asterinas NixOS Overview - Asterinas NixOS integrates a pure Rust kernel with the mature software ecosystem of NixOS, transitioning from a kernel project to a usable system [1]. - The project has been developed over three years, contributing 130,000 lines of Rust code and supporting over 230 Linux system calls [2]. - The release signifies a milestone, enabling Asterinas to support a user space for installation and experience [2]. Group 2: Technical Advantages - Asterinas NixOS is built on the innovative "framekernel" architecture, providing high performance and security compared to traditional OS kernels [2]. - It inherits the "system state stability reproduction" feature from NixOS, allowing developers to automatically construct systems based on specified software and configuration needs, ensuring consistent results across different machines and times [3]. - The integration with NixOS's extensive software ecosystem, which includes over 120,000 packages, allows Asterinas NixOS to avoid starting from scratch, facilitating a quicker transition to a mature system [3]. Group 3: Industry Trends and Future Plans - The trend in the industry is shifting towards native memory safety, with major players like Microsoft planning to migrate core codebases to Rust by 2030 [1][4]. - Asterinas NixOS aims to establish a reproducible and comparable baseline in real application loads, positioning itself at the forefront of memory-safe kernel engineering [4]. - The latest version, Asterinas 0.17.0, meets core user needs for desktop and server environments, with plans for deployment in secure computing and trusted data spaces by 2026 [4].

截至发稿，品茗科技市值为114亿元。 每经头条（nbdtoutiao）——对话马斯克脑机接口"一号受试者"：大脑植入芯片23个月，我正重新夺回 人生的独立 每经AI快讯，品茗科技（SH 688109，收盘价：145元）12月26日晚间发布公告称，2025年9月24日，公 司控股股东、实际控制人莫绪军、股东李军、陶李义、李继刚及新余灵顺灵创业投资合伙企业（有限合 伙）与通智清研（北京）科技产业发展合伙企业（有限合伙）共同签署了《股份转让协议》，转让方拟 将其所持有的公司约1255万股无限售流通股（占公司总股本的约15.92%），通过协议转让的方式转让 给受让方。本次股份协议转让事项已通过上海证券交易所合规性审核，并于2025年12月26日取得中国证 券登记结算有限责任公司上海分公司出具的《证券过户登记确认书》，过户日期为2025年12月25日，过 户数量约为1255万股（占公司股份总数的约15.92%），股份性质为无限售流通股，本次股份协议转让 过户登记手续已办理完毕。 （记者 曾健辉） 2024年1至12月份，品茗科技的营业收入构成为：软件业占比100.0%。 ...

Core Viewpoint - Kingdee International has maintained its leading position in the Chinese low-code and no-code software development market with an 18.1% market share, as reported by IDC, marking five consecutive years at the top [1] Group 1: Financial Performance - Since 2020, Kingdee International has faced significant financial challenges, accumulating losses of 1.476 billion yuan over five and a half years [2][9] - In the first half of 2025, the company reported a revenue of 3.192 billion yuan, a year-on-year increase of 11.24%, while its cloud business revenue reached 2.673 billion yuan, growing by 11.9% [9][10] - The net loss for the first half of 2025 was approximately 977.38 million yuan, representing a 55.14% reduction in losses compared to the previous year [9][12] Group 2: Business Segments - Kingdee's cloud subscription revenue for the first half of 2025 was 1.684 billion yuan, with a year-on-year growth of 22.1% and a gross margin of 96.2% [10] - The company’s financial services, including its credit platform, have faced scrutiny due to regulatory penalties, raising concerns about compliance and sustainability [2][5] - Kingdee's financial segment has been under pressure, with a focus on consumer finance through its loan app, Fangxin Wallet, which has faced numerous consumer complaints [6][18] Group 3: Strategic Initiatives - Kingdee International is pivoting towards AI integration and overseas market expansion as key growth strategies, with plans to fully transition into an AI-driven enterprise management company by 2025 [15][17] - The company has begun to establish a presence in Southeast Asia and the Middle East, with overseas revenue growing significantly but still representing a small portion of total income [16][18] - Kingdee's commitment to AI has led to a contract signing amount of 150 million yuan related to AI in the first half of 2025 [15]

Core Insights - Billionaire investor Mark Cuban advocates for new graduates to pursue jobs in small to medium-sized businesses (SMBs) to leverage their skills in agentic AI [1][3] - Cuban emphasizes that AI agents, which can perform tasks autonomously, represent a significant opportunity for job seekers to add value in SMBs, where such expertise is often lacking [2][4] Group 1: Job Market Insights - Cuban suggests that new graduates should focus on SMBs rather than large corporations, as their AI skills will be more valuable and noticeable in smaller firms [3] - The rise of agentic AI is highlighted, with spending projected to reach $155 billion by 2030, indicating a growing market for AI technologies [2] Group 2: AI Adoption Trends - A report indicates that the adoption of agentic AI in SMBs increased from 50% in December 2024 to 82% in May 2025, showcasing a rapid embrace of this technology [4] - The current job market shows that one-fourth of America's unemployed are college graduates, reflecting challenges in the employment landscape for this demographic [4]

Group 1 - The core viewpoint of the news is that Beijing Defeng New Journey Technology Co., Ltd. (Defeng Technology) has had its IPO application returned by the Hong Kong Stock Exchange (HKEX), marking the first "tied" case in two and a half years, which requires the company to wait at least 8 weeks before reapplying [1] - The "tied" mechanism is a criticism notification system implemented by HKEX to enhance the quality of listed companies, with a total of 25 applications returned since its inception, involving major underwriters like Goldman Sachs and Morgan Stanley [1] - Defeng Technology, established in March 2015, focuses on AI-enabled Industrial Internet of Things (AIoT) production optimization software solutions, having completed over 600 projects primarily for state-owned enterprises in China [2] Group 2 - The company had previously attempted to list on the Shanghai Stock Exchange's Sci-Tech Innovation Board in 2022 but terminated those plans in October 2024, subsequently applying for a listing on HKEX in November 2023 [2] - Defeng Technology's IPO involves a special agreement granting investors rights such as redemption rights and anti-dilution rights, with a revised IPO target date extended to December 31, 2026 [2] - The company has incurred redemption liabilities of RMB 1.224 billion, RMB 1.455 billion, and RMB 1.462 billion for the years 2023, 2024, and the first half of 2025, respectively, due to the granted redemption rights [3]

Core Viewpoint - The article discusses the emergence of Triton as a user-friendly programming tool for AI chip development, aiming to lower the barriers for developers previously reliant on complex languages like CUDA [1][2][3]. Group 1: Triton Overview - Triton allows developers to write high-performance GPU code in a Python-like syntax, making it accessible to a broader audience [3]. - The integration of Triton with the PyTorch ecosystem enhances its usability and performance, making AI chip programming more approachable [3]. Group 2: Triton Next Conference - The Triton Next conference is scheduled for January 9, 2026, in Beijing, organized by the FlagOS community and the Beijing Academy of Artificial Intelligence [4][16]. - The conference aims to explore the current state and future developments of Triton, including discussions on its compiler and potential applications [5][6]. Group 3: Conference Agenda - The morning sessions will focus on the foundational principles of Triton, recent academic research, and the latest developments in the FlagOS community [7]. - Afternoon sessions will highlight practical applications of Triton, featuring insights from various teams on building next-generation AI models and addressing hardware compatibility challenges [11]. Group 4: Workshops and Training - The conference will include hands-on workshops designed to help developers apply Triton in real-world scenarios, covering topics like operator training and compiler usage [15][18].

Core Insights - The article discusses the global surge in AI technology applications, highlighting the importance of efficient digital tools for businesses and individuals to compete globally [1][3] - Baidu's product "MiaoDa" aims to break down traditional software development barriers, allowing users to realize their creative ideas through natural language without needing coding skills [1][4] Industry Trends - There is a growing demand for efficient application development tools as AI technology penetrates the global market [3] - China possesses the largest application market and diverse practical scenarios, providing a fertile ground for the iteration of AI tools [3] Company Developments - Baidu's "MiaoDa" has achieved significant success, serving over 10 million users and creating social value exceeding 5 billion [4] - The product allows users to complete the entire software development process through AI, fundamentally changing traditional software development methods [4] Competitive Advantage - "MiaoDa" builds a differentiated core competency by focusing on productivity creation in the intelligent era, contrasting with traditional development models [5] - The product operates on a "human proposes ideas, AI delivers everything" model, positioning itself at the forefront of human-machine collaboration [5] Future Outlook - Baidu plans to enhance AI technology innovation and improve the global adaptability of "MiaoDa," combining domestic experience with local market needs [5] - The company believes that the global promotion of "MiaoDa" will unlock a vast creative market, expanding from 30 million professional developers to billions of potential creators worldwide [5][6]

JarvisEvo团队 投稿 量子位 | 公众号 QbitAI 下面就来了解一下详细情况吧～ 自我评估和修正 研究背景与动机 近年来，基于指令的图像编辑模型虽然取得了显著进展，但在追求"专业级"修图体验时，仍面临两大核心挑战： 1. 指令幻觉 (Instruction Hallucination)： 现有的文本思维链 (Text-only CoT) 存在信息瓶颈。模型在推理过程中"看不见"中间的修图结果，仅凭文本"脑补"假设进行下一步操作的 视觉结果，容易导致事实性错误，无法确保每一步都符合用户意图。 一句话让照片变大片，比专业软件简单、比AI修图更可控！ 腾讯混元携手厦门大学推出 JarvisEvo ——一个统一的图像编辑智能体模拟人类专家设计师，通过 迭代编辑、视觉感知、自我评估和自我反 思 来"p图"。 "像专家一样思考，像工匠一样打磨" 。JarvisEvo不仅能用Lightroom修图，更能"看见"修图后的变化，并自我评判好坏，从而实现无需外部 奖励的自我进化 。 2. 奖励黑客 (Reward Hacking)： 在强化学习进行偏好对齐的过程中，策略模型（Policy）是动态更新的，而奖励模型（R ...

Retirement is complicated and you only get once chance to do it right. Don't miss out because you didn't know what was out there.The Retirement Forum provides actionable ideals, a high-yield safe retirement portfolio, and macroeconomic outlooks, all to help you maximize your capital and your income. We search the entire market to help you maximize returns.Microsoft ( MSFT ) is one of the largest tech companies, with a more than $3.5 trillion market cap. The company is incredibly well positioned with artific ...