过去几年间，在田间地头，在办公楼宇，在工厂车间，金融机构用算法丈量普惠客群的信用半径。中国普惠型小微企业贷款余额从2017年末的8.8万亿元飙 至2024年末的33万亿元上方，年均复合增速达到20.7%，金融服务的核心需求从"有没有"转为"好不好"。 进入2025年后，AI逐步从实验室走向规模化应用。在金融行业，AI跨越单点工具辅助阶段渗透至金融业务的各环节，普惠金融的齿轮加速转动。普惠金融 的"最后100米"，不再是物理距离的阻隔，而是信任、成本与合规的三重考题。AI能否成为解题钥匙？答案藏在每一次模型迭代、每一次监管沙盒的测试 里。 20254 ■ fi AI员工接连上岗 从北京商报记者的调研结果来看，过去十余年间，普惠金融的发展大致可以分为可得性、便捷性、精准性三个阶段，实际上就是普惠客群金融服务需求 从"有没有"转向"好不好"的过程。 普惠金融发展的关键阶段，AI被寄予厚望。而普惠金融的核心矛盾，始终是信任与成本的博弈，同时，随着AI应用的深入，科技伦理的治理受到更多关 注。 2025年以来，以大模型和智能体为代表的生成式AI技术迎来突破性发展，AI从执行指令的"自动化工具"演进为具备推理能力和自主决 ...

Core Insights - The core viewpoint of the articles emphasizes the transformative role of AI in enhancing inclusive finance, particularly for small and micro enterprises in China, with a significant increase in loan balances and a shift in focus from availability to quality of financial services [1][10]. Group 1: Growth of Inclusive Finance - The balance of inclusive loans for small and micro enterprises in China surged from 8.8 trillion yuan at the end of 2017 to over 33 trillion yuan by the end of 2024, achieving a compound annual growth rate of 20.7% [1]. - By the third quarter of 2025, the balance of inclusive loans for small and micro enterprises reached 36.5 trillion yuan, reflecting a year-on-year growth of 12.1% [10]. Group 2: AI Integration in Financial Services - Since 2025, generative AI technologies have evolved from automation tools to business partners, leading to systematic and large-scale applications in the financial sector [5]. - AI applications in finance have expanded from isolated attempts to comprehensive solutions, enhancing efficiency in credit approval, fraud detection, and investment research [5][6]. Group 3: Challenges in Trust, Cost, and Compliance - The development of inclusive finance has transitioned through three stages: availability, convenience, and precision, highlighting the shift in financial service demands from "whether" to "how good" [7]. - Trust issues arise as traditional risk assessment methods struggle with the unique characteristics of the inclusive customer base, leading to reliance on alternative data and concerns over algorithmic fairness [7][8]. - The costs associated with AI implementation, including model training and compliance verification, pose significant challenges for financial institutions, potentially eroding profits [8]. Group 4: Innovations and Solutions - Financial institutions are increasingly collaborating to address the challenges in inclusive finance, focusing on technology innovation and industry cooperation [9]. - AI technologies are evolving towards lighter and more precise models to reduce dependency on large datasets and lower implementation costs [9]. - Customized AI applications are being developed to cater to specific scenarios, such as the "data credit" model in rural finance, which replaces traditional collateral methods [9]. Group 5: Future Trends and Regulatory Framework - The gradual improvement of regulatory frameworks is establishing a risk baseline for the large-scale application of AI in finance, with initiatives like regulatory sandboxes allowing for innovation while managing risks [9]. - The integration of AI in inclusive finance is expected to enhance productivity, improve service quality, and lead to ongoing advancements in technology regulation [10].

AI员工接连上岗 2025年以来，以大模型和智能体为代表的生成式AI技术迎来突破性发展，AI从执行指令的"自动化工 具"演进为具备推理能力和自主决策的"业务伙伴"，除了AI客服外，还有更多AI员工上岗。 北京商报记者向业内十余家金融机构调研发现，当前AI在金融领域的应用不再是零星的点状尝试，而 是呈现出体系化、规模化的特征。从信贷审批的分钟级响应到反欺诈的秒级拦截，从投研报告的智能生 成到普惠服务的全域覆盖，AI向金融领域更深层次的业务赋能演进。 过去几年间，在田间地头，在办公楼宇，在工厂车间，金融机构用算法丈量普惠客群的信用半径。中国 普惠型小微企业贷款余额从2017年末的8.8万亿元飙至2024年末的33万亿元上方，年均复合增速达到 20.7%，金融服务的核心需求从"有没有"转为"好不好"。 进入2025年后，AI逐步从实验室走向规模化应用。在金融行业，AI跨越单点工具辅助阶段渗透至金融 业务的各环节，普惠金融的齿轮加速转动。普惠金融的"最后100米"，不再是物理距离的阻隔，而是信 任、成本与合规的三重考题。AI能否成为解题钥匙？答案藏在每一次模型迭代、每一次监管沙盒的测 试里。 以智能投顾为例，从业机构 ...

Core Insights - The September 2025 Bank Social Media Index (SMI) highlights a significant rise in the rankings of several banks, with Industrial Bank's Douyin account "Qian Da Zhang Gui" achieving a remarkable jump to the second position due to its effective content strategy [1] - New entrants to the top rankings include Postal Savings Bank, Citic Bank, and Huishang Bank, indicating a shift in social media engagement within the banking sector [1] Group 1: Performance Highlights - Industrial Bank's Douyin account "Qian Da Zhang Gui" rose from outside the top rankings to second place, showcasing its balanced operation across various platforms including video accounts and public accounts [1] - Postal Savings Bank's content titled "New Atmosphere of Local Specialties" received 22,000 likes, emphasizing the bank's role in supporting local agriculture through comprehensive financial services [1] - The "100 Small Shops" series by MyBank highlighted a small business owner's innovative approach to offering affordable meal options, resonating well with users and contributing to the bank's social media presence [1] Group 2: Content Themes - The top content on Douyin this month featured diverse themes such as product promotion, brand awareness, financial education, and community support, reflecting a broad engagement strategy [1][2] - MyBank's focus on AI in finance was evident in its content, which showcased the practical applications of AI technology in credit services, aligning with current industry trends [2] - NewNet Bank's engagement with young audiences during the back-to-school season aimed to establish emotional connections and foster brand loyalty among future customers [2] Group 3: Social Media Engagement - The top-performing content on WeChat included discussions around personal consumption loan policies, indicating a growing interest in financial literacy among consumers [2] - Various banks utilized the back-to-school theme to create relevant content, such as Hengfeng Bank's practical financial advice for college students, which ranked sixth in user engagement [1][2]

Core Insights - The banking industry is entering a golden decade for AI applications, with significant changes expected in service models due to AI's capabilities [1] - The new "310" model in the AI era emphasizes comprehensive user understanding and real-time interaction, contrasting with the previous model focused on online credit services [1][2] - AI is not merely an automation tool but exhibits human-like cognitive and execution abilities, marking a fundamental shift in banking services [1] Group 1: AI Implementation in Banking - The first step in AI integration involves equipping employees with AI assistants to enhance individual efficiency and promote equitable financial services [3] - The second step focuses on transforming service paradigms from product provision to comprehensive solutions that include financial products, services, and business advisory [3] Group 2: Addressing Small and Micro Enterprises' Needs - NetEase Bank aims to stabilize credit limits for small and micro enterprises by providing "AI credit experts," achieving a consistency rate of 90% in AI credit decisions compared to human approvals [3] - AI-driven marketing strategies are being developed to accurately identify user needs and deliver tailored financial products at the right time [3] - The bank is also working on stabilizing investment returns for small and micro enterprises through intelligent operational engines, reducing yield fluctuations by 5 basis points compared to market averages [3]