Core Insights - The rapid development of large model technology has significantly lowered the barriers for AI innovation in the financial industry, shifting focus from basic models to innovative applications of large models [1][3] - Companies are intensively developing intelligent agent applications tailored to various financial business scenarios, leading to a diverse landscape of large model applications [1][3] Group 1: Large Model Applications in Finance - The application effects of large models show significant differentiation, with challenges in meeting high accuracy requirements and providing quality user experiences in core financial business scenarios [1] - Tencent's strategy in the large model field emphasizes the integration of AI capabilities across various business scenarios, enhancing user experience and iterating solutions based on user pain points [3][5] - The "Cloud-Data-Model-Application" flywheel model proposed by Tencent outlines how financial enterprises can optimize foundational models using their business scenarios and data [3][5] Group 2: Financial AI Solutions - Tencent Cloud provides a comprehensive solution for financial institutions, integrating technical capabilities and rich ecosystems to support AI application development [5][7] - The intelligent agent development platform facilitates efficient construction of AI applications for financial institutions, enabling the implementation of specific use cases [9] - Tencent's financial cloud has developed a complete matrix of large model applications covering core business scenarios in banking, asset management, and insurance [10][11] Group 3: Specific Use Cases - The enterprise knowledge base scenario addresses challenges in knowledge fragmentation and retrieval efficiency in the financial sector, with Tencent's solution promoting knowledge integration [12][15] - The credit due diligence assistant significantly reduces the report generation time from 10 days to 1 hour, enhancing efficiency by tenfold [20] - The insurance agent assistant optimizes workflows and improves the quality of service by leveraging AI to assist agents in their daily tasks [24][26] Group 4: Future Trends - The integration of data and AI is seen as a core trend, moving from isolated technological breakthroughs to deep industry restructuring [36][39] - Companies are encouraged to view data governance and AI model development as an integrated process, enhancing overall efficiency and breaking down data silos [36][40] - Tencent's Data+AI solution aims to help clients unlock data value and achieve monetization through various AI applications in the financial sector [39][41]

6月27日消息，今天上午举行的全球人工智能开发与应用大会上，腾讯云大数据基础产品中心总经理程 彬透露，腾讯云已经构建完善的"Data+AI"能力，今年下半年将发布数据智能体产品。 程彬表示，数据作为传统数据平台、AI大模型平台的共同"石油原料"，"Data+AI"不仅仅是数据的简单 叠加，更是深度融合与创新，实现从"让数据说话"到"与数据智能对话"的跃迁。 数据显示，2024年国内数据生产总量首次突破40ZB，达到41.06ZB，同比增长25%，增速较去年提高 2.56个百分点；人均数据生产量约为31.31太字节（TB），相当于1万多部高清电影，同比增长25.17%， 数据生产总量和人均产量实现同步跃升。 最新发布的《Data+AI下一代数智平台建设指南》报告显示，Gartner研究表明，非结构化数据占当今 组织数据的70%至90%。受生成式AI计划、多模态数据处理需求的爆炸式增长以及合规性压力的推 动，企业对非结构化数据管理的需求急剧增长。 因此，非结构化数据处理支出在数据管理总支出中所占的份额将越来越大。 随着AI大模型对于数据的需求增加，传统数据平台在应对生成式AI带来的新型数据需求时，正面临严 峻挑战 ...