Core Insights - The conference focused on the integration of artificial intelligence (AI) in asset management, highlighting the historical opportunities presented by "AI + Asset Management" [2][3] - The event gathered over 300 participants from various sectors, including government, finance, and technology, to discuss the application and development trends of AI in asset management [1] Group 1: AI Integration in Asset Management - The Shanghai Lingang New Area aims to become a benchmark financial technology hub by leveraging its unique advantages in financial openness and innovation [2] - The Intelligent Investment Research Technology Alliance (ITL) has grown from 72 to 333 member institutions over five years, indicating a significant expansion in the sector [2] - AI technologies, particularly large models, are transforming asset management processes such as research, advisory, trading, and risk control [3] Group 2: Challenges and Solutions in AI Development - Key challenges in AI model development include the misalignment of AI expansion laws with hardware capabilities, error accumulation in multi-agent collaboration, and ensuring AI safety and value alignment [4] - Solutions proposed include the "Federated Teacher-Student Model" for collaborative learning between general and specialized models [4] - The need for high-quality vertical data, foundational technology platforms, and enhanced AI safety and ethical governance was emphasized as critical for AI's successful implementation in finance [4] Group 3: Practical Applications of AI - AI is being applied in various industries, including satellite design, new materials development, and semiconductor manufacturing, presenting investment opportunities [6][7] - The integration of AI in semiconductor manufacturing is particularly highlighted, with a focus on improving operational efficiency and increasing domestic production rates [6][7] - The financial sector is seeing a shift from "information disparity" to "model disparity," necessitating improvements in data governance and organizational structures to fully leverage AI capabilities [8] Group 4: Future Directions and Ecosystem Development - The establishment of a trusted data space in the securities industry aims to reduce the complexity and cost of applying large models, fostering AI innovation [8] - The importance of a multi-disciplinary team with a blend of curiosity and cross-functional skills was identified as essential for overcoming challenges in AI integration [8][9] - The Shanghai Asset Management Association is focusing on enhancing the global competitiveness of Shanghai as an asset management center through digitalization and AI integration [10]

大模型挑战与金融AI实践 在主题演讲环节，杨强院士深入剖析了人工智能大模型发展面临的五大挑战，包括AI扩展定律与硬件 摩尔定律的不协调、多智能体协同中的错误叠加、人工智能安全与价值观对齐等，并提出了"联邦师生 大模型"的解决方案，通过联邦学习实现通用大模型与垂直小模型的协同学习。 2025年10月17日，第六届1024资管科技开发者大会（ITDC 2025）在上海临港中心（世界顶尖科学家论 坛永久会场）滴水厅成功举办。大会由中国（上海）自由贸易试验区临港新片区管理委员会指导，由上 海资产管理协会、智能投研技术联盟（ITL）、上海临港新片区商会、上海临港新片区投资控股（集 团）有限公司主办。 本次大会是上海资产管理协会主办的"全球资产管理中心上海国际活动周2025"系列重要活动之一，同时 也是智能投研技术联盟（ITL）成员机构发展年会。本次大会以"人工智能+，从产业AI到金融AI"为主 题，汇聚来自政府部门、金融基础设施、银行、保险、公募基金、券商资管、金融科技企业等领域的 300余位嘉宾，共同探讨人工智能技术在资产管理行业的前沿应用与发展趋势。 "AI+资管"正迎来历史性机遇 在大会开幕仪式上，中国（上海）自 ...