Investment Rating - The report does not explicitly state an investment rating for the industry Core Insights - The communication industry is transitioning from 4G/5G/5G-A to 6G, marking the beginning of a new era of connectivity that empowers vertical industries such as industrial internet, smart grids, and intelligent transportation [4] - The development of self-intelligent networks (AN) is crucial, with a focus on integrating AI and intelligent agents to enhance network autonomy and operational efficiency [5][6] - The TM Forum's self-intelligent network white paper 7.0 outlines a framework for self-intelligent networks, emphasizing the need for structured methods to design and integrate intelligent agents across the network [14] Summary by Sections 1. Definition of Self-Intelligent Network Agents - Self-intelligent network agents (AN Agents) are defined as autonomous systems that make decisions based on their perceptions and goals without human intervention, categorized into digital assistants and self-closed-loop agents [13][14] 2. High-Level Architecture of Self-Intelligent Networks - The architecture focuses on a three-layer model with autonomous domain capabilities, emphasizing the need for collaboration between different layers and the integration of AI technologies [15][19] 3. Key Technologies and Development Directions - The report identifies critical technologies such as communication large models, digital twins, intelligent collaboration, and intelligent learning as essential for the functionality of self-intelligent network agents [34][41] 4. Building Large Models in Self-Intelligent Network Agents - The report discusses the importance of specialized, pre-integrated, and co-evolving large models to enhance the capabilities of self-intelligent network agents [17][18] 5. Interface Protocols for Self-Intelligent Network Agents - Various paradigms for integrating intelligent agents are explored, including A2A-T protocols that facilitate seamless communication between agents [19][24] 6. Standards Progress and Implementation Recommendations - The report emphasizes the need for a standardized ecosystem to support the interoperability of self-intelligent network agents across different vendors [23][24] 7. Transition from Traditional Networks to Self-Intelligent Network Architectures - The evolution towards self-intelligent network architectures is driven by the need for enhanced operational efficiency and the ability to manage complex network environments autonomously [26][27] 8. Huawei ADN Intelligent Agent Practices - Huawei's ADN practices demonstrate the application of intelligent agents in real-world scenarios, showcasing significant improvements in operational efficiency and fault management [10][11] 9. Future Outlook for Self-Intelligent Network Agents - The report anticipates a shift from high-value scenario autonomy to comprehensive network autonomy, enabling operators to achieve end-to-end management and improved service quality [7][9]