Committed to connecting the world

  •  

ITU-T work programme

[2025-2028] : [SG20] : [Q7/20]

[Declared patent(s)]  - [Associated work]

Work item: Y.EI-AIS
Subject/title: Evaluation framework and indicators of IoT AI agent services
Status: Under study 
Approval process: AAP
Type of work item: Recommendation
Version: New
Equivalent number: -
Timing: 2028-Q4 (Medium priority)
Liaison: ITU-T SG13
Supporting members: China Academy of Information and Communications Technology, Beijing University of Posts and Telecommunications
Summary: Currently, the deep integration of large AI models and IoT technologies is driving fundamental changes in industrial paradigms. As a core carrier connecting the physical world and the digital world, the AI agent system in Internet of Things has evolved from simple command execution to an intelligent service entity with environmental perception, intention understanding, autonomous decision-making, and collaborative execution capabilities. However, existing agent standards mostly focus on underlying communication protocols or single algorithm performance, lacking systematic classification and quantitative evaluation of agents, leading to vague definitions of agent capabilities and uneven product services, which severely restricts the large-scale deployment and cross-platform interoperability of agents in the IoT field. This draft Recommendation will provide the system architecture for AI agent systems in Internet of Things, specifies the evaluation requirements and evaluation indicators for AI agent systems in IoT environments. By establishing an evaluation indicator system covering six major categories and twenty-two detailed indicators, including accuracy, robustness, explainability, efficiency, collaboration, and security, this recommendation constructs an evaluation indicator system covering six core dimensions. It will specify qualitative detailed evaluation requirements under each dimension, providing a unified, implementable, and traceable evaluation basis for the global AI agent in IoT industry, supporting cross-platform interoperability and trustworthy ecosystem construction.
Comment: -
Reference(s):
  Historic references:
-
Contact(s):
Zhen GUO, Editor
Yang LIU, Editor
Ziqin SANG, Editor
Xiaojing SHAO, Editor
Bin XIE, Editor
Zihang YIN, Editor
ITU-T A.5 justification(s):
Generate A.5 drat TD
-
[Submit new A.5 justification ]
See guidelines for creating & submitting ITU-T A.5 justifications
First registration in the WP: 2026-05-27 13:55:33
Last update: 2026-07-16 16:35:57