|
Work item:
|
QSTR.ACC
|
|
Subject/title:
|
Signalling and protocol considerations on networking for AI-agent collaboration
|
|
Status:
|
Under study
|
|
Approval process:
|
Agreement
|
|
Type of work item:
|
Technical report
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
2028-12 (Medium priority)
|
|
Liaison:
|
ITU-T SG13, ITU-T SG17, ITU-T SG21, 3GPP SA1, IETF Network Working Group
|
|
Supporting members:
|
China Telecom, China Unicom, China Information Communication Technologies Group (CICT)
|
|
Summary:
|
With the rapid advancement of artificial intelligence (AI) technologies, extensive research has been conducted specifically on large language models (LLMs) and multi-agent systems. AI-agents are evolving from "isolated intelligence" toward "collective intelligence." Under this trend, AI-agents are no longer merely standalone program modules; instead, they are becoming autonomous entities that must collaborate with one another, share information, and jointly accomplish complex tasks.
However, the current AI-agent ecosystem is highly fragmented. AI-agents developed by different vendors, platforms, and frameworks often rely on proprietary communication mechanisms and lack unified standards. This results in poor interoperability, high integration costs, and low collaboration efficiency, which significantly hinder the scalable development and value realization of AI-agent networks.
To better leverage the advantages of AI-agent communication, protocols such as ACP, A2A, and MCP have been proposed. These protocols aim to establish a standardized, efficient, and scalable communication framework for heterogeneous AI-agents, enabling AI-agents to collaborate freely within an AI-agent communication network.
Currently, ITU-T SG13 has initiated several ongoing work items on AI-agents. [ITU-T Y.supp.NAC-roadmap] “Networking for AI agent collaboration - Standardization status and roadmap” focuses on describing the standardization status and roadmap of networking for AI-agent collaboration, collecting related Recommendations and Technical reports, and providing a roadmap to the possible standardization direction of networking for AI-agent collaboration. [ITU-T YSTR.NAC] “Framework of networking for AI agent collaboration in future network” studies the requirements and framework of networking for AI-agent collaboration in future networks (as shown in Figure 6-4). [ITU-T YSTR.ACN] “AI agent communication network in IMT-2020 networks and beyond” studies the AI-agent communication network in the context of IMT-2020 and beyond. ITU-T SG11 has initiated one ongoing Technical Report [ITU-T QSTR.SPACN] “Signalling and protocol for AI-agent communication network in IMT-2020 networks and beyond” studies the signalling and protocol for AI-agent communication network in IMT-2020 networks and beyond.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2026-03-13 14:30:07
|
|
Last update:
2026-03-13 14:39:51
|
|
