Guidelines  ·  2026-11-20

IETF draft-jeong-nmrg-agentic-ai-framework-01: 'An Agentic AI Framework for the Management of Networks, Security, and AI Services' (revision)

GuidelinesMedium impactGlobal
Revision -01 of the IRTF Network Management Research Group Internet-Draft 'An Agentic AI Framework for the Management of Networks, Security, and AI Services' (last updated 2026-11-16). It specifies an intent-based agentic AI framework that enforces human-language intent as translated policy across a User System, Cloud System, Edge System, and Device, targeting network forwarding elements, network security functions (firewall, web filter, DDoS mitigator), IoT devices, physical AI robots, software-defined vehicles, UAVs, and urban air mobility. Components include AI agents and a knowledge repository, with inter-agent interfaces using the A2A protocol; it also includes a security-considerations section and a 'changes from -00' appendix.
Extends agentic-AI standardization into network/security service management: describes how agents autonomously manage security functions and enforce intent-based policy across network/edge/cloud — directly relevant to securing and operating agent-driven security infrastructure.
Track the framework and A2A interfaces as agentic network-security management matures; consider for intent-based security design.
IETF Datatracker — draft-jeong-nmrg-agentic-ai-framework-01
See this in the live feed Explore related AI security and governance findings — updated every morning.
Open the feed →