← Reading listAgendaReviewdatatracker ↗ draft-jadoon-dawn-layered-agent-discovery-00 · none · 8 pp · 2026-11-02 · offline copy



DAWN                                                        M. A. Jadoon
Internet-Draft                                              S. Robitzsch
Intended status: Informational                   InterDigital Europe Ltd
Expires: 6 May 2027                                      C. J. Bernardos
                                        Universidad Carlos III de Madrid
                                                         2 November 2026


Discovery of AI Agents for Communication Services: Use Case and Problem
                               Statement
              draft-jadoon-dawn-layered-agent-discovery-00

Abstract

   This document presents a use case and problem statement for the
   initial discovery of AI agents supporting end-to-end communication
   services within an organization.  The use case considers AI agents
   associated with different functions of the Internet protocol suite,
   including application, transport, Internet-layer, and related
   network-operation functions.  These agents may share general
   capabilities but differ in their communication-related operational
   scopes and responsibilities.  The document examines the minimum
   public discovery information needed to distinguish such agents within
   a known organizational discovery scope.

Status of This Memo

   This Internet-Draft is submitted in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF).  Note that other groups may also distribute
   working documents as Internet-Drafts.  The list of current Internet-
   Drafts is at https://datatracker.ietf.org/drafts/current/.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   This Internet-Draft will expire on 6 May 2027.

Copyright Notice

   Copyright (c) 2026 IETF Trust and the persons identified as the
   document authors.  All rights reserved.





Jadoon, et al.             Expires 6 May 2027                   [Page 1]

Internet-Draft  AI Agent Discovery for Communication Ser   November 2026


   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents (https://trustee.ietf.org/
   license-info) in effect on the date of publication of this document.
   Please review these documents carefully, as they describe your rights
   and restrictions with respect to this document.  Code Components
   extracted from this document must include Revised BSD License text as
   described in Section 4.e of the Trust Legal Provisions and are
   provided without warranty as described in the Revised BSD License.

Table of Contents

   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   2
   2.  Terminology . . . . . . . . . . . . . . . . . . . . . . . . .   3
   3.  Use Case  . . . . . . . . . . . . . . . . . . . . . . . . . .   4
   4.  Problem Statement . . . . . . . . . . . . . . . . . . . . . .   5
   5.  Security Considerations . . . . . . . . . . . . . . . . . . .   6
   6.  IANA Considerations . . . . . . . . . . . . . . . . . . . . .   6
   7.  Informative References  . . . . . . . . . . . . . . . . . . .   6
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .   7

1.  Introduction

   The Internet protocol suite separates application, transport,
   Internet, and link-layer responsibilities [RFC1122] [RFC1123].  Each
   layer performs specific functions and interacts with other layers
   through established interfaces and protocols.  This separation
   supports interoperability and allows the implementation and evolution
   of protocols independently of one another.

   AI agents can augment these functions through adaptive decision-
   making and automation within their respective operational boundaries.
   For example, an application-associated agent may adapt media
   generation according to application requirements, while a transport-
   associated agent may assist with congestion control and transmission
   behavior.  Similarly, an Internet-layer-associated agent may assist
   with routing and forwarding decisions, while agents associated with
   network operations may support packet scheduling, prioritization, and
   resource management.

   In a layered agentic architecture, AI agents may operate
   independently within their functional boundaries and may cooperate
   with each other to support an end-to-end communication service.  Each
   agent may use observations and interfaces relevant to its associated
   functions.

   Research on cross-layer agentic networking illustrates cooperation
   among specialized agents [SANET], while work on intent-aware
   transport intelligence for immersive applications illustrates



Jadoon, et al.             Expires 6 May 2027                   [Page 2]

Internet-Draft  AI Agent Discovery for Communication Ser   November 2026


   adaptation at a particular protocol layer [XR-TRANSPORT].  The
   architectural motivation for agentic augmentation of the layered
   protocol suite is further discussed in [LAYERED-AGENTS].

   The ETSI Technology Radar identifies emerging ICT trends and assesses
   their potential implications for future standardization activities
   [ETSI-RADAR].  The deployment of specialized AI agents across
   communication functions raises interoperability considerations when
   agents need to cooperate to support shared service objectives.  One
   prerequisite for such cooperation is the ability to discover agents
   exposing the required communication services and identify their
   associated protocol-layer responsibilities.

   Before such interactions can occur, a Discovering Entity may need to
   obtain information about other AI agents that are discoverable within
   a given organizational scope.  Similar high-level descriptions of
   communication-related services may refer to agents operating within
   different functional and operational boundaries.

   This use case considers initial discovery within an organization, one
   of the deployment contexts identified by the DAWN Working Group
   [DAWN-CHARTER].  Subsequent capability exchange, agent selection,
   authorization, and operational coordination are outside the scope of
   this document.

2.  Terminology

   This document uses the terminology defined in [DAWN-TERMS].  The
   following additional terms are defined for the purposes of this
   document.

   Communication Function:  In this document, a specialization of the
      DAWN term _Function_, referring to processing that supports an
      end-to-end communication service.  Examples include application
      adaptation, transport optimization, routing, forwarding, and
      packet scheduling.  A communication function describes the general
      service provided by an agent, rather than its detailed
      capabilities or internal implementation.

   Layered Agentic Architecture:  An architecture in which AI agents are
      associated with different communication functions across the
      Internet protocol suite and related network operations.  These
      agents may cooperate to support end-to-end communication service
      objectives while operating within their respective functional
      responsibilities.

   Layer-Associated AI Agent:  An AI agent whose advertised




Jadoon, et al.             Expires 6 May 2027                   [Page 3]

Internet-Draft  AI Agent Discovery for Communication Ser   November 2026


      communication-related service is associated with one or more
      protocol-layer functions or related network operations.

   Operational Domain:  An administrative or technological environment
      within which an AI agent provides its advertised communication
      function.  An Operational Domain may be a subdivision of an
      organization's Discovery Scope.  It describes the context of the
      agent's functional responsibility and does not, by itself,
      establish authorization to interact with the agent.

   In this document, the unqualified term _agent_ refers to an AI agent
   unless otherwise stated.

3.  Use Case

   Consider an interactive augmented-reality (AR) application in an
   organization supporting maintenance in an industrial facility.  A
   user device exchanges application traffic with a service through an
   access network and one or more intermediate networks.  The access may
   be wired or wireless, and the service may be hosted locally or
   remotely.  Application traffic and network load can affect the
   quality and continuity of the communication service.

   The organization employs multiple agents to assist with different
   aspects of the communication service.  An application-associated
   agent may adapt the resolution of AR content according to application
   requirements.  A transport-associated agent may monitor transport
   performance and assist with transmission adaptation through
   interfaces permitted by the relevant transport implementation.
   Similarly, Internet-layer-associated agents may support routing and
   forwarding operations, while agents associated with network
   operations may assist with packet scheduling, prioritization, or
   resource management.

   These agents may use different observations, interfaces, and
   operational controls.  For example, an application-associated agent
   may observe media quality and application performance, while a
   transport-associated agent may observe round-trip time, loss, or
   transmission behavior.  A forwarding-associated agent may use routing
   or topology information, whereas a scheduling-associated agent may
   rely on queue occupancy, traffic priority, or link conditions.

   Although these agents contribute to the same end-to-end communication
   service, their operational responsibilities remain associated with
   different communication functions.






Jadoon, et al.             Expires 6 May 2027                   [Page 4]

Internet-Draft  AI Agent Discovery for Communication Ser   November 2026


   Within the organization, multiple AI agents may expose related high-
   level service descriptions.  For example, a transport-associated
   agent and an Internet-layer-associated agent might both advertise
   services concerned with traffic optimization.  However, the former
   may support transport adaptation, while the latter may support
   routing or forwarding operations.  The same high-level description
   does not necessarily communicate these different protocol-layer
   associations.

   Suppose a transport-associated AI agent needs to discover other AI
   agents exposing network-operation services within the organization's
   infrastructure.  Acting as the Discovering Entity, it initiates
   discovery within the known organizational Discovery Scope.  The
   discovery mechanism may return information about multiple AI
   resources that advertise related communication services.

   For initial discovery, the relevant question is whether the returned
   public information is sufficient to distinguish the advertised
   communication functions and their protocol-layer associations, and to
   provide the reachability information necessary for subsequent
   interaction.

4.  Problem Statement

   The DAWN requirements draft identifies entity type, offered services,
   reachability information, communication protocols, and relevant
   security parameters as components of Minimum Discoverable Information
   (MDI) [DAWN-REQ].

   In the layered communication architecture described in Section 3, AI
   agents may expose similar high-level service descriptions while
   supporting different communication functions and protocol-layer
   responsibilities.  For example, a general description such as traffic
   optimization may refer to an agent supporting transport adaptation,
   routing, or packet scheduling.  These agents may have different
   operational scopes even though their advertised services are related.

   The use case therefore raises the question of whether offered-service
   information, together with the Discovery Scope and other available
   discovery properties, is sufficient to distinguish the communication
   functions and protocol-layer responsibilities associated with
   discovered AI agents.

   *Problem Statement:* What minimum public discovery information is
   needed to distinguish and contact AI agents exposing similar
   communication-related services but associated with different
   protocol-layer or operational-domain responsibilities, within a known
   organizational Discovery Scope?



Jadoon, et al.             Expires 6 May 2027                   [Page 5]

Internet-Draft  AI Agent Discovery for Communication Ser   November 2026


   This problem concerns the information made available during initial
   discovery rather than detailed capability assessment or agent
   selection.  Depending on the discovery mechanism, the necessary
   distinctions might be expressed through offered-service descriptions,
   existing discovery properties, or additional discovery context.  This
   document does not presume that new mandatory MDI fields or a
   particular discovery architecture are required.

5.  Security Considerations

   This document describes an initial-discovery use case and does not
   specify a discovery protocol or security mechanism.

6.  IANA Considerations

   This document makes no requests of IANA.

7.  Informative References

   [DAWN-CHARTER]
              Internet Engineering Task Force, "Discovery of Agents With
              Names (DAWN) Working Group Charter", 22 October 2026,
              <https://datatracker.ietf.org/doc/charter-ietf-dawn/01/>.

   [DAWN-REQ] King, D. and A. Farrel, "Requirements for the Discovery of
              Agents With Names (DAWN)", Work in Progress, Internet-
              Draft, draft-king-dawn-requirements-02, 19 October 2026,
              <https://datatracker.ietf.org/doc/html/draft-king-dawn-
              requirements-02>.

   [DAWN-TERMS]
              Farrel, A., Yao, K., Schott, R., and N. Williams,
              "Terminology for the Discovery of Agents With Names
              (DAWN)", Work in Progress, Internet-Draft, draft-farrel-
              dawn-terminology-06, 28 October 2026,
              <https://datatracker.ietf.org/doc/html/draft-farrel-dawn-
              terminology-06>.

   [ETSI-RADAR]
              European Telecommunications Standards Institute (ETSI),
              "ETSI Technology Radar", n.d.,
              <https://www.etsi.org/technologies/technology-radar/>.









Jadoon, et al.             Expires 6 May 2027                   [Page 6]

Internet-Draft  AI Agent Discovery for Communication Ser   November 2026


   [LAYERED-AGENTS]
              Jadoon, M. A., Robitzsch, S., and C. J. Bernardos,
              "Agentic AI Architectural Principles for Autonomous
              Computer Networks", Work in Progress, Internet-Draft,
              draft-jadoon-nmrg-agentic-ai-autonomous-networks-00, 2
              March 2026, <https://datatracker.ietf.org/doc/html/draft-
              jadoon-nmrg-agentic-ai-autonomous-networks-00>.

   [RFC1122]  Braden, R., Ed., "Requirements for Internet Hosts -
              Communication Layers", STD 3, RFC 1122,
              DOI 10.17487/RFC1122, October 1989,
              <https://www.rfc-editor.org/rfc/rfc1122>.

   [RFC1123]  Braden, R., Ed., "Requirements for Internet Hosts -
              Application and Support", STD 3, RFC 1123,
              DOI 10.17487/RFC1123, October 1989,
              <https://www.rfc-editor.org/rfc/rfc1123>.

   [SANET]    Xiao, Y., Li, X., Zhou, H., Li, Y., Gao, Y., Shi, G.,
              Zhang, P., and M. Krunz, "SANet: A Semantic-Aware Agentic
              AI Networking Framework for Cross-Layer Optimization in
              6G", IEEE Transactions on Mobile Computing Vol. 25, 2026,
              DOI 10.1109/TMC.2026.3691804, 2026,
              <https://doi.org/10.1109/TMC.2026.3691804>.

   [XR-TRANSPORT]
              Huynh, V. S. H., Qian, P., Wang, N., Udora, C., and R.
              Tafazolli, "User-Intent Aware Transport-Layer Intelligence
              for Frame Synchronisation in Multi-Party XR Application",
              IEEE Gaming, Entertainment, and Media Conference 2024,
              DOI 10.1109/GEM61861.2024.10585795, 2024,
              <https://doi.org/10.1109/GEM61861.2024.10585795>.

Authors' Addresses

   Muhammad Awais Jadoon
   InterDigital Europe Ltd
   London
   United Kingdom
   Email: muhammad.awaisjadoon@interdigital.com


   Sebastian Robitzsch
   InterDigital Europe Ltd
   London
   United Kingdom
   Email: sebastian.robitzsch@interdigital.com




Jadoon, et al.             Expires 6 May 2027                   [Page 7]

Internet-Draft  AI Agent Discovery for Communication Ser   November 2026


   Carlos J. Bernardos
   Universidad Carlos III de Madrid
   Madrid
   Spain
   Email: cjbc@it.uc3m.es














































Jadoon, et al.             Expires 6 May 2027                   [Page 8]