sources · source-BATCH-2026-005-003

AI Agent Authentication and Authorization

Accepted discovery candidate candidate-BATCH-2026-001-009; contributes standards document; Internet-Draft evidence or normative context to governed AI-agent identity.

Open the canonical original source

Source at a glance

Source type: working paper
Publisher: Internet Engineering Task Force
Source or access date: 2026-06-01
Analysis depth: deeply analyzed
Rights treatment: link and paraphrase
Human review: Murray Newlands · 2026-08-16

What the source reports or argues

The draft models an agent as a workload that requires a stable identifier and cryptographically bound credentials.
It proposes preserving user or system delegation context in authorization decisions and audit trails.

Important limitations

Revision 02 is an expiring work in progress.
No interoperability tests or deployment outcomes are reported.

Source-located statements

6 reviewed statements are indexed from this source.

Revision 02 treats an AI agent as a workload that must authenticate to the tools, services, models, and other systems with which it interacts. (Revision 02 HTML, Section 3 "Agents are workloads", paragraphs 1-3)
The draft requires each participating agent to receive one WIMSE identifier. (Revision 02 HTML, Section 5 "Agent Identifier", normative paragraph)
Agent credentials should expire quickly, must state an expiration time, and must be cryptographically tied to the agent identifier. (Revision 02 HTML, Section 6 "Agent Credentials", paragraphs 1-3)
The proposed Agent Identity Management System separates identity, credential provisioning, authentication, authorization, policy, lifecycle events, and observability into a common control model. (Revision 02 HTML, Section 4 "Agent Identity Management System", Figure 2 and component definitions)
Authorization and audit records should retain the human or system delegation context when an agent acts for another principal. (Revision 02 HTML, Section 3 "Agents are workloads", paragraph after Figure 1)
The draft assembles existing identity and authorization specifications into an agent framework instead of claiming a wholly new protocol family. (Revision 02 HTML, Section 13 "Security Considerations", opening paragraph)

Evidence lineage and transparency

Relationship field	Linked identifiers
author ids	person-BATCH-2026-005-pieter-kasselman, person-BATCH-2026-005-jeff-lombardo, person-BATCH-2026-005-yaron-rosomakho, person-BATCH-2026-005-brian-campbell, person-BATCH-2026-005-nick-steele, person-BATCH-2026-005-aaron-parecki

Machine review: machine checked. Human review: approved. Workflow: published.

Complete structured record

source id: source-BATCH-2026-005-003
canonical title: AI Agent Authentication and Authorization
alternate titles
source type: working_paper
series or parent source: draft-klrc-aiagent-auth-02
publisher: Internet Engineering Task Force
channel: Unknown
speaker ids
author ids: person-BATCH-2026-005-pieter-kasselman, person-BATCH-2026-005-jeff-lombardo, person-BATCH-2026-005-yaron-rosomakho, person-BATCH-2026-005-brian-campbell, person-BATCH-2026-005-nick-steele, person-BATCH-2026-005-aaron-parecki
institutional author: IETF Internet-Draft authors
organization references
recorded at: Unknown
event date: Unknown
published at: 2026-06-01
updated at: Unknown
duration seconds: Unknown
language: lang-en
translated title: Unknown
translation method: Unknown
geography of speaker
geography of organization
geography discussed: global standards context
study geography: global standards context
original url: https://www.ietf.org/archive/id/draft-klrc-aiagent-auth-02.html
canonical url: https://www.ietf.org/archive/id/draft-klrc-aiagent-auth-02.html
archived url: Unknown
embed url: Unknown
doi: Unknown
canonical identity status: verified
repost status: original_or_authoritative_rendition
original source id: Unknown
rights status: link_and_paraphrase
ownership status: third_party
relationship to off: none_identified; Executive AI Research snapshot contains no matching production records
transcript status: Unknown
transcript source: Unknown
transcript republication permission: Unknown
chapter markers
analysis basis: Complete official IETF HTML rendition reviewed with stable section and paragraph locators.
topics: topic-ai-agents, topic-ai-governance, topic-cybersecurity, topic-non-human-identity
executive roles: role-ciso, role-cio, role-cto, role-general-counsel, role-board-director
original abstract: Unknown
inclusion rationale: Accepted discovery candidate candidate-BATCH-2026-001-009; contributes standards document; Internet-Draft evidence or normative context to governed AI-agent identity.
source quality dimensions: {"attribution_strength":"high","methodological_transparency":"not_applicable_non_empirical","independence":"multi_vendor_author_group","bibliographic_stability":"medium_due_to_expiring_draft","evidence_strength":"strong_for_current_proposal_weak_for_interoperability_outcomes","unresolved":"Future revisions may change normative language or identifier choices."}
methodology quality: {"design":"non-empirical technical standards proposal","transparency":"not_applicable_non_empirical","human_review_required":true}
study design: non-empirical technical standards proposal
sample: {"size":"not applicable","sampling_method":"not applicable"}
population: AI-agent workloads and the tools, services, models, systems, and users that interact with them
date range: {"fieldwork":"not applicable","version":"revision 02; expires 2026-12-03"}
funding: not reported
sponsor: not reported
peer review status: working document; not an RFC and not peer reviewed
findings: The draft models an agent as a workload that requires a stable identifier and cryptographically bound credentials., It proposes preserving user or system delegation context in authorization decisions and audit trails.
limitations: Revision 02 is an expiring work in progress., No interoperability tests or deployment outcomes are reported.
correction ids
retraction status: active_work_in_progress_not_retracted_as_of_2026-08-15
content hash: e34f335c56546a6d91d88d43159c9f80d4dcfbbad020d41c31607fc56848befa
accessed at: 2026-08-15
verification status: metadata_content_and_version_machine_verified_human_approved
source depth: deeply_analyzed
publication status: published
workflow status: published
machine review status: machine_checked
human review status: approved
reviewed by: Murray Newlands
reviewed at: 2026-08-16T23:17:22Z

Provenance and revision history

{
  "provenance": [
    {
      "source_url": "https://www.ietf.org/archive/id/draft-klrc-aiagent-auth-02.html",
      "accessed_at": "2026-08-15",
      "retrieval_method": "official source retrieval and machine-assisted review",
      "exact_locator": "revision 02; expires 2026-12-03",
      "content_hash": "e34f335c56546a6d91d88d43159c9f80d4dcfbbad020d41c31607fc56848befa",
      "batch_id": "BATCH-2026-005",
      "prompt_id": "OEII-EVIDENCE-RESEARCH",
      "prompt_version": "2.0",
      "notes": "Human review pending."
    }
  ],
  "revision_history": [
    {
      "changed_at": "2026-08-16T23:17:22Z",
      "changed_by": "Murray Newlands",
      "summary": "Approved for the governed-identities pilot release under the exact scope, exclusions, rights treatment, and limitations recorded in issue #18.",
      "batch_id": "BATCH-2026-005"
    }
  ]
}

Open machine-readable record