I-D Action: draft-sharif-ai-model-lifecycle-attestation-01.txt

Internet-Draft draft-sharif-ai-model-lifecycle-attestation-01.txt is now
available.

   Title:   Cryptographic Attestation for AI Model Lifecycle: From Training Data to Inference Output
   Author:  Raza Sharif
   Name:    draft-sharif-ai-model-lifecycle-attestation-01.txt
   Pages:   14
   Dates:   2026-08-27

Abstract:

   This document defines a cryptographic attestation framework for
   the complete lifecycle of artificial intelligence models, from
   training data provenance through model weight signing,
   quantization verification, deployment attestation, and per-
   inference output signing.  The framework creates an unbroken
   chain of cryptographic evidence binding each inference output to
   the specific model version, training data, and deployment
   configuration that produced it.

   The framework uses ECDSA P-256 digital signatures, SHA-256
   hash functions, Merkle trees for corpus attestation, and JSON
   Web Key Sets (JWKS) for key discovery.  It addresses documented
   threats including model distillation attacks, quantization
   poisoning, training data manipulation, silent model degradation,
   and inference output tampering.

   This specification complements the Agent Trust Transport Protocol
   (ATTP) [draft-sharif-attp-agent-trust-transport], MCPS message
   signing [draft-sharif-mcps-secure-mcp], and the Agent Audit
   Trail format [draft-sharif-agent-audit-trail] to provide end-to-
   end cryptographic verification from data ingestion to consumer
   delivery.

The IETF datatracker status page for this Internet-Draft is:
https://datatracker.ietf.org/doc/draft-sharif-ai-model-lifecycle-attestation/

There is also an HTMLized version available at:
https://datatracker.ietf.org/doc/html/draft-sharif-ai-model-lifecycle-attestation-01

A diff from the previous version is available at:
https://author-tools.ietf.org/iddiff?url2=draft-sharif-ai-model-lifecycle-attestation-01

Internet-Drafts are also available by rsync at:
rsync.ietf.org::internet-drafts


_______________________________________________
I-D-Announce mailing list -- [email protected]
To unsubscribe send an email to [email protected]