Answers are grounded in this paper and cite their sources.
Technical specification demonstrates a tamper-evident audit record architecture for policy decisions, highlighting verifiable canonical logging in automated systems.
IETF individual Internet-Draft specifying the Governed Action Receipt (GAR): a compact, tamper-evident record binding actor, action, policy digest, decision outcome (closed vocabulary — UNKNOWN is never PASS), byte-level subject digests, and evidence references. Canonicalized with RFC 8785, carried in DSSE envelopes (Ed25519) or honestly unsigned under the *.unsigned.json naming convention, and linkable into append-only hash-chained logs with the tail-truncation limitation stated explicitly. The specification matches the reference implementation (szl-receipts 14.0.0, Apache-2.0) statement for statement; every worked example is computed by the library at build time and independently re-derived by the validator. idnits 2.17.1: 0 errors. Files: draft-lutar-governed-action-receipt-00.txt (rendered, 33pp), .md (kramdown source), build_draft.py, validate_draft.py.
No takes yet. Share an insight, caveat, or question.
Stephen Paul Jr. Lutar (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: