DSES: Decision-Sequence Evidence Schema
Abstract
Open specification for reconstructable human-AI decision evidence. The v0.1 sequence layer defines event, exposure, information-state, capture-level, and integrity vocabularies; an I2 single export establishes internal consistency only unless compared with a separately held checkpoint or supported by verified external anchoring. The v0.2.0-rc13 outcome layer adds denominator-closed eligibility manifests, outcome linkage, pre-consensus adjudication under declared charters, content-addressed executable rules, recomputable reliance metrics (RAIR, RSR, EAR), and governance constraints on unit of analysis and individual-level derivation. The shipped candidate supplies RFC 8785 commitments, signatures, RFC 9162 proofs, reader-supplied external anchor trust, release lint, an adversarial suite, and a reference verifier, with unimplemented and attestation-only claims named explicitly. Specification text is licensed CC BY 4.0 (LICENSE-SPEC.md); reference code is licensed MIT (LICENSE-CODE.md).
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 14 September 2026