Prediction is not authorization.
Research findings remain separate from publication and delivery authorization.
Decision Reliability
GDR-SE
GDR-SE Runtime Enforcement v0.1-R1.1 determines whether a StructEvidence research result is sufficiently complete, current and traceable to be published, commercially delivered or used in monitoring.
GDR-SE Runtime Enforcement v0.1-R1.1Evaluation Mode: RUNTIME_EVALUATED
Abstention is valid.
When evidence is insufficient, stale or unstable, withholding a conclusion is a valid scientific output.
Counterfactual required.
Every authorization record states what would falsify the conclusion and what evidence would distinguish alternatives.
Runtime Outcomes
ALLOW_PUBLICATIONALLOW_WITH_LIMITATIONSREFRESH_REQUIREDHUMAN_REVIEW_REQUIREDABSTAINVETO
GDR-SE evaluates freshness and expiry against an explicit UTC evaluation timestamp. Authorization precedence and Gate effects are versioned configuration inputs whose hashes are recorded with each evaluation.
GDR-SE produces categorical release decisions, not scores, ratings or rankings.
G3 Evidence Freshness is evaluated using RDL Freshness Policy v0.1a where a configured profile exists. EVENT_INVALIDATED, UNDER_REVIEW and critical unconfigured states set explicit fail-safe facts for GDR-SE aggregation.