From Intent to Execution Grant: An Execution-Boundary Conformance Profile for High-Risk AI Actions

RESUMO

AI agents increasingly propose actions with external consequences: financial transfers, infrastructure changes, deployments, disclosures, physical actuation. We specify EBL-Core, an execution-boundary conformance profile deciding whether a candidate AI-generated action may receive action-scoped execution authority under explicit conditions, via a verifiable Decision Derivation and Execution Release Contract. Across 100 trials, 32 concurrent Redemption attempts yielded exactly one successful Redemption per trial, and 100 Revoke-Redeem races ended in valid terminal outcomes.