Constitutional Replay-Authority: A Category-Defining Paper for Computational Legitimacy and Replay Verification
Larry Otto · Zenodo (CERN European Organization for Nuclear Research) · 2026
Constitutional Replay-Authority is proposed as a new infrastructure category dedicated to computational legitimacy and replay verification. The paper examines a growing challenge in artificial intelligence and automated decision systems: determining whether, long after a consequential decision has been made, it remains possible to establish what information was available, what policies governed the decision, what decision resulted, and whether those conditions can be authoritatively verified. The paper argues that existing categories such as logging, observability, explainability, provenance, governance, and immutable record systems address related concerns but were not specifically created to solve this challenge in its entirety. Constitutional Replay-Authority is introduced as a distinct category focused on preserving the relationship between computational decisions and the information, policies, and conditions under which those decisions were made, thereby enabling future verification of computational legitimacy. The concepts discussed herein are associated with infrastructure innovations described in U.S. Patent Pending Application No. 64/078,167, filed May 29, 2026.