From Corrigibility to Answerability: A correction contract for scoped authority and verified consequences

Read full text
Authors: mFLO 111
Cited by: 8

Abstract

A system can accept an intervention without the intervention being justified, and can explain a justified correction without changing the affected outcome. This paper specifies a correction contract connecting four elements: the referent at issue, scoped authority, supporting evidence, and verified consequences. A fictional service-restoration workflow distinguishes acceptance, execution, observation, and closure. It handles standing delegation, changed mandates, partial repair, uncertain execution, duplicate requests, and reopening after new evidence. A contract worksheet and twelve authored transition scenarios make the proposed requirements inspectable; they are design materials, not implemented-system tests. The proposal complements technical corrigibility and overlaps with meaningful human control, normative interfaces, and established case-management models. Those existing resources can express the same distinctions. The proposed contribution is a correction-specific review procedure that makes the scope of a completion claim explicit across system boundaries. Its practical advantage over competent ordinary workflow design remains unmeasured. Authentication, legitimate jurisdiction, incentive compatibility, privacy enforcement, and reliable downstream observation remain separate implementation or normative burdens. Research status: Conceptual systems-design working paper; contract worksheet and twelve authored scenarios; no system or model experiment. Not peer reviewed. Contents: Paper in PDF, HTML, and Markdown, README, supplementary materials, license notices, and file manifest. Version 1.1 revises the unpublished version 1.0 draft. Licenses: All author-owned paper and accompanying research/design materials in this package use CC BY 4.0. No software is included. See LICENSE.md for scope. Related work: The conceptual antecedent is mFLO 111, Invariant Relation and Non-Substitution, version 1.1, https://doi.org/10.5281/zenodo.22805469.

Fields of Study & Subjects

Safety Systems Engineering in Autonomy Business Process Modeling and Analysis Access Control and Trust

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
17 September 2026
Document type
Preprint