PROBAITI Control Framework Specification - Version 1.0
PROBAITI Control Framework Specification — Version 1.0 defines the controlled semantic architecture for designing, classifying, governing, testing and maintaining controls within the PROBAITI evidence-first AI governance and assurance methodology. The specification establishes the canonical PROBAITI control object and the methodological requirements governing control identity, objectives, applicability, ownership, execution, dependencies, operational outputs, testing interfaces, exceptions, compensating controls, remediation, change management, revalidation, traceability and historical preservation. The framework distinguishes control design, implementation, operating effectiveness, outcome effectiveness, and change and continuity as separate assessment layers. A control is not considered effective merely because a policy, document or expected output exists; conclusions must remain bounded by reproducible evidence and testing. The specification also defines controlled approaches to mapping PROBAITI controls to external legal, regulatory, standards-based and organisational frameworks without implying automatic equivalence, compliance inheritance, certification or external endorsement. Conformance is governed through explicit quality gates covering control identity, objective, applicability, specification, ownership, dependencies, outputs, testing, exceptions, change, traceability and history. This Control Framework operates within the normative boundary established by the PROBAITI Core Methodology Specification — Version 1.0 and supports the wider PROBAITI publication ecosystem. It does not override the Core Methodology or independently establish legal compliance. Version 1.0 was approved as the PROBAITI canonical control-framework baseline on 14 September 2026. The framework has not yet completed independent industry validation. No claim is made that it constitutes certification, accreditation, regulatory approval, statutory conformity assessment, or endorsement by a regulator, standards body or certification body.
Authors
- Cem Yılmaz
Institutions
- Centro Universitário Curitiba (BR)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-09-16
- DOI
- https://doi.org/10.5281/zenodo.22793672
- Primary Topic
- Safety Systems Engineering in Autonomy
- Type
- article
- Field-Weighted Citation Impact
- 0.00