Enregistré dans:
| Auteur principal: | |
|---|---|
| Format: | Recurso digital |
| Langue: | anglais |
| Publié: |
Zenodo
2026
|
| Accès en ligne: | https://doi.org/10.5281/zenodo.19357683 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866901113344622592 |
|---|---|
| author | Truong, Narnaiezzsshaa |
| author_facet | Truong, Narnaiezzsshaa |
| contents | <p>This record documents Phase 17 of APR‑Lite, a cryptographic governance layer designed to preserve human decision authority over AI‑influenced outputs in regulated environments. Phase 17 introduces topology retention: signed views of a tenant’s governance‑relevant state captured at specific governance moments, allowing the system to answer questions about past configuration using direct attestations rather than reconstructed history.</p> <p>Topology views are limited to the aspects of state on which governance decisions operate and are captured at a small number of constitutionally significant triggers. These include automated capture around lifecycle transitions that alter operational structure, automated capture around actions that suspend activity, and manual capture initiated by authorized governance personnel when needed.</p> <p>The authority model preserves the distinction between actors and historians established in prior phases. Tenant‑bound signatures continue to record acts, while a distinguished authority key attests to the corresponding topology views, maintaining a separation between the entities that exercise authority and the entity that notarizes the evolution of state over time.</p> <p>Attestations produced in this phase are verifiable through the existing proof interfaces, and the history‑verification surface is extended to report on the presence of these topology views alongside existing integrity indicators. Phase 17 achieved full success on its smoke‑test suite for the extended topology‑retention pipeline in worker version 8.2.0‑p17.0. A more detailed treatment of the topology model, trigger semantics, and verification behavior is available under restricted access to qualified requesters.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_19357683 |
| institution | Zenodo |
| language | eng |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Topology Retention for Human-AI Governance Substrates: GLOBAL-Signed Constitutional Surface Snapshots at Governance Moments | APR-Lite Phase 17 — Technical Note Truong, Narnaiezzsshaa <p>This record documents Phase 17 of APR‑Lite, a cryptographic governance layer designed to preserve human decision authority over AI‑influenced outputs in regulated environments. Phase 17 introduces topology retention: signed views of a tenant’s governance‑relevant state captured at specific governance moments, allowing the system to answer questions about past configuration using direct attestations rather than reconstructed history.</p> <p>Topology views are limited to the aspects of state on which governance decisions operate and are captured at a small number of constitutionally significant triggers. These include automated capture around lifecycle transitions that alter operational structure, automated capture around actions that suspend activity, and manual capture initiated by authorized governance personnel when needed.</p> <p>The authority model preserves the distinction between actors and historians established in prior phases. Tenant‑bound signatures continue to record acts, while a distinguished authority key attests to the corresponding topology views, maintaining a separation between the entities that exercise authority and the entity that notarizes the evolution of state over time.</p> <p>Attestations produced in this phase are verifiable through the existing proof interfaces, and the history‑verification surface is extended to report on the presence of these topology views alongside existing integrity indicators. Phase 17 achieved full success on its smoke‑test suite for the extended topology‑retention pipeline in worker version 8.2.0‑p17.0. A more detailed treatment of the topology model, trigger semantics, and verification behavior is available under restricted access to qualified requesters.</p> |
| title | Topology Retention for Human-AI Governance Substrates: GLOBAL-Signed Constitutional Surface Snapshots at Governance Moments | APR-Lite Phase 17 — Technical Note |
| url | https://doi.org/10.5281/zenodo.19357683 |