ZTEK: A Zero Trust Epistemic Kernel for AI-Assisted Research Workflows

Fuente: Zenodo
Salvato in:
Dettagli Bibliografici
Autore principale: Papanokechi, Papanokechi
Natura: Recurso digital
Lingua:inglese
Pubblicazione: Zenodo 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866901779219742720
author Papanokechi, Papanokechi
author_facet Papanokechi, Papanokechi
contents <p>This paper introduces <strong>ZTEK: a Zero Trust Epistemic Kernel</strong>, the enforcement layer of the AI Epistemic Governance Stack. ZTEK provides an architectural boundary between AI‑generated output and the verified knowledge base, ensuring that no claim enters organizational workflows without reproducible provenance and independent verification. The protocol is grounded in five documented failure modes from a multi‑year AI‑assisted mathematical research program and is organized into five pillars: assume fabrication, require provenance, verify at the boundary, enforce least‑confidence assertion, and re‑verify on context change. The paper includes a decision gate, trust‑tier system, empirical validation, and implementation artifacts such as a provenance JSON schema, reproduce commands, and audit‑log examples.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_19591564
institution Zenodo
language eng
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle ZTEK: A Zero Trust Epistemic Kernel for AI-Assisted Research Workflows
Papanokechi, Papanokechi
AI governance, epistemic reliability, verification, Zero Trust, provenance, research workflows, ZTEK, AEAL,
<p>This paper introduces <strong>ZTEK: a Zero Trust Epistemic Kernel</strong>, the enforcement layer of the AI Epistemic Governance Stack. ZTEK provides an architectural boundary between AI‑generated output and the verified knowledge base, ensuring that no claim enters organizational workflows without reproducible provenance and independent verification. The protocol is grounded in five documented failure modes from a multi‑year AI‑assisted mathematical research program and is organized into five pillars: assume fabrication, require provenance, verify at the boundary, enforce least‑confidence assertion, and re‑verify on context change. The paper includes a decision gate, trust‑tier system, empirical validation, and implementation artifacts such as a provenance JSON schema, reproduce commands, and audit‑log examples.</p>
title ZTEK: A Zero Trust Epistemic Kernel for AI-Assisted Research Workflows
topic AI governance, epistemic reliability, verification, Zero Trust, provenance, research workflows, ZTEK, AEAL,
url https://doi.org/10.5281/zenodo.19591564