MKS.

Operational AI assurance · ODA3 Institute

GAISSF™ v1.0Global AI Safety and Security Framework

The umbrella control framework for translating AI governance requirements into operational and assessable practice.

Open the authoritative framework ↗

From institutional framework to operational practice.

My work on GAISSF™ focuses on the operational layer between high-level governance requirements and observable system behaviour: controls, runtime mechanisms, evidence, provenance, verification, findings, and bounded assurance.

GAISSF™ organizes AI safety and security requirements across nine domains and 59 controls. It provides the umbrella architecture for relating governance, technical security, human oversight, regulatory alignment, incident practices, and evidence-based assessment.

  1. 01Requirement
  2. 02Control
  3. 03Runtime mechanism
  4. 04Evidence and provenance
  5. 05Verification and bounded assurance