CPAG Lab is CIAGI’s applied research and validation laboratory for demonstrating how AI governance failures can create physical consequences in OT environments, and how AAIGF-E aligned controls can detect, challenge, and stop unsafe actions.
Prompt injection through operational data feeds provides the research basis for demonstrating how trusted operational context can be manipulated.
Governance controls provide the validation, oversight, escalation, and action inhibition mechanisms demonstrated by the lab.
AI Consequence Propagation explains how degradation and authority drift can move from AI output to operational consequence.
A tank is physically full, but the AI layer receives misleading information indicating that the tank is empty. The system recommends filling the tank, creating an overfill risk.
Physical proof of AI failure and control response at cybersecurity, OT, and AI governance events.
Show organizations what AAIGF-E protects against before a formal assessment or pilot.
Hands on learning for students, practitioners, webinars, workshops, and university programs.
A concrete tool for regulators, standards bodies, and policy makers considering sector specific AI controls.
Edge computing and controlled scenario orchestration.
Physical signal capture, validation, actuation, and inhibition.
Water tank, pump, flow, and level scenarios with visible outcomes.
Failure injection, assurance checks, human oversight, and control evidence.
Discuss conference demonstrations, pilot validation, educational collaboration, academic research, or policy engagement.