From Edge AI to Governed Autonomous Edge Intelligence

Abstract

Artificial intelligence is moving from advisory systems toward autonomous agents that can perceive environments, reason about goals, invoke tools, access services, delegate tasks, and act across digital and physical infrastructures. At the same time, computation is moving from centralized cloud environments toward distributed edge devices, gateways, local nodes, private infrastructure, and cloud services. These changes create an architectural problem that cannot be addressed by placing conventional security controls around an AI model. The article introduces SGAEIA and asks how autonomous systems can receive enough authority to perform useful actions while keeping that authority bounded, verifiable, auditable, and revocable.

This page provides the author-written abstract and bibliographic metadata for discovery. The Zenodo record is the persistent source for citation, files, version, and license information.