The advancement of Artificial Intelligence (AI) presents new challenges in risk management, requiring structured frameworks that effectively integrate ethical, technical, and engineering aspects. Existing ontology-based frameworks often lack sufficient specificity to comprehensively represent multi-faceted AI risks, thereby limiting their real-world applicability. This research presents a robust, ontology-driven AI risk management framework that systematically incorporates ethical principles such as fairness, transparency, and accountability. Utilizing Boxology diagrams for structured visual representations, this framework clarifies the relationships among AI system components, risks, impacts, and consequences. These diagrams inform the development of a semantic, RDF-based ontology explicitly linked to requirements engineering practices, facilitating proactive risk identification and mitigation throughout the AI system development lifecycle. By leveraging semantic web technologies, the framework supports automated querying, compliance tracking, and cross-domain governance, thereby improving regulatory alignment, decision-making, and stakeholder trust. Ultimately, this research contributes to establishing an integrated, scalable approach to AI risk governance, advancing ethical innovation and sustainable technology development.