Artificial Intelligence (AI) systems demand comprehensive governance frameworks that ensure ethical, transparent, and accountable practices across their entire lifecycle. This paper presents the TAFES framework—comprising Transparency, Accountability, Fairness, Ethics, and Safety—as a theoretically grounded and practically implementable approach for responsible AI. Through systematic analysis of existing global frameworks and an applied healthcare use case (Nurse Linda), TAFES demonstrates how ethical principles can be operationalized through lifecycle processes spanning design, development, deployment, and decommissioning. Unlike regulatory or risk-centric models such as the NIST AI RMF or EU AI Act, TAFES integrates moral philosophy with engineering implementation to bridge the gap between ethical intent and practical execution.