The rapid development and widespread adoption of artificial intelligence technologies necessitate their alignment with ethical principles throughout design and deployment processes. This study comparatively examines five major international reports on AI ethics published by UNESCO, the European Commission, the OECD, the European Parliament, and the Council of Europe. Employing document analysis and thematic coding methods, the research identifies transparency, responsibility, and trustworthiness as core ethical principles commonly emphasized across these frameworks. However, the findings reveal a fundamental pattern: while organizations demonstrate normative convergence around shared ethical values, they exhibit operational divergence in how these principles are framed, justified, and institutionalized within distinct governance contexts. This produces an ethical implementation gap, the distance between widely endorsed norms and their practical realization in policy frameworks. UNESCO pursues culturally adaptive norms through education and capacity building; the European Commission and European Parliament enforce compliance through binding regulations; OECD coordinates flexible implementation across member states; and the Council of Europe anchors AI ethics within human rights law. These divergent governance logics reflect deeper philosophical disagreements about whether ethical AI should be realized through voluntary norms, technical standards, binding legislation, or rights-based frameworks. The study demonstrates that contemporary challenges in AI ethics stem not from a lack of shared principles but from incompatible mechanisms for their operationalization. By analytically unpacking this implementation gap, the research advances AI ethics scholarship beyond principle identification and provides actionable insights for policymakers and practitioners.