<![CDATA[The increasing use of artificial intelligence (AI) in higher education offers new opportunities for enhancing distance learning. However, limited research exists on how AI-supported environments can simultaneously promote academic achievement, learner autonomy, sustainability awareness, and values-based education, especially within Islamic higher education. This study addresses this gap by developing and evaluating an AI-supported adaptive navigation model that integrates personalized learning with Sustainable Development Goals (SDG)-oriented and values-based principles. Using a quasi-experimental design, 84 undergraduate students from an Islamic institution in Uzbekistan were assigned to either an experimental group (n=42) using the AI platform or a control group (n=42) receiving traditional online instruction. Data were collected via achievement tests, engagement questionnaires, system analytics, motivation scales, and technology acceptance measures, then analyzed through descriptive statistics, t-tests, regression, and ANCOVA. Results showed the experimental group achieved significantly higher learning outcomes (M=82.63) than the control (M=74.18). They also exhibited greater engagement, autonomy, and sustainability awareness, illustrating the positive impact of the AI approach. Students expressed favorable views on ethical technology use, inclusive education, and the social benefits aligned with SDG 4. Overall, AI-supported adaptive navigation enhances learning by combining personalized feedback, structured pathways, and values-driven experiences. Theoretically, this study contributes to the emerging concept of values-based AI in education by combining Islamic educational philosophy, the Community of Inquiry framework, and Self-Determination Theory, highlighting how intelligent environments support meaningful learning, empowerment, and ethics. Practically, it underscores the potential of ethical digital environments to foster sustainable, inclusive, value-driven higher education within Islamic contexts.]]>
