Rapid advances in digital technologies and the expansion of artificial intelligence (AI) have led to the emergence of new approaches in architecture and urban environmental management. In this context, data-driven architecture, as an emerging paradigm, has significant potential to optimize resource consumption, enhance energy efficiency, and strengthen environmental resilience. This study aims to develop a conceptual framework for data-driven architecture in the era of AI with a focus on environmental sustainability. The research is qualitative and employs a systematic literature review and thematic analysis approach. Data were collected from credible national and international sources in the fields of smart architecture, AI, the Internet of Things, digital twins, and environmental sustainability. The analysis followed Braun and Clarke’s thematic analysis method, resulting in more than 150 initial codes organized into coherent themes. The findings reveal that data-driven architecture in the AI era is structured around four main themes: (1) data-driven infrastructure and architectural digitalization, (2) intelligent analysis and adaptive decision-making, (3) sustainability and environmental resilience, and (4) governance, interaction, and implementation context. The results further indicate that this approach operates through a dynamic cycle comprising real-time data collection, intelligent analysis, adaptive decision-making, and continuous learning. The proposed conceptual framework suggests that integrating data-driven technologies and AI with sustainability principles can facilitate the development of smart, sustainable, and resilient buildings and cities. This study contributes to the theoretical advancement of data-driven architecture and provides a foundation for future policymaking and the design of intelligent urban environments.