Software architecture is a knowledge-intensive and decision-centric discipline in which architects continuously reason about trade-offs, quality attributes, and long-term system evolution. Architectural knowledge is documented through heterogeneous artifacts such as architectural decision records (ADRs), design documents, issue discussions, and code-level dependencies. Despite the increasing availability of such data, architectural activities largely remain manual, experience-driven difficult to empirically validate. Recent advances in artificial intelligence (AI), natural language processing (NLP), machine learning (ML), computational intelligence, and data-driven software engineering enable new opportunities to support architectural reasoning in a systematic and evidence-based manner. These techniques can assist in extracting architectural knowledge from unstructured text, identifying architecture smells and technical debt, analyzing architectural evolution, and supporting architectural decision-making through data-driven recommendations. This workshop focuses on intelligent and data-driven approaches for software architecture, emphasizing validated and explainable methods that augment architects’ capabilities rather than replace human judgment. The workshop aims to bring together researchers and practitioners from software architecture, empirical software engineering, and AI-driven SE to discuss methodological foundations, empirical evidence, limitations, and open research challenges. The outcome will be a shared research agenda for responsibly integrating intelligent methodologies into software architecture research and practice.