In this study, we explore the role of image-generating AI in architectural design, focusing on sketch-to-image AI tools and their ability to interpret hand-drawn sketches. Five AI tools, namely, fabrie, LookX AI, PromeAI, mnml.ai, and Rerender AI, were analyzed for their comprehension and expressiveness in architectural contexts. To evaluate the effectiveness of these AI tools, architectural experts were surveyed to identify key architectural features present in sketches. These expert insights were then compared with AI-generated outputs to assess their accuracy in spatial perception, material expression, and spatial context awareness. Fabrie excelled in spatial interpretation and material representation, efficiently converting conceptual sketches into detailed visualizations. PromeAI demonstrated strong creative flexibility, supporting iterative design processes with diverse customization options. However, some contextual inconsistencies and missing environmental elements were noted. Importantly, in this study, we discuss how AI-generated imagery can be integrated with sensor-based feedback loops and material-aware simulation tools to enhance the design process. By combining AI visualization with sensor-informed environmental data and performance metrics, architects can more effectively evaluate spatial quality and environmental responsiveness in early-stage design. The study highlights AI’s potential as a complementary tool in early-stage design, enhancing rapid ideation, visualization, and design automation. By integrating AI into workflows, architects can expand creativity and efficiency while exploring a wider range of design possibilities. Furthermore, advancements in AI learning models, prompt engineering, and collaborative design processes particularly those that incorporate sensor-derived data are emphasized to strengthen AI’s role in bridging digital tools with human-driven creativity in architectural practice.