This research introduces a method leveraging generative AI and graph theory to accelerate scientific discovery in bioinspired materials. It transforms scientific papers into knowledge graphs, revealing hidden connections between concepts and enabling AI to propose innovative material designs. The study demonstrates how these graphs facilitate question answering, hypothesis generation, and knowledge transfer across disciplines, even connecting art and science through isomorphism analysis. New knowledge is dynamically incorporated into graphs through agentic modeling and literature integration, further enhancing discovery. The methodology integrates multimodal inputs like artistic images to guide material design, culminating in AI-generated microstructures with detailed manufacturing processes. This work demonstrates a powerful human-AI collaborative workflow for exploring complex interactions and generating novel insights, showcasing the potential to revolutionize research across various fields.