Imagine using artificial intelligence to compare two seemingly unrelated works – biological tissue and “9. Beethoven’s Symphony. At first glance, a living system and a musical masterpiece may seem unrelated. But a novel artificial intelligence method developed by Markus J. Buehler, the McAfee Professor of Engineering and professor of civil and environmental engineering and mechanical engineering at MIT, fills this gap by uncovering common patterns of complexity and order.
“By combining generative AI with graph-based computational tools, this approach reveals completely new ideas, concepts and designs that were previously unimaginable. We can accelerate scientific discovery by teaching generative AI to make novel predictions about never-before-seen ideas, concepts and designs,” says Buehler.
Recent open access research published in demonstrates an advanced artificial intelligence method that integrates generative knowledge acquisition, graph-based representation, and multimodal knowledgeable graph reasoning.
The work uses charts developed using methods inspired by category theory as a central mechanism for teaching the model to understand symbolic relationships in science. Category theory, a branch of mathematics that deals with abstract structures and the relationships between them, provides a framework for understanding and unifying diverse systems by focusing on objects and their interactions rather than their specific content. In category theory, systems are viewed in terms of objects (which can be anything from numbers to more abstract entities such as structures or processes) and morphisms (arrows or functions that define the relationships between these objects). Using this approach, Buehler was able to teach an AI model to systematically reason about sophisticated scientific concepts and behaviors. The symbolic relationships introduced through morphisms make it clear that AI does not just draw analogies, but engages in deeper reasoning that maps abstract structures across domains.
Buehler used this modern method to analyze a set of 1,000 research articles on biological materials and transformed them into a knowledge map in the form of a graph. The chart showed how different pieces of information were related to each other and allowed you to find groups of related ideas and key points that connected multiple concepts.
“What’s really interesting is that the graph is scale-free, highly connected, and can be effectively used to make graph inferences,” says Buehler. “In other words, we teach AI systems to think about graph-based data to help them build better models to represent the world and increase their ability to think and discover new ideas to enable discovery.”
Scientists can operate this framework to answer sophisticated questions, find gaps in current knowledge, propose modern material designs and predict how materials might behave, and connect concepts that have never been linked before.
The artificial intelligence model discovered unexpected similarities between biological materials and ‘9. Symphony,” which suggests that there are patterns of complexity in both cases. “Just as cells in a biological material interact with each other in a complex but organized way to fulfill their function, Beethoven’s Ninth Symphony arranges notes and motifs to create a complex but coherent musical experience,” says Buehler.
In another experiment, a graph AI model recommended the creation of modern biological material inspired by the abstract patterns in Wassily Kandinsky’s painting “Composition VII.” Artificial intelligence suggested a modern composite material based on mycelium. “The result of this material combines an innovative set of concepts encompassing the balance of chaos and order, tunable properties, porosity, mechanical strength and complex chemical functionality,” notes Buehler. Drawing inspiration from abstract painting, artificial intelligence created a material that combines durability and functionality, while being adaptable and capable of performing various roles. Applications could lead to the development of pioneering, sustainable building materials, biodegradable alternatives to plastics, wearable technologies and even biomedical devices.
With this advanced AI model, scientists can draw on knowledge from music, art and technology to analyze data from these fields to identify hidden patterns that can open a world of pioneering possibilities in material design, research and even music and visual arts.
“Graph-based generative AI achieves a much higher degree of novelty, allows exploration of capabilities and technical details than conventional approaches, and establishes a broadly useful framework for innovation by revealing hidden connections,” says Buehler. “This study not only contributes to the field of bio-inspired materials and mechanics, but also sets the stage for a future where interdisciplinary research based on artificial intelligence and knowledge graphs can become a tool for scientific and philosophical inquiry as we look to other future work.” ”