GNoME (Graph Networks for Materials Exploration) is an AI tool developed by Google DeepMind to help scientists discover new materials faster and more efficiently. It focuses on finding new inorganic crystals, which are important for technologies like batteries, solar panels, and computer chips.

Using machine learning, GNoME has predicted the structures of over two million new materials, of which about 380,000 are considered very stable. These stable materials could be used for future breakthroughs in superconductors (materials that can conduct electricity without resistance), advanced batteries, and more efficient electric vehicles.

Instead of the slow trial-and-error methods usually used in labs, GNoME analyzes how atoms connect in a material to predict which combinations will be stable. This AI-driven process can rapidly speed up discovery, saving years of time and effort.