Crystal Predictions: How AI is Changing the Game

AI is stepping into the world of crystal prediction. It's not just about guessing if a crystal can be made. It's also about figuring out how to make it and what ingredients are needed. This is a big deal because it could change how we create new materials. A new tool called CSLLM is leading the way. It uses AI to predict if a crystal can be made, how to make it, and what starting materials are best. The AI was trained on a lot of data, including crystals that can and can't be made. It's really good at its job, with an accuracy of 98. 6% for predicting if a crystal can be made. That's way better than older methods. But it's not just about accuracy. The AI can also handle complex crystals that it wasn't trained on. It can predict the best way to make common crystals and even identify the best starting materials. This is a big step forward in materials science. With this AI, scientists can now look at thousands of theoretical crystal structures and figure out which ones can be made. They can also predict important properties of these crystals. This could lead to new materials for all sorts of applications.