Google DeepMind and Lawrence Berkely National Laboratory revealed that they were able to discover 2.2 million new crystals using the AI system called GNoME.
GNoMe Finds Nearly 800 Years Worth of Knowledge
Using the Graph Networks for Materials Exploration (GNoME), researchers were able to make a breakthrough discovery of more than 2 million new crystals. It also included 380,000 stable materials that can be utilized to power future technologies.
"GNoME shows the potential of using AI to discover and develop new materials at scale," Google DeepMind shared. External researchers from all over the world were responsible for experimentally validating the 736 new structures.
The company also announced that the GNoME predictions are now available within the research community. Moreover, the 380,000 stable materials will be contributed to the Materials Project and will be added to the online database.
Using AI to Boost Progress in Materials Science
Two papers have been published under Nature detailing the new crystals in discovery and how the AI predictions can be used for autonomous material synthesis. Over the last decade, the previous computational approaches were only able to discover 28,000 new materials, a small number compared to the capability of GNoMe.
"By giving scientists the full catalog of the promising 'recipes' for new candidate materials, we hope this helps them to test and potentially make the best ones," Google DeepMind shared upon announcing the release of the database.
In addition, both Google DeepMind and the Berkeley Lab showed optimism about the potential of AI tools such as GNoME to revolutionize materials discovery and shape the future of the industry.
