DeepMind, Google’s main artificial intelligence group, claims to have discovered “800 years’ worth of knowledge” after identifying 2.2 million new crystals. DeepMind stated that the materials discovered during the research might be utilized to alter industries while also opening up entirely new pathways for future discoveries. Approximately 380,000 of the 2.2 million crystals are apparently stable enough for creating next-generation technologies ranging from better electric car batteries to superconductors for ultra-efficient computers. DeepMind created a cutting-edge neural network technology called GNoME (Graph Networks for Materials Exploration) to find the crystals.
External researchers independently created 736 of the novel materials revealed by GNoME, putting DeepMind’s breakthrough to the test
Amil Merchant and Ekin Dogus Cubuk of DeepMind noted in a blog post that adopting GNoME saved them centuries of “painstaking experimentation” that would have been required to uncover the new materials. “With GNoME, we’ve multiplied the number of technologically viable materials known to humanity,” the team said in a statement. “GNoME demonstrates the promise of employing artificial intelligence to find and develop novel materials at scale… We expect that GNoME and other AI tools can assist revolutionize materials discovery today while also shaping the field’s future.” External researchers independently created 736 of the novel materials revealed by GNoME, putting DeepMind’s breakthrough to the test.
“Among these candidates are materials that have the potential to develop future transformative technologies ranging from superconductors, powering supercomputers, and next-generation batteries to boost the efficiency of electric vehicles,” the blog post stated. The research was detailed in a study, titled ‘Scaling deep learning for materials discovery, published in the journal Nature. The researchers behind the new tool said it can “reach unprecedented levels of generalisation, improving the efficiency of materials discovery by an order of magnitude”.
Others who were not participating in the study referred to GNoME as the “ChatGPT for chemistry,” alluding to the immensely popular artificial intelligence chatbot released one year ago. “Scientific discovery is the next frontier for AI,” said Carla Gomes, co-director of the Cornell University AI for Science Institute and a non-participant in the study. “That’s why I find this so exciting.”
