Google leveraged its artificial intelligence laboratory, known as DeepMind, to employ the emerging technology for predicting the molecular structure of over a million new materials. This breakthrough is believed to have wide-ranging benefits in enhancing technology across various real-world domains, such as renewable energy and computing.
This infiltration results in a tenfold increase in the number of known stable materials. These materials still require manufacturing and testing, which could take a long time, ranging from months to even years.
The recent progress is expected to accelerate the discovery of new materials needed for purposes like energy storage, solar panel utilization, and the development of high-performance semiconductor chips.
The artificial intelligence lab announced in a research paper published in the scientific journal Nature that in the near future, approximately 400,000 virtual designs of materials could be created under lab conditions.
Possible applications of this research include the production of batteries, solar panels, and computer chips with enhanced performance. Discovering and synthesizing new materials can be costly and time-consuming.
It took about twenty years of research for the lithium-ion batteries currently used in various commercial tech devices like smartphones, laptops, and electric cars to become operational.
Ekin Dogus Kubuk, a researcher at DeepMind, stated, “We aspire that significant improvements in experiments, independent homogenization, and machine learning models will greatly reduce the time gap extending between 10 years and 20 years to be more manageable.”
Google tested artificial intelligence on data from the Materials Project, a global research group established at the Lawrence Berkeley National Laboratory in 2011, including current research on around 50,000 known materials.
Google confirmed that it is now sharing its data with the research community with the aim of accelerating further progress in materials discovery.
Google emphasized that the industry tends to somewhat avoid risks when it comes to cost increases, and new materials often take some time before becoming cost-effective. It would be a real achievement if we could reduce that time period.
Google is now focusing on shifting its attention to predicting the ability to manufacture these new materials in the lab after using smart forecasting to estimate the stability of these materials.
