A team of scientists at Northwestern University has used machine learning to discover a new version of an alloy called metallic glass, reports Northwestern Now. Metallic glass is lighter and stronger than steel and is more resistant to wear and corrosion, which is why it’s often used as a protective coating and alternative to steel. Like any alloy, it is created by rearranging the atoms of various metals in new ways. The problem is, over the last 50 years only a few thousand of the millions of possible combinations of the ingredients used to make metallic glass have been studied. That is, until researchers turned to AI.
Using artificial intelligence, the scientists at Northwestern University have already been able to discover three new blends of ingredients that form metallic glass–and AI does it 200 times faster than the scientists themselves could have done. The results mean that new alloys could reach the market decades before they normally would thanks to AI. This, in turn, could have a massive impact on the industrial and structural design of objects large and small, from smartphones to skyscrapers.