One of my favorite stories in this package comes from my colleague David Rothman, who took A Closer Look at AI for Materials Research. AI could transform the process of discovering new materials—innovations that could be especially useful in the world of climate technology, where new batteries, semiconductors, magnets, and more are needed.
But the field has yet to prove that it can produce materials that are truly new and useful. Can AI really speed up materials research? What might this look like?
For researchers hoping to find new ways to power the world (or cure disease, or achieve any number of other big, important goals), a new material could change everything.
The problem is that inventing materials is difficult and slow. Just look at plastic: the first fully synthetic plastic was invented in 1907, but it wasn't until about the 1950s that companies began producing the wide variety we're familiar with today. (And, of course, while plastic is incredibly useful, it also causes its fair share of complications for society.)
Materials science has come to a bit of a standstill in recent decades—David has been covering the field for nearly 40 years, and in that time, he says, there have been only a few major commercial breakthroughs. (Lithium-ion batteries are one of them.)
Can AI change everything? The prospect is tempting, and companies are rushing to check it out.
Lila Sciences, based in Cambridge, Massachusetts, works to use artificial intelligence models to discover new materials. Not only can a company train an AI model using all the latest scientific literature, but it can also connect it to an automated laboratory so it can learn from experimental data. The goal is to speed up the iterative process of inventing and testing new materials and looking at research in ways that people might otherwise miss.
At an MIT Technology Review event earlier this year, I was able to listen to David's interview with Rafael Gomez-Bombarelli, one of the co-founders of Lila. In describing what the company is working on, Gomez-Bombarelli acknowledged that the discovery of materials for artificial intelligence is not yet a big breakthrough. More.
