February 15, 2019
MIT Technology Review | By David Rotman
The biggest impact of artificial intelligence will be to help humans make discoveries we couldn’t make on our own.
Regina Barzilay’s office at MIT affords a clear view of the Novartis Institutes for Biomedical Research. Amgen’s drug discovery group is a few blocks beyond that. Until recently, Barzilay, one of the world’s leading researchers in artificial intelligence, hadn’t given much thought to these nearby buildings full of chemists and biologists. But as AI and machine learning began to perform ever more impressive feats in image recognition and language comprehension, she began to wonder: could it also transform the task of finding new drugs?
The problem is that human researchers can explore only a tiny slice of what is possible. It’s estimated that there are as many as 1060 potentially drug-like molecules—more than the number of atoms in the solar system. But traversing seemingly unlimited possibilities is what machine learning is good at. Trained on large databases of existing molecules and their properties, the programs can explore all possible related molecules.
Drug discovery is a hugely expensive and often frustrating process. Medicinal chemists must guess which compounds might make good medicines, using their knowledge of how a molecule’s structure affects its properties. They synthesize and test countless variants, and most are failures. “Coming up with new molecules is still an art, because you have such a huge space of possibilities,” says Barzilay. “It takes a long time to find good drug candidates.”
By speeding up this critical step, deep learning could offer far more opportunities for chemists to pursue, making drug discovery much quicker. One advantage: machine learning’s often quirky imagination. “Maybe it will go in a different direction that a human wouldn’t go in,” says Angel Guzman-Perez, a drug researcher at Amgen who is working with Barzilay.
“It thinks differently.”
Others are using machine learning to try to invent new materials for clean-tech applications. Among the items on the wish list are improved batteries for storing power on the electric grid and organic solar cells, which could be far cheaper to make than today’s bulky silicon-based ones.
Such breakthroughs have become harder and more expensive to attain as chemistry, materials science, and drug discovery have grown mind-bogglingly complex and saturated with data. Even as the pharmaceutical and biotech industries pour money into research, the number of new drugs based on novel molecules has been flat over the last few decades. And we’re still stuck with lithium-ion batteries that date to the early 1990s and designs for silicon solar cells that are also decades old.
The complexity that has slowed progress in these fields is where deep learning excels. Searching through multidimensional space to come up with valuable predictions is “AI’s sweet spot,” says Ajay Agrawal, an economist at the Rotman School of Management in Toronto and author of the best-selling Prediction Machines: The Simple Economics of Artificial Intelligence.
In a recent paper, economists at MIT, Harvard, and Boston University argued that AI’s greatest economic impact could come from its potential as a new “method of invention” that ultimately reshapes “the nature of the innovation process and the organization of R&D.”
Iain Cockburn, a BU economist and coauthor of the paper, says: “New methods of invention with wide applications don’t come by very often, and if our guess is right, AI could dramatically change the cost of doing R&D in many different fields.” Much of innovation involves making predictions based on data. In such tasks, Cockburn adds, “machine learning could be much faster and cheaper by orders of magnitude.”
In other words, AI’s chief legacy might not be driverless cars or image search or even Alexa’s ability to take orders, but its ability to come up with new ideas to fuel innovation itself.
Ideas are getting expensive
Late last year, Paul Romer won the economics Nobel Prize for work done during the late 1980s and early 1990s that showed how investments in new ideas and innovation drive robust economic growth. Earlier economists had noted the connection between innovation and growth, but Romer provided an exquisite explanation for how it works. In the decades since, Romer’s conclusions have been the intellectual inspiration for many in Silicon Valley and help account for how it has attained such wealth.
But what if our pipeline of new ideas is drying up? Economists Nicholas Bloom and Chad Jones at Stanford, Michael Webb, a graduate student at the university, and John Van Reenen at MIT looked at the problem in a recent paper called “Are ideas getting harder to find?” (Their answer was “Yes.”) Looking at drug discovery, semiconductor research, medical innovation, and efforts to improve crop yields, the economists found a common story: investments in research are climbing sharply, but the payoffs are staying constant.
From an economist’s perspective, that’s a productivity problem: we’re paying more for a similar amount of output. And the numbers look bad. Research productivity—the number of researchers it takes to produce a given result—is declining by around 6.8% annually for the task of extending Moore’s Law, which requires that we find ways to pack ever more and smaller components on a semiconductor chip in order to keep making computers faster and more powerful. (It takes more than 18 times as many researchers to double chip density today as it did in the early 1970s, they found.) For improving seeds, as measured by crop yields, research productivity is dropping by around 5% each year. For the US economy as a whole, it is declining by 5.3%.
The rising price of big ideas
It is taking more researchers and money to find productive new ideas, according to economists at Stanford and MIT. That’s a likely factor in the overall sluggish growth in the US and Europe in recent decades. The graph below shows the pattern for the overall economy, highlighting US total factor productivity (by decade average and for 2000–2014)—a measure of the contribution of innovation—versus the number of researchers. Similar patterns hold for specific research areas.