“These are hard to roll over,” geologist Wilson Bonner assures me as the all-terrain vehicle he’s been riding suddenly tilts to one side, hurling me toward the churning mud beneath their wheels. I. We were heading toward a wooded hill in rural Ontario, Canada, on a chilly autumn day, toward a site that Bonner’s owner, the startup KoBold Metals, said represents for the combination of advanced artificial intelligence with one of humanity. oldest industries.
We actually completed the half-hour trek with relative ease, eventually passing a ring of broken trees and bushes ripped apart into a bulldozed mud flat. A black pipe about the size of my arm protrudes from the ground—the top end of a hole nearly a kilometer deep, punched into the ground by a truck-sized rig lying dormant nearby. It’s not much to look at, but the hole could mark a step toward the future of mining, an industry crucial to the world’s transition to renewable energy.
As the world begins to shift from fossil fuels to greener alternatives, there is an increasingly fierce global competition to find the vast quantities of cobalt, lithium and other metals needed to make them. all the electric car batteries, solar panels and wind turbines that we are using. need. But finding new mineral deposits is always difficult and expensive, and it’s getting harder and harder. Most of the world’s easy-to-find reserves have been tapped. The rest tend to be in remote areas and deep underground. Miners in general only 1 in 100 exploratory boreholes come up with anything.
KoBold Metals, a four-year-old startup, is one of the few companies trying to make the process faster, cheaper and more efficient by applying artificial intelligence. KoBold has built a massive database that combines all the information it can find about the Earth’s crust—the equivalent of 30 million pages of geological reports, soil samples, satellite images, documents scholarly research and centuries-old handwritten field reports. A team of data scientists converts all this disparate information into something machine-readable—for example, scanning written reports with optical character reading software or normalizing geographic information. physically recorded in various digital formats.
All of that is run through machine learning algorithms that identify patterns in the geology and other characteristics of places where metals have been found in the past. The algorithms can then be extended on the full database to find promising locations with unexplored similar patterns, creating a series of maps that indicate where the find is likely to be. target metals.
Backed by investors including venture firm Andreessen Horowitz and Bill Gates’ Breakthrough Energy Ventures, KoBold’s first exploration groups began operations last summer, explorating in areas in Zambia, Greenland and Canada—including the Ontario site near Crystal Lake.
KoBold is looking for copper, cobalt, nickel, lithium and rare earths — key components of electric car batteries and other renewable energy technologies. The International Energy Agency predicts demand for all of those metals could quadruple by 2050, and demand for some, like cobalt and nickel, could grow 40-fold. All told, the agency estimates the collective market for minerals needed for “clean energy technology”—everything from renewable energy sources to batteries and grids—will grow more than fivefold by 2050. to about $400 billion.