Companies all over the world are using AI to identify rare earth minerals, reducing the need to drill unnecessarily in the exploration phase. 

BY CHRIS STOKEL-WALKER

Illustration: Getty Images

Rare earth metals, a clutch of 17 elements used in consumer electronics, including smartphones and electric vehicles, are a precious bounty. But finding them can be tricky. That’s a problem because some of these metals, such as copper and lithium, are expected to see demand outstrip supply by as early as 2035, according to the International Energy Agency (IEA). 

As a result, more work is going into the exploration and production (E&P) of those rare earths. But before you can get them out the ground, you first have to find them.

That’s where AI is coming in handy. 

“Mining is a complex economic equation that relies on resources,” says Grant Sanden of Geologic AI, an AI-powered minerals exploration company. “Resources are interesting because of the cost and sparse data.” You don’t drill where you don’t think there’s a good chance of striking it lucky. But you can’t always tell where your odds will improve, particularly if it’s a part of the world that hasn’t been explored before.

Featured Video

How CoreWeave Bought and Borrowed Its Way to AI Domination

AI can help filter through what data there is—and fill in the gaps along the way—while bringing down the cost. 

Sanden: “What we’ve done is said: ‘We have the ability to model thousands of variables at once now. Let’s let computers figure it out, as opposed to just doing it by hand, which is the traditional way of doing it.’” 

His Calgary, Alberta-based company, founded in 2013, recently raised

© Inc.com