Will AI replace geologists in mineral discovery?

With its supreme qualities as an electrical conductor, the red metal is the one commodity that can’t be replaced in a future electrified economy.

Its why Goldman Sachs has labelled copper the New Oil.

There’s many reasons to be optimistic about stocks leveraged to this metal — But are there any risks?

Right now, we have high interest rates in the west and deflation in the east… Neither bodes well for a commodity tied exclusively to growth.

Yet copper continues to buck the doom and gloom sentiment… Futures are consolidating well above their April lows.

Meanwhile, major copper producers like Ivanhoe Mines (TSX: IVN) and Southern Copper (NYSE: SCCO) both hovering near all-time highs.

It’s widely accepted by the mining industry that underinvestment in new discovery will bring on shortages.

That was the assessment made by S&P Global Market Intelligence last year.

But let’s play devil’s advocate for a moment…

Could technology solve the looming supply problem?

According to the high profile New York based fund manager Goehring & Rozencwajg (G&R) there’s plenty of room for innovation in the ‘archaic field’ of mineral exploration meaning the idea of supply shortages could be overblown (emphasis added):

“Additional supply could come from introducing a new exploration tool from Ivanhoe Electric known as the “Typhoon” geophysical survey system.

As opposed to the oil and gas industry, which has made considerable advancements in exploration technology over the past 70 years, the mining industry largely relies on technologies used for hundreds of years.

Geologists walk along the surface with a pick hammer and a sampling “grab bag,” looking for rock outcrops and soil alterations.

A geologist from the eighteenth century would feel right at home today. Given the relatively primitive techniques, it stands to reason that most significant deposits are still located reasonably near the surface.

The oil and gas industry has developed exploration tools to look deep into the earth’s crust. A lack of technology restricted the mining industry to what could be found visually near surface.”

As a former geologist that’s worked in all phases of exploration across many different types of commodities, including copper, I disagree with G&R.

Yes, geoscientists conduct field work, it’s part of a process called ‘ground-truthing’ ensuring the data in the office matches what’s seen on the ground.

But more than that, exploration geologists have been using advanced software to identify chemical anomalies in the soil and computer programs to measure the probable geometries of ore bodies sitting ‘unseen’ below the surface, for years.

Computer science is an important skill for the profession… Software is ingrained with fieldwork.

The reality on the ground couldn’t be more different to G&R’s observations… Mineral exploration has been ‘technified’ for decades. 

But could companies like Ivanhoe Electric be the game changer?

I’d argue the Typhoon system described by G&R is just a reworking of current systems.

Just like the oil and gas industry, mineral exploration has been using deep penetrating geophysical surveys for decades.

A technology known as Induced polarisation (or ‘IP’ for short) has helped geologists find copper ore bodies lying hundreds of metres below the surface.

That puts a dent on the idea that ‘the most significant deposits are still located reasonably near the surface.’

IP has been uncovering these for decades.

As a former copper hunter, we used IP technology with Equinox Minerals in Zambia… A company I worked for back in 2010.

The technology assisted with two major copper discoveries.

But that wasn’t led by the IP geophysics… Rather the expertise from its operator, Alex Copeland.

He was a pioneer in this technology.

Alex held decades of experience in geophysics and mineral discovery… It was his time in the field refining and tweaking he’s technology that became the catalyst for the major discoveries.

And that’s the key point here…

It’s not so much the technology, rather the skill of the operator to interpret the noise from the data that matters in mineral exploration.

As Alex explained to me, being able to adapt the technology to the different conditions was crucial.

What worked in Australia didn’t necessarily work in Zambia.

Each deposit was different… IP surveys offer clues in one location… But those same clues are useless in other areas.

Simply put, nature is not a standardised playing field that can be plugged into an algorithm.

For decades new technologies have been touted as the game changer…. But rather than lead discovery, they are simply tools assisting operators narrow down their search.

A dwindling game of chasing deeper, lower grade and more marginal deposits.

But once again, a new ‘game changer’ is arriving in the form of Artificial Intelligence.

KoBold Metals, a company that uses AI and machine learning to hunt for raw materials has been peddled as the new generation in mineral discovery.

A company determined to prove computer scientists based in California can replace geologists and boots on the ground.

Yet despite receiving large capital injections from prominent billionaires, Jeff Bezos, Bill Gates and Richard Branson, KoBald is yet to make its first discovery.

In fact, rather than lead discovery, KoBold is using its deep pockets to buy up deposits previously discovered by those ‘primitive’ geologists!

In December 2022, the company acquired Mingomba copper project in Zambia.

No doubt, they’ll be a lot of hype in the years ahead about AI’s ability to change the landscape in mineral discovery but history proves there’s nothing new under the sun…

More supply won’t come from technological breakthrough it can only arise from higher prices.

James Cooper is a geologist and mining analyst. You can follow his Twitter Feed @JCooperGeo