IBM Watson Forms Historic AI Exploration Partnership with Goldcorp

IBM and Goldcorp are proving that adding AI capabilities to a company can be a literal goldmine.

A new innovative offering from IBM Canada and the Vancouver-based Goldcorp is helping the latter better predict where gold mineralization might occur. IBM Exploration with Watson is the computing company’s new solution that enables Goldcorp to access powerful search and query capabilities across a huge breadth of exploration datasets. Watson is IBM’s AI computing solution that can be plugged into a company’s workflow to improve and streamline existing systems.

“The potential to radically accelerate exploration target identification combined with significantly improved hit rates on economic mineralization has the potential to drive a step-change in the pace of value growth in the industry,” said Todd White, COO of Goldcorp.

Goldcorp’s mine in Red Lake, Ontario was constructed to harvest gold from the earth. The new IBM Exploration with Watson platform allows Goldcorp to use spatial analytics, machine learning and predictive models to point out what information gathered from the Red Lake excavations is most valuable to the company, then develop and extrapolate geological models in a short time and without the need for other expensive forms of modeling.

“Applying the power of IBM Watson to these unique challenges differentiates us in the natural resources industry,” said Mark Fawcett, an IBM Canada partner. “We are using accelerated computing power for complex geospatial queries that can harmonize geological data from an entire site on a single platform. This is the first time this solution has been ever used, which makes this project all the more significant.”

Through this new partnership, IBM Watson was able to verify its value by providing support to targets already planned and located by traditional methods, then going on to identify new targets which were subsequently verified. Many of these locations have yet to undergo drilling, but the first target was successful in terms of yielding the predicted mineralization at the estimated depth. In non-mining terms, this essentially means IBM Watson helped Goldcorp find gold within a mineral deposit without having to go through costly scanning processes.

“Timelines are short in mining and exploration. I am excited to see the improvements we can make with the data platform and gold mineralization predictions,” said Maura Kolb, Goldcorp’s exploration manager at Red Lake Gold Mines. “These tools can help us view data in totally new ways. We have already begun to test the Watson targets from the predictive model through drilling, and results have been impressive so far.”

The technology will continue to see use as Goldcorp expands the scope of their IBM partnership throughout 2019. They first began working together in 2017 but unveiled the scope of how much Watson could help in early 2018 As Goldcorp generates more and more data from drilling reports, land surveys and other types of collections, IBM will have more to work with and begin to optimize the process even further.

Goldcorp has long been in the business of finding new ways to innovate in mining. They are the co-hosts of Disrupt Mining, a pitch competition where companies compete to win $1 million and work with Goldcorp themselves.