Last summer, mining startup KoBold made headlines with the discovery of one of the world’s largest copper deposits in Zambia. Now, another startup, Earth AI, has identified promising deposits of critical minerals in overlooked parts of Australia. While the scale of these findings is still uncertain, the use of artificial intelligence to analyze field data is proving to be a game-changer in mineral exploration.
“The actual, real frontier [in mining] is not so much geographical as it is technological,” said Roman Teslyuk, founder and CEO of Earth AI.
Earth AI has pinpointed copper, cobalt, and gold in the Northern Territory, as well as silver, molybdenum, and tin in New South Wales, 310 miles (500 kilometers) northwest of Sydney.
Harnessing AI for Smarter Mining
Earth AI originated from Teslyuk’s graduate research. A native of Ukraine, he pursued a doctorate at the University of Sydney, where he gained insight into Australia’s mining industry. Since the 1970s, companies have been required to submit exploration data to a national archive, yet few have utilized this vast resource.
“For some reason, nobody’s using them,” Teslyuk said. “If I could build an algorithm that can absorb all that knowledge and learn from the failures and successes of millions of geologists in the past, I can make much better predictions about where to find minerals in the future.”
Initially, Earth AI functioned as a software company, offering predictive models to mining firms. However, potential customers hesitated to invest in an unproven technology. “Mining is a very conservative industry,” Teslyuk explained. “Everything outside of the approved dogma is considered heresy.”
From Prediction to Discovery
To validate its findings, Earth AI developed its own drilling equipment to confirm the accuracy of its predictions. The company joined Y Combinator’s spring 2019 cohort and spent the following years refining its technology. In January, Earth AI secured $20 million in Series B funding.
Although both Earth AI and KoBold use AI-driven exploration, Teslyuk emphasizes a key difference. Earth AI’s algorithms are designed to scan vast areas rapidly, identifying overlooked deposits with greater efficiency.
“The way we used to explore for metals in the past, the 20th century, it just takes very, very long. It takes decades to find something,” Teslyuk said. “With the modern pace of the world, you just can’t wait for that long.”
