Last summer, the mining startup KoBold made headlines when it announced the discovery of one of the world’s largest copper deposits in Zambia, a find not seen in over a decade.
Recently, another startup, Earth AI, revealed to TechCrunch its discovery of promising deposits of critical minerals in parts of Australia that have been overlooked by other mining companies for decades. Although the size of these deposits is yet to be compared with KoBold’s findings, the news indicates that future supplies of critical minerals may emerge through a combination of field data and artificial intelligence analysis.
Roman Teslyuk, the founder and CEO of Earth AI, explained to TechCrunch that the frontier in mining is now more technological than geographical. Earth AI has identified deposits of copper, cobalt, and gold in the Northern Territory, as well as silver, molybdenum, and tin in New South Wales, located 310 miles (500 kilometers) northwest of Sydney.
Earth AI, which originated from Teslyuk’s graduate studies, was founded following his doctoral pursuit at the University of Sydney. There, Teslyuk became acquainted with Australia’s mining industry, where the government owns mineral deposit rights and issues six-year leases. Since the 1970s, exploration companies have been required to submit their data to a national archive. Despite the availability of such data, Teslyuk noted that it has been largely underutilized. He believed that by creating an algorithm capable of learning from historical geological successes and failures, better predictions about future mineral locations could be made.
Initially, Earth AI focused on developing software to predict potential deposit locations and sought customers interested in exploring these sites. However, potential clients were reluctant to invest in unproven technology. Recognizing the conservative nature of the mining industry, Earth AI decided to create its own drilling equipment to validate the potential of the sites identified by its software. The company joined Y Combinator’s spring 2019 cohort and has since worked on refining its hardware and software. In January, Earth AI secured a $20 million Series B investment.
While Earth AI also employs AI to search for minerals like KoBold, Teslyuk stated that their approach differs. Earth AI’s algorithms are designed to efficiently scan large areas to detect deposits that might have been missed by traditional methods.
Teslyuk remarked that traditional methods of metal exploration, characteristic of the 20th century, are time-consuming and often take decades to yield results. In today’s fast-paced world, such extended timelines are impractical.
