Mining start-up Earth AI, which has a presence in the United States and Australia, has discovered deposits of critical minerals in parts of Australia “that other mining outfits had ignored for decades.”
According to reports, these future supplies of critical minerals are likely to emerge from a combination of field data parsed by artificial intelligence (AI), providing a glimpse into how the industry will operate in the coming days.
While speaking with TechCrunch, Roman Teslyuk, founder and CEO of Earth AI, stated, “The actual, real frontier [in mining] is not so much geographical as it is technological.”
His company has identified deposits of copper, cobalt, and gold in Australia’s Northern Territory, and silver, molybdenum, and tin at another site in New South Wales, 310 miles (500 kilometres) northwest of Sydney.
Earth AI emerged from Teslyuk’s graduate studies. The Ukraine native, while working toward a doctorate at the University of Sydney, became familiar with Australia’s mining industry. There, the government owns the rights to mineral deposits and leases them for six-year terms. According to Teslyuk, under the current rule, exploration companies are required to submit their data to a national archive.
“For some reason, nobody’s using them. 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,” he said.
In today’s episode of the “Start-up of the Week,” International Finance will discuss in detail how Earth AI is redefining 21st-century mining activities.
Knowing The Company In Detail
Earth AI, in its early days, started as a software company focused on making predictions about potential deposits before approaching customers who might be interested in exploring sites further. However, customers were hesitant to invest, partly due to their reluctance to spend millions on the predictions of an “unproven technology.”
To address this hesitancy, Earth AI developed its own drilling equipment to prove that the sites it identified were as promising as its software suggested. The company was accepted into Y Combinator’s Spring 2019 cohort and spent the next few years refining its hardware and software.
In January 2025, Earth AI raised a USD 20 million Series B. Even though the company uses AI to search for minerals, the start-up takes a different approach in terms of algorithms, as the latter has been trained to scan wide areas quickly and efficiently to find deposits that might otherwise have been overlooked.
With the global economy’s rush toward a net-zero transition, Earth AI’s prospects look bright, given the explosion in demand for critical metals.
Earth AI is now aiming to help mineral explorers discover untapped critical metal deposits at half the cost and in a fraction of the time through its AI-driven field data parsing method (the process of analysing and converting data from one format to another, typically to make it more structured, readable, and usable for analysis or storage), offering its clients three significant benefits: the lowest drilling costs, precise prospect detection, and minimal environmental impact.
“With accurate AI-driven prospect detection, modular hardware, and streamlined operations, Earth AI can move from detecting a prospect to drilling in just three to six months. Our revenue model is independent of service profits, and our process is four times faster than traditional methods. Once a discovery is made, we partner with explorers and development companies to bring these deposits into production,” the start-up stated.
Meet The Key Technology
Earth AI is decoding complex geology and turbocharging exploration drilling to scale up mineral exploration. With the industry’s most precise mineral detection technology, fundamental geology, and responsible drilling, the start-up now possesses the ability to move from detecting a prospect to drill testing in just three to six months. At the heart of these efforts remains Earth AI’s “Mineral Discovery Engine,” a system of key processes and technologies that bring precision, efficiency, and scalability to critical minerals exploration.
The “Mineral Discovery Engine” has its own prediction tools, which Earth AI claims are the “most accurate in the mining industry,” improving discovery success rates from 0.5% to 66%.
“By studying multiple datasets and learning to identify mineralisation, our AI replicates and scales the mind of a geoscientist. Our models train on 400 million geological cases from the entire continent, creating new knowledge and formulating conclusions. This experience, achieved through extensive training, allows us to deliver repeatable, high-quality predictions,” the venture explained.
To understand the origin of mineralisation, the start-up applies a generalist approach to studying and understanding the geology of the area. With every project, data leads the way. Earth AI then applies a unique set of field and laboratory methods to decode the mineral system, form hypotheses, and test them through drilling. This approach leads to the detection of brand-new prospects.
Earth AI has added another first to its achievements by developing the first modular drilling system for self-sufficient mineral exploration in remote desert environments. From drilling hardware and fluid treatment systems to crew housing, the start-up has meticulously optimised every step in the process while minimising environmental impact.
Earth AI’s mineral exploration activities follow three stages: discovery (one year), feasibility study (three years), and mine construction (two years). During this five-year cycle, the “Mineral Discovery Engine” and its “Mineral Targeting Platform” predict areas with possible mineral enrichment, followed by the start-up’s field teams sampling the sites to confirm or deny the presence of minerals.
Earth AI then conducts generalist geological research to understand the mineral system and plan drilling. During this phase, each drill hole tests a hypothesis. The learnings derived from this process are then applied to refine future hypotheses. Next, the mining alliance partners working with Earth AI take the deposit through the feasibility stage, and upon clearance, the metals enter the production stage. After feasibility confirms the site’s economic viability, mine construction begins.
A single mine in Australia, with a lifespan ranging from eight to over 100 years, can produce annual revenues from USD 50 million to USD 3 billion. Earth AI uses drill samples of economic-grade ore as its “success criteria,” after which it gains a perpetual royalty from the mine’s revenue.
At the same time, commercial miners partnering with the start-up benefit from its efficient, modular drilling design and streamlined operations, translating into best-in-class drilling economics. Additionally, the drilling system requires no groundwork and causes close to zero environmental impact, while the onboard mud filtration system minimises water wastage and prevents soil contamination, reducing downtime and costs.
The Willow Glen Success Story
Another success story of Earth AI is its discovery of a new gold system at its molybdenum-gold project in Willow Glen. The discovery, made using AI software to drill 650 metres of rock, yielded seven intercept samples of mineralised gold. The discovery was of economic-grade gold, at approximately 1.14 grams of gold per tonne of rock. Earth AI will follow up with continued mapping and drilling to assess the nature of the mineralisation and the size, grade, and economic value of the deposit.
Earth AI acquired the project in May 2021, following a state-wide analysis using its patented AI software to develop predictive clusters of potential base metal targets. Predictive targets were subsequently investigated, and areas of surface metal enrichments became targets for the maiden drilling campaign that commenced in December 2022. By January 2025, six diamond drill holes had been drilled, totalling 3,000 metres in length.