How GeologicAI uses high-resolution data to reduce uncertainty in mining

Every drilling program begins with the same objective: gather enough information to understand what lies beneath the surface. The quality of early geological data influences nearly every decision that follows. Yet, despite decades of advances in mining technology, geological logging remains a time-intensive process. As demand for copper, rare earth elements and other critical minerals grows, mining companies face mounting pressure to shorten exploration timelines. One of the slower hills to climb is drill core logging. GeologicAI is among a growing group of mining technology companies seeking to address that challenge. Unlike artificial intelligence (AI)-focused exploration firms that concentrate on predicting where deposits may be located, GeologicAI’s focus is on improving the quality and speed of geological information collected after drilling begins. Through AI, automation and sensing technologies, the company seeks to improve geological understanding and reduce uncertainty throughout the mining life cycle. Founded in 2013 by engineer and geoscientist Grant Sanden, the company combines robotics, cloud computing, AI and a suite of sensing technologies to create detailed digital representations of drill core. It works to provide geologists and engineers with more complete geologic information earlier in the project life cycle. “The biggest lever that we have in mining is geological understanding,” said Sanden. “Capturing uncertainty, modeling things better than we have in the past, and modeling them with high resolution makes a big difference in outcomes.”With a background in engineering and geoscience, Sanden noticed a disconnect emerging across the exploration industry. Advances in geostatistics and predictive modeling were outpacing the quality of the geological information being collected. While sophisticated modeling techniques were becoming available, the underlying data sets often remained limited by traditional logging methods. “We set out to create a system that could capture everything we could practically capture at a reasonable cost and could scan all the core in a mine site with multiple sensors,” Sanden said. Although GeologicAI was established in 2013, its early years looked considerably different than the company it is today. For its first six years, the company focused primarily on oil sands mining, operating largely as a research consultancy while developing technology in collaboration with several major oil sands producers. In 2019, the company expanded into hard rock mining and shifted toward a commercial technology platform. “We launched into core scanning and hard rock,” said Sanden. “That was a big break for us, and we created it as more of a venture business than a consulting business. We built units that went on site. That was our Gen Three units, and that was our initial fleet deployment. Now we’re on Gen Four.” Tools for geological understanding GeologicAI’s approach is intended to supplement geologist workflows with additional information. Rather than replacing geological expertise, the technology is intended to free geologists from repetitive tasks and allow them to spend more time interpreting geological relationships. GeologicAI applies machine learning models designed to identify geological relationships within large integrated data sets. By analyzing thousands of variables simultaneously, the system can help geologists recognize lithologies, mineralization trends and geochemical patterns. “The computer can say how certain it is,” said Sanden. “Certain lithologies are super easy, usually about 80 to 85 percent for a computer to predict … so the geologists can focus on the interesting stuff and do a much better job, much quicker, because their work is more targeted,” Sanden said. For decades, geological logging has relied on visual observations, laboratory assays, and the expertise of geologists interpreting drill core. While those methods remain fundamental to mineral exploration, they also present limitations. Laboratory results can take weeks to return, logging practices can vary between geologists, and many geological characteristics simply are not visible to the human eye. This idea became the foundation of GeologicAI’s multisensory core scanning platform. Instead of relying on a single analytical technology, the company’s systems combine hyperspectral imaging, X-ray fluorescence (XRF), laser-induced breakdown spectroscopy (LIBS), high-resolution RGB imaging, laser tomography, magnetic measurements and AI-assisted analytics to build a digital representation of drill core. Rather than functioning as individual data streams, the information is combined into a single integrated geological data set that can be analyzed using machine learning, statistical modeling and resource evaluation tools. Each technology contributes a different layer of information.“We invented the multisensor core scanning approach, which was: Hyperspectral’s great; it can do molecular. XRF is great, and LIBS; they can do elements,” said Sanden. “Let’s put them all together, and add cameras, of course, the industry is already using for high resolution, and laser tomography and magnetics for textures and high resolution. That’s rocket fuel for prediction. We can do a much better job in geoscience because we have a better data source.”While XRF is effective for many heavier elements, LIBS provides information about lighter elements and rare earth elements, allowing for a more comprehensive geochemical characterization of drill core. Exploration advantages One of the most immediate applications of high-resolution geological data is during exploration drilling, where companies must decide whether additional drillholes are justified and where they should be located. Traditional workflows often require waiting weeks for laboratory assays before modifying a drill program. During that time, drilling equipment may continue operating without the benefit of the latest geological information. “You’ve got automated models, and therefore you have much more risk-optimized decisions. That compounds, saving half your drill costs and things like that,” Sanden said. GeologicAI’s sensor data can provide early indications that support faster decision-making. This could give operations the opportunity to allocate capital differently and make more informed decisions. “Some of our flagship customers are doing a lot of work, not just with our technology, but using our technology with their insight that high-resolution geoscience and decision making makes a difference. And we used to say, ‘There’s a same site, you can get a 5 percent improvement.’ Some of our customers are seeing a 20. That’s a very big deal,” he added. The company also applies its technology to drill-spacing studies. By integrating scanned core data with resource modeling workflows, companies can evaluate whether existing drilling provides sufficient confidence for resource classification or whether additional drilling is required. “We can take the data you have, model what is the drill spacing … and do that in a much more scientific way than is traditionally done,” Sanden said. Those data sets can feed directly into automated geological models that update as new information becomes available. Advancing in a risk-averse industry Even with its advances, adoption required patience. Resource estimates, mine plans and capital decisions carry substantial financial consequences, making mining companies naturally cautious about adopting unfamiliar tools. “Mining is a notoriously risk-averse industry,” said Sanden. “You have to look for that pocket of a person or a group of people within a company who can actually impact change and want to impact change. That’s actually a very rare thing.” Fortunately enough, Sanden said, GeologicAI started its early trials with large-sized companies. “We actually thought we’d start at the mid-tier, but we actually started at the largest companies in mining,” he said. “Because they had the innovation teams or people who wanted to trial and see the value of this technology.” More recently, in July 2025, GeologicAI announced a $44 million Series B financing that included participation from BHP Ventures and Rio Tinto, two of the world’s largest mining companies. GeologicAI stated that the companies were joining not only as investors but also as customers and collaborators. As countries seek to strengthen domestic supplies of critical minerals, interest has grown in technologies that can help mining companies advance mineral resources more efficiently. GeologicAI sees opportunities extending beyond individual projects and into larger questions of capital allocation across the industry. “I think there’s an opportunity to actually do this at a very big scale, at the country level,” he said. “We’re doing capital allocation for companies at their own deposit scale, but there’s no reason the world shouldn’t elevate its data so that the mining industry can be much more efficient at targeting some of these game-changing deposits that have the real potential.” Currently, GeologicAI works across commodities including copper, iron ore, gold and zinc while increasingly supporting rare earth and critical mineral projects. The technology is particularly valuable in geologically complex environments where variability can significantly influence resource estimates and mine planning. Growing focus has been placed particularly on copper, which is essential to electrification and data infrastructure growth. “There’s an obvious copper problem that’s hitting the industry,” Sanden said. “If it takes 17 years to bring on new deposits, that’s too long.” While GeologicAI’s technology can assist with the early identification of rare earth mineralization, geology is only one piece of a larger supply-chain challenge. The company sees the technology’s value extending beyond exploration. The same data sets generated during drilling can support resource modeling, geometallurgical studies, geotechnical assessments, mine planning and grade control. Rather than existing as standalone exploration data, the information can become part of an evolving digital representation of an orebody throughout the life of a project. As demand for critical minerals continues to grow, the ability to generate faster, more comprehensive geological information may increasingly shape how companies evaluate the deposits needed to support new technology infrastructure.

How GeologicAI uses high-resolution data to reduce uncertainty in mining

Authored by: Carrie Smith

To view the past issues of the UCA magazine you must be a member of SME/UCA.

Mining Engineering’s Weekly Rundown: Episode 73

Get your SME Foundation VIP Access Conference Pass to MINEXCHANGE 2027

“Check the box” and help support the SME Foundation

Mining Engineering’s Weekly Rundown: Episode 73

Get your SME Foundation VIP Access Conference Pass to MINEXCHANGE 2027

“Check the box” and help support the SME Foundation

To view the past issues of the UCA magazine you must be a member of SME/UCA.

AI-ready digital twin for mineral processing plants: Maximizing copper production with dynamic water management

Rare earth elements in the global economy: Usage, recovery and the quest for supply security — A review

Related Articles

Get your SME Foundation VIP Access Conference Pass to MINEXCHANGE 2027

“Check the box” and help support the SME Foundation

The Art and Science of the Flowsheet

UCA Division officer nominations

Mining Engineering’s Weekly Rundown: Episode 73

Get your SME Foundation VIP Access Conference Pass to MINEXCHANGE 2027

“Check the box” and help support the SME Foundation

Mining Engineering’s Weekly Rundown: Episode 73

Get your SME Foundation VIP Access Conference Pass to MINEXCHANGE 2027

“Check the box” and help support the SME Foundation

Use Single Sign On