AI Finds Ore: KoBold Metals' AI Discovery Strategy

The race for critical minerals essential for the clean energy transition and advanced technologies is intensifying, and artificial intelligence is emerging as a pivotal tool in this pursuit. KoBold Metals, a company focused on leveraging AI for mineral exploration, is at the forefront of this shift. In a recent interview on "Pioneers of AI," KoBold Metals' co-founder and president, Josh Goldman, detailed how the company is using AI to revolutionize the discovery of vital resources like copper and lithium.

Who is Josh Goldman?

Josh Goldman, co-founder and president of KoBold Metals, brings a unique blend of scientific rigor and business acumen to the resource exploration sector. With a PhD in physics, Goldman possesses a deep understanding of complex systems and data analysis. His background includes experience in private equity and the oil and gas industry, where he honed his skills in identifying and developing high-value assets. This diverse experience has shaped his approach to applying cutting-edge technology, particularly AI, to solve fundamental challenges in resource discovery. Goldman's transition from physics and finance to the mining sector underscores a growing trend of cross-disciplinary innovation driving progress in traditional industries.

The AI Advantage in Exploration

Goldman explained that traditional mineral exploration is a time-consuming and often inefficient process. Historically, it relied heavily on geological intuition, extensive fieldwork, and basic data analysis. However, the sheer volume of available geological and geophysical data has grown exponentially, making manual analysis increasingly impractical. KoBold Metals' core innovation lies in its ability to process and analyze these massive datasets using AI.

The full discussion can be found on Pioneers of AI's YouTube channel.

The company utilizes AI algorithms to identify subtle patterns and correlations within geological data that human geologists might miss. This includes analyzing satellite imagery, seismic data, magnetic surveys, and geochemical samples. By training AI models on known mineral deposit locations and their associated geological characteristics, KoBold Metals can predict the probability of finding similar deposits in new areas. This predictive capability significantly narrows down the search space, allowing exploration teams to focus their efforts more effectively.

Goldman highlighted the sheer scale of data involved, stating, "We're collecting many terabytes of data day and today... and we have to put that all into models and combine it with other data sets." This data-intensive approach is crucial for the success of their AI models, enabling them to learn from a vast array of geological information.

Transforming Discovery with Data

The company's approach addresses a fundamental challenge: the scarcity of readily accessible mineral deposits. As easily discoverable resources are depleted, exploration must increasingly focus on more complex and hidden deposits. AI provides the tools to sift through vast amounts of subtle geological indicators, increasing the likelihood of finding these elusive resources.

Goldman elaborated on the process: "We can see when we see those patterns, and then we can use that data to make predictions about where the resources are going to be found." He likened the process to an AI learning to recognize patterns in a vast dataset, enabling it to make informed predictions about future discoveries.

Addressing the Critical Minerals Gap

The demand for minerals like copper and lithium is projected to surge in the coming decades, driven by the electrification of transportation and the expansion of renewable energy infrastructure. Goldman pointed out the critical need for these materials, noting that by mid-century, the global demand for copper, for example, will need to increase significantly beyond current mining rates.

KoBold Metals' AI platform aims to accelerate the discovery of these vital resources, thereby helping to meet this growing demand. By identifying promising exploration targets more efficiently, the company can shorten the timeline from discovery to production, which is crucial for the energy transition. Goldman emphasized that their AI models help them to both identify existing known deposits more effectively and to find entirely new ones that might have been missed by traditional methods.

The Role of "Ground Truth" Data

A key aspect of KoBold Metals' success is its ability to leverage existing geological data and integrate it with new findings. Goldman explained that much of the historical geological data is publicly available, but it is often fragmented and difficult to analyze. KoBold Metals' AI platform aggregates and standardizes this information, creating a comprehensive dataset that can be used to train and validate their models.

He highlighted the importance of this "ground truth" data, explaining, "The vast majority of the ore deposits that people have found are still out there to be discovered... and what's left to find are deposits that are either very close to the surface or are hidden somehow." The AI's ability to identify these subtle, often hidden, indicators is what sets KoBold Metals apart.

The Future of AI in Mining

Goldman also touched upon the broader implications of AI in the mining industry, emphasizing that it's not just about finding minerals but also about optimizing the entire value chain. AI can be applied to improve mine planning, operational efficiency, safety, and environmental management.

Regarding the future, Goldman expressed optimism about the continued integration of AI in resource exploration. He believes that AI will become an indispensable tool for geologists and mining companies, enabling them to discover and extract critical resources more sustainably and efficiently. The company's ongoing work with AI and its commitment to data-driven decision-making position it as a key player in shaping the future of mineral resource development.