– What value do geological archives offer today, and how does MINTEL use them to assess the potential of exploration sites?
– Archives are an enormous repository of knowledge. Our predecessors carried out extremely valuable work, and failing to use this data in decision-making would be a serious mistake. MINTEL focuses on reducing the time required to process and analyse archival data, organising it into a clear structure and producing concise reports. By reading such a report, a geologist can gain an understanding of a particular site and determine whether it is worth exploring further.
– Is your software prone to errors?
– A key feature of our product is traceability: every statement and every figure can be traced back to its original source. We save geologists time, enabling them to evaluate more sites and select the most promising assets with genuine potential to advance to the production stage. In this way, we reduce uncertainty and accelerate decision-making.
– Could you explain the specifics of working with geological archives and share some of the results you have achieved?
– During the Soviet era, geological exploration was conducted across almost the entire territory of Kazakhstan, but exploration programmes focused primarily on near-surface deposits. Some sites were classified as unpromising because of their mineral grades. Researchers mentioned these sites in their reports but did not pursue further exploration.
Finding such references, often contained in a single paragraph, requires reading thousands of pages. By analysing several reports, our team discovered that one project had been considered promising in 1958, reclassified as unpromising in 1965, and reassessed in 1975, when researchers challenged the 1965 conclusion and determined that the project's potential had been underestimated.
This information can only be uncovered by examining all the original sources and assessments from each period.
– Which regions do you consider most attractive for expanding your project?
– One advantage of our platform is that the entire CIS region shares a common geological heritage rooted in the Soviet geological school. As a result, the formats and GOST standards used to prepare geological reports were essentially identical. This means our model can easily be applied in other CIS countries. We are particularly interested in Kyrgyzstan, Uzbekistan, and the Baltic states.
– Could you explain the role of artificial intelligence in the MINTEL project?
– Our product uses several AI models that function as agents. However, because our core value lies in traceability and the ability to link every finding to its original source, we have designed the platform to be as deterministic as possible to prevent AI hallucinations and inaccuracies.
We therefore deliberately limit the role of AI: if there is a risk that the agents have made an error while processing data, the potentially unreliable information is simply excluded from the report
