At the forum of mining industry leaders as part of the MiningWorld Russia 2026 exhibition, a discussion was held “Digital VS production: who makes decisions and what to invest in?” The director of information technology and innovation of Rosgeologia, Pavel Solovyov, took part in it. The main question: who to trust to manage production processes – algorithms or live specialists?
According to Solovyov, artificial intelligence capable of significantly speeding up data processing, but cannot be held responsible for decisions made. If you transfer control to the system, the problem of risk distribution arises. “This is an effective tool, but it must be treated with understanding and a great deal of caution,” he emphasized.
Today Rosgeo is already actively using AI. RPA services have been implemented in many processes. Artificial intelligence systems help process geological information and current applications. The company is developing ambitious plans to automate monitoring and control of geological exploration work. According to Solovyov, these solutions will be put into productive use in the near future.
One of the key questions is how long does it take to train AI work predictably and give useful advice. Rosgeo has accumulated a large amount of production information on monitoring field work for solid minerals. The process has been going on for ten months. Now the company can carry out forecasting and analysis with a high degree of probability. Nine months turned out to be enough to accumulate a data array that ensures high accuracy of forecasts.
During the discussion, participants concluded that the industry is searching for a balance between digital tools and production expertise. Key questions remain open: who is responsible for errors, how to build interaction with solution providers, and how to evaluate the effectiveness of investments in digital technologies.
The positions of vendors and subsoil users today coincide: everyone is faced with a shortage of resources, so only short projects that give a quick effect are in demand. Artificial intelligence helps speed up processes, but you cannot rely on it without control. A compromise is needed, where the person retains the last word, and the number remains a tool for supporting decisions.
Source: Rosgeology
Image: Rosgeologia
