Russian diamond miner Alrosa has developed an artificial intelligence system to aid in the search for new diamond deposits in geologically complex regions. The technology is expected to significantly cut exploration time and generate economic benefits worth hundreds of millions of roubles per project.

Exploration for new deposits has slowed in recent decades, as most major kimberlite pipes were discovered in the 20th century through traditional methods. Future finds are likely to be hidden deep underground, requiring advanced technologies. Alrosa’s AI system, built on large language models and neural networks, analyses vast geological datasets to identify promising sites more accurately.

“Our new system is a Russian development, and we are already testing it. Once trials are complete, it will be scaled across all Alrosa territories,” said CEO Pavel Marinychev.

Developed by Alrosa’s Digital Laboratory, the system draws on over 50 years of the company’s geological and geophysical data, fully digitised and integrated into its geodata bank and corporate GIS. By rapidly processing this information, the AI can guide geologists with targeted recommendations and improve decision-making in exploration.

Global diamond reserves are currently estimated at 1.7-1.9 billion carats, sufficient for only a few more decades at present production levels. More than half of these reserves are located in Russia.