20:03, 17 February 2026

AI to Help Breed Future Racing Champions in Russia

Horse breeders begin using neural networks to guide selection

Russian horse breeders have begun using artificial intelligence in the selection of racing horses. Algorithms analyze the competitive performance of potential parents and help identify pairs most likely to produce strong offspring.

Searching for Champions Before Birth

Specialists upload performance data into the system and initiate the analysis. The model detects patterns that are difficult for the human eye to spot. The goal is to identify traits that can guide the choice of parents for future winners.

Initial calculations show that stallions successful at distances of 1,800–2,000 meters are particularly valuable for breeding. Their inherited qualities appear suitable for both sprinters and stayers.

“We input the parameters we consider important. In racing, not only speed but also endurance matters. Pure speed is valuable, but it is inherited only up to a point. Without endurance, an animal cannot deliver the required result. That applies even in sprint races. Moreover, these traits are inherited in a nonlinear way,” said Alena Akimova, head of production at the Pyatigorsk Hippodrome.

The system also confirmed that winners of top-level group races are more likely to become strong breeding parents.

A New Approach to Selection

The AI analyzes performance data from Russian and American horses. Since races in both countries are held on dirt tracks, the results can be compared directly. Russia has about 5,000 Thoroughbreds and roughly 1,500 Arabian racehorses. Each year, breeders purchase around 200 foals from abroad.

Selection focuses less on physical appearance and more on early competitive results. If a horse performs strongly at age two and maintains form at three and four, that suggests durability and long-term potential.

The big data methodology originated in livestock breeding but required adaptation for equestrian sports.

“In dairy farming it’s simpler. They have annual data on milk yields. In our field it’s more complicated because much depends on people. A horse may show clear potential, but what matters is whose hands it ends up in. It could be a super champion, but due to oversight or injury it may never reach its full potential. Statistics matter, but in this sport much still depends on the people working with the horse,” Akimova said.

Artificial Intelligence Combined With Human Expertise

Egor Melentyev, CEO of the Rosplemkonzavod association, said AI can accelerate pairing decisions and simplify preliminary selection of breeding stock.

“We could ask AI to identify the most successful sires and analyze bloodlines. That would be a strong support tool for breeders. Combined with human expertise, it would be beneficial, but we cannot rely entirely on artificial intelligence. It will likely suggest breeding the best with the best, yet racing, jumping and movement qualities are complex traits. Achieving them still requires human talent,” he said.

Vladimir Tishkin, coach of Russia’s national equestrian team, added that algorithms help analyze pedigrees more quickly and form an initial assessment of a horse.

Experts agree that artificial intelligence is becoming a practical tool in horse breeding, but producing a champion still requires experience and intuition.

Culture, sports and media