Russia Deploys AI-Powered Vehicles to Hunt Dangerous Giant Hogweed

Neural network–equipped patrol cars now scan hundreds of routes to stop the spread of a toxic plant.

In Russia, artificial intelligence is being used to track down a dangerous invasive plant — Sosnowsky’s hogweed. This towering weed can grow several meters tall, and its sap causes severe chemical burns on human skin. Researchers have trained a neural network to detect the plant at every stage of its growth.

The AI system is integrated into the “Gorodovoy” mobile monitoring network. Eight vehicles equipped with neural-network technology now patrol St. Petersburg along 186 routes covering every district of the city. When the system automatically identifies hogweed, it records the violation and issues the property owner a fine of 100,000 rubles (about $1,100). Follow-up inspections are also automated: the AI verifies the site during subsequent patrols.

This new approach allows every district in the city to be scanned within just three to four weeks. Neural network–based monitoring has dramatically increased efficiency compared to traditional methods, which could take months — a crucial improvement during the plant’s rapid growth season.

Russia’s anti-hogweed arsenal also includes drones. These smart aerial systems can precisely map the plant’s spread before evenly applying herbicides, minimizing environmental impact while eradicating the threat.