In Pakistan-administered Gilgit-Baltistan, an initiative is underway to mitigate conflicts between villagers and the endangered snow leopard population, estimated to be only 4,000 to 6,000 globally, with around 300 in Pakistan. A significant cause of snow leopard deaths is retaliation for livestock losses.

To address this, a collaboration between a conservation organization and a local university has developed AI-powered cameras. These cameras, equipped with solar panels and lithium batteries for the harsh mountain environment, aim to detect the presence of snow leopards. Once a snow leopard is identified, the system is designed to send text message alerts to villagers, giving them time to secure their livestock.

The AI model, trained over three years, can differentiate between humans, other animals, and snow leopards with high accuracy. Ten cameras are currently being tested across three villages at high altitudes. While the technology has shown promise in detecting snow leopards, challenges include ensuring consistent cellular service for alerts and protecting the equipment from environmental factors like landslides.

Gaining the trust and cooperation of local communities has also been crucial. Initial skepticism and concerns about privacy led to instances of tampering with the cameras and delays in obtaining consent for deployment in some areas. The conservation organization is working to build understanding of the ecological importance of snow leopards, which prey on animals that can overgraze grasslands vital for villagers’ livestock.

Beyond the AI cameras, the project plans to trial deterrents such as smells, sounds, and lights near camera sites to discourage snow leopards from approaching villages. While the AI technology offers hope, it is considered part of a broader strategy to protect these elusive “ghosts of the mountains.”