‘Don’t shoot. Listen’: The AI-aided quest to understand Yellowstone’s wolves

Jeff Reed, a Montana rancher and former Silicon Valley software engineer, has spent five years building the Cry Wolf Project, a bioacoustics effort based near Yellowstone that uses about 50 automated recording units to gather roughly 500,000 hours of soundscape data a year. AI models he built extract more than 8,000 hours of continuous wolf vocalizations from that data, reportedly the largest such database in the world.

The project sits alongside the long-running Yellowstone Wolf Project, led by chief wildlife biologist Dan Stahler, which currently tracks eight packs totaling 84 wolves. Wolves were eradicated from the park by 1926 and reintroduced in 1995 to rebalance an ecosystem overrun by unchecked elk and deer.

Stahler's team, whose own methods rely on costly and risky helicopter collaring operations, has welcomed Reed's AI-driven approach as a complementary tool for studying wolf communication, including the chorus howls packs perform together.

The full dispatch is available from the source below.