AI Warfare Becomes Real for US Military With Project Maven

On a summer evening in 2020 at Fort Liberty, a sprawling US Army installation in North Carolina, soldiers from the 18th Airborne Corps pored over satellite images on the computers in their command post. They weren’t the only ones looking. Moments earlier, an artificial intelligence program had scanned the pictures, with instructions to identify and suggest targets. The program asked the human minders to confirm its selection: a decommissioned tank. After they decided the AI had it right, the system sent a message to an M142 Himars—or High Mobility Artillery Rocket System, the wheeled rocket launcher that’s a mainstay of America’s artillery forces—instructing it to fire. A rocket whistled through the air and found its mark, destroying the tank. The explosion was unlike any other in the hundreds of live-fire exercises that occur each year on Fort Liberty’s 146,000 acres of training grounds. In fact, it had no precedent in the Army. For the first time, American soldiers had struck a target located and identified by an AI program. Colonel Joseph O’Callaghan, the fire-support coordinator for the 18th and the leader of its AI targeting efforts, keeps a picture of the moment in a conference room, the tank engulfed in a bright fireball. “It showed us the art of the possible,” he says. Less than four years after that milestone, America’s use of AI in warfare is no longer theoretical. In the past several weeks, computer vision algorithms that form part of the US Department of Defense’s flagship AI effort, Project Maven, have located rocket launchers in Yemen and surface vessels in the Red Sea, and helped narrow targets for strikes in Iraq and Syria, according to Schuyler Moore, the chief technology officer of US Central Command.

Full feature : US military operators started out skeptical about AI, but now they are the ones developing and using Project Maven to identify targets on the battlefield.