Pentagon's special ops chief warns against rushing AI into combat

Adm. Frank Bradley, commander of U.S. Special Operations Command (SOCOM), delivered a sobering reality check on military AI during the annual special forces conference in Tampa, Florida. He told attendees that while AI holds transformative potential for warfare, troops "have to be very careful about how we come to its employment and its inspiration into the delivery of lethality." Bradley stressed that of all systems deployed on today's battlefields, "very few, if any of them, are actually using true AI at the edge." He clarified that this does not mean it's impossible—"We absolutely believe it is"—but urged caution in integrating AI into combat operations, especially when it involves decisions about lethal force.

Bradley's comments come as the Pentagon aggressively pushes for AI-powered battlefield tools, from autonomous drones to real-time targeting systems. His call for restraint highlights a growing tension within military leadership between speed of adoption and the need for rigorous testing, ethics, and reliability. The challenge of deploying "true AI at the edge"—systems that can learn and adapt in disconnected, high-stakes environments—remains significant. Bradley's warning suggests that while the military invests heavily in AI research, responsible deployment demands deliberate validation and oversight before such tools influence lethal actions.