American soldiers stationed across Europe now train for a single catastrophic day: tracking and striking over 1,500 targets while stopping as many as 1,200 incoming missiles and drones within 24 hours. Commanders believe artificial intelligence may offer the only realistic path through that volume of data.
This planning benchmark stems directly from Russia’s relentless bombardment campaign against Ukraine, where ballistic missiles, cruise missiles, and one-way attack drones arrive in mixed salvos on a weekly basis. The sheer velocity and quantity of threats now outpace what human analysts alone can process.
Brig. Gen. Steven P. Carpenter, who leads Multi-Domain Command Europe, described the core problem without softening it. “Our challenge is processing and prioritizing that data, so the right shooter receives the right target at the right time,” he told War on the Rocks.
Carpenter emphasized that AI serves to assist analysts rather than replace them. That distinction matters, particularly for complex targets such as dual-use facilities that demand closer human scrutiny.
The threat picture got sharper over the summer. Russia deployed its new jet-powered drones against Ukraine daily throughout August, the first month that upgraded models appeared in Ukrainian Air Force records every single day. Monthly attack counts for those faster drones jumped from 350 in June to 1,600 in July, then 2,800 in August.
Cruising above 300 mph, these aircraft outrun Kyiv’s interceptor drones entirely. The weekly mass attacks, often involving hundreds of drones and missiles at once, explain why NATO planners now rehearse for saturation rather than isolated strikes.
During February’s Dynamic Front exercise, commanders set the operational bar: allied forces must track and strike 1,500 targets per day while intercepting between 600 and 1,200 missiles and drones. Col. Jeff Pickler said handling that scale in large-scale combat “will require taking advantage of automation and artificial intelligence technology.”
Meanwhile, the Germany-based 56th Multi-Domain Command used the same exercise to practice shooting down up to 1,200 ballistic missiles in a day. Army leaders are also exploring decoy drone fleets to confuse enemy targeting and high-altitude platforms for detecting electronic warfare systems.
A homegrown AI system, still in beta, helped sort battlefield data during a joint U.S. Army and NATO exercise this year. The system can assess available capabilities and recommend responses ranging from conventional strikes to electronic warfare and cyber operations.
Carpenter acknowledged AI’s strength with speed and scale, yet warned that “the challenge lies in creating adequate workflows to ensure the warfighter still validates the targeting data.” NATO’s parallel Next Generation Targeting experiment, tested during DiBaX26, showed a similar philosophy. Operators from 11 Allies processed over 6,000 simulated targets and more than 100 live ones, with human approval required at every engagement decision.
The stakes continue rising. Dynamic Front merges next year with Arcane Thunder to simulate electronic jamming alongside lethal munitions, suggesting the Army expects the worst day to keep getting harder.
