Image: Lockheed Martin ©
Lockheed Martin Skunk Works, together with the U.S. Air Force Test Pilot School and industry partners, has demonstrated sensor-driven autonomy on a fighter aircraft for the first time. An artificial intelligence agent used data from an operational sensor to carry out successful air intercepts against a live target, as per a recent press statement.
Closing the Loop Between Sensors and Combat Action
Across eight flights, the X-62 Variable In-flight Simulation Test Aircraft completed 27 AI-controlled intercepts. The aircraft, fitted with Lockheed Martin’s Legion Pod, tracked a live T-38 jet and fed secure data to an onboard AI agent. That agent then autonomously piloted the fighter into a tactical intercept position, marking a shift from simulated targeting data to real-time sensor streams.
Skunk Works’ “Supermassive” AI agent generation capability drove the pace of the programme. Engineers integrated and ground-tested the agents with the X-62 in just three months.
Industry and Air Force Voices
Ron Fehlen, vice president and general manager at Lockheed Martin Skunk Works, commented that the trials show the company’s AI can reliably close the sensor-to-action loop aboard an operational combat aircraft. He added that the autonomous agents processed classified infrared search and track feeds while executing combat-critical manoeuvres in real time.
Stacy Kubicek, vice president and general manager of Lockheed Martin Sensors and Global Sustainment, said reliable sensor data only delivers its full advantage once linked seamlessly to AI-driven action.
What Comes Next
Skunk Works has partnered on the X-62 programme for decades, supplying open software and hardware architectures for flight testing. The aircraft’s forthcoming Mission Systems Upgrade will integrate combat systems, sensors and airborne AI agents within a next-generation mesh network, building on this test framework.
Learn more on Lockheed Martin’s website
