The well-intentioned author of the article "The Killing Ma-chines," which appeared in the Atlantic last year, offers a lengthy description of a Hellfire missile strike by a remotely piloted aircraft (RPA). The story's protagonist, a "19-year-old American soldier" who entered Air Force basic military training straight out of high school, became an MQ-1 Predator crew member upon graduation. Reportedly, on his very first mission at the controls, the "young pilot" observed a troops-in-contact situation on the ground. The "colonel, watching over his shoulder, said, 'They're pinned down pretty good. They're gonna be screwed if you don't do something.' " The narrative goes on to describe the Hellfire missile strike and the psychological effect it had on the Airman. To a sophisticated military audience, the factual inconsistencies in this account are apparent. Air Force RPAs are crewed by Airmen, not Soldiers. The 19-year-old Airman (an enlisted rank) cannot be an Air Force pilot (an officer rating). The article claims that during his first time at the controls, this Airman finds himself on a combat mission in-theater. In reality, he would have become familiar with the controls at initial qualification training, prior to arriving at his first combat squadron. Furthermore, when colonels speak to Airmen about life-and-death combat decisions, they tend to do so in terms of direct orders rather than leading suggestions. How can Mark Bowden, notable historian and author of such well-received books as Black Hawk Down, commit such factual errors? The answer is simple. Information about Air Force RPA operations is rarely available-and when it is, it usually proves unreliable. This article contends that because an information vacuum exists with respect to US RPA operations, well-meaning people cannot gain adequate knowledge to develop and share an informed opinion on the most important RPA questions. It calls this dearth of information "the epistemic problem."