Inside the Air Force's Secret Drone Pilot Training Classroom

Learning how to drop bombs and fire Hellfire missiles is more like sitting in a regular college classroom than you might expect. There are hundreds of pages of text to digest, continual testing of knowledge, and nervous, eager students anxious to please their instructors. I know, because I spent a week at the Air Force’s drone pilot training school last year.

We were sitting in a cramped briefing room in the 9th Attack Squadron’s new headquarters at Holloman Air Force Base in New Mexico, across the parking lot from its former, dingier quarters. The furniture was a motley assortment of old and new chairs, tables, and couches that had been scavenged from across the base. Like every building at Holloman, the headquarters was tan, mirroring the surrounding desert landscape. In the briefing room, the walls and ceiling tiles were white; a large whiteboard adorned the south wall. Only a hanging video screen disrupted the color scheme.

Patrick, a senior instructor who spent years piloting drones and engaging in “super secret squirrel stuff” in Afghanistan, led the class. Tall, angular, and pale, with a jutting nose and a wide, quick smile, he was relentlessly energetic, alternately instructing and cajoling the students. “The first thing to think about,” Patrick told the class’ two students—Paul, a pilot, and Justin, a sensor operator—“is the intent of the attack: what does the attack controller, or whoever’s in charge, want to happen on the ground?” The Air Force requested that I use first names only in exchange for weeklong access at Holloman as part of my research for a book on the future of warfare.

And the future, I learned, is like the past: In matters of war, there is tension between what members of the military feel is right and what their work requires. I observed this in the discord between trainers' rhetoric about how much they disliked killing people—they repeated this to me frequently—and their unabashed excitement, also expressed frequently, about the times they were able to launch strikes and kill "bad guys." Hating killing, but enjoying the chance to kill. The competing impulses may have seemed irreconcilable, but they were everywhere.

* * *

About an hour into class, Patrick told his students that different units would allow them different degrees of control over their attacks. “Sometimes you’ll be handcuffed,” he said. “They’ll say, ‘Run in from this specific angle, and drop a guided bomb unit on that specific point of impact.’” Other times, he said—and this is more enjoyable—“They’ll go, ‘I want you to kill those guys right there,’ and you’ll get to tailor your options to what makes sense to you.”

“So what makes sense?” Patrick asked Paul and Justin. “If you want to disable a car, or kill the people inside, you might have different aim points, right?”

Paul, a former security forces officer with experience in Afghanistan, was square-jawed and broad-shouldered, a classic linebacker physique, and patchy brown hair. Drone pilots saved his life at least four times in Afghanistan, he said, and he wanted to return the favor. He’d only had a six-day break between his last deployment and drone pilot training school.

Justin had joined the Air Force partly for money—he was in the third year of his bachelor’s degree in business, and already $20,000 in debt. An intelligent perfectionist, he talked proudly of his 4.0 GPA. (You could watch him berate himself in class whenever he made a mistake.) He styled his hair and wore expensive black-frame glasses. 

“To disable the car?” Justin said. “I’d go through the engine.”

“What if we want to kill the passengers inside the car?” Patrick asked.

Paul thought for a moment. “It depends on which person you’re trying to kill,” he replied. “But if it’s all of them, I’d aim for the center mass of the roof.”

Patrick gave an emphatic yes. “And roofs are so small,” he said, “that most people will tell you to go center mass of the roof no matter what. Even if there’s a high value individual in the back left seat, the roof is so small, you want to make sure your sensor hits it. Because if he slips off to the side and he lases glass, what’s going to happen?”

Paul started to answer: “The laser’s gonna be reflected all over the glass and—”

Patrick jumped in. “It ain’t gonna work, right? So if all of your high-value individuals are inside, hit the center mass of the roof and you’re gonna kill everybody in there. The car’s gonna blow up.”

This wasn’t always the goal, he reminded the students. “If you’re just killing one HVI, say if the people get out of the vehicle, make sure you know which one is important. If you don’t have a specific person, go for the biggest group.”

Patrick’s lesson pointed to the less-than-ideal ground intelligence upon which drone pilots sometimes operate. There is an assumption that any individual in “BadGuyLand” (Patrick’s term) is, well, a bad guy, and therefore a potential open target. For all the talk of precision strikes, in other words, war is still war. Next up was a discussion of the weapons at each drone operator’s disposal. “Give me a quick run-down,” Patrick said. “What does a Papa do?”

Paul had his weapons down pat. “A Papa is designed to kill tanks and armored vehicles,” he answered. “So you’re gonna use that on people in the open. It’s better to use it on movers, as well.

“A GBU-12,” Paul continued, unprompted. “is a larger laser-guided bomb. Me, personally, if I can use that, it’s more of a sure thing. So if collateral damage isn’t a factor, I’m gonna use that.”

“So he’s gonna have time to run,” Paul said.

Patrick moved on to the Hellfire missile, the weapon of choice against people who are moving in the open. He directed his first question to Paul: “If you’re shooting a person walking around,” he asked, how should you go about it?

Paul took a moment to think. “Do I have time constraints on the shot?”

Patrick was dismissive. “There’s always a time constraint,” he said.

Paul adopted a firmer tone. “I’d probably try to shoot more towards the front,” he said, “because if I’m shooting off to the side, the missile might end up hitting behind him or off to the side, and not actually kill him.”

“Absolutely,” Patrick said, his impatience gone. “Placement matters a lot.”

“What’s the other one big factor?” the instructor prodded.

“The sonic boom time,” Paul answered.

“Yes,” Patrick said, with even more enthusiasm than usual.

This was one of his favorite topics as an instructor. An understanding of sonic boom time is, he said, what separates mediocre drone pilots from skilled ones. “If I shoot 90 degrees off azimuth, what’s my sonic boom time going to be?” (Azimuth refers to the angle between an object and a point of reference. In other words, it's the angle at which the pilot aims the laser that guides the bomb or missile to the target he wants to hit.)

Paul was silent.

“Eight to 10 seconds maybe?” Patrick offered. “It’s a long time.”

“So he’s gonna have time to run,” Paul said.

Patrick nodded vigorously. “That’s a fun one, right?” he said. “You shoot, the boom happens, and then they run, from the boom time on, and you have to track them.”

To drive the point home, Patrick engaged in a brief exercise. “BOOM!” he shouted, then counted silently to eight. “Okay, there’s the impact.”

He stood still for a moment, looking at Paul and Justin to make sure they understood the importance of this lesson. “You’re gonna run pretty darn far if you hear the boom, and you know you’re a bad guy, and you know good guys are overhead,” Patrick said. In Afghanistan and elsewhere, he pointed out, “The tactics have spread, and everyone knows that if you hear a boom, you run, because you might live.”

One of a drone operator’s primary goals, of course, is to make sure targets don’tsurvive. This was now being complicated by the fact that, after years of drone strikes, the enemy had learned to adapt. More and more “bad guys” were becoming increasingly adept at beating the system.

Patrick then offered a solution. “If you shoot 20 off azimuth,” he asked, “what’s the sonic boom now?”

“It’s gonna be like one-and-a-half to two seconds,” Paul said.

“Yes. Two seconds,” Patrick nodded.

“They’re not gonna have time to do anything,” Paul said.

“Yes,” Patrick said, pleased. “So you’re making your sensor’s job way easier. If it’s an eight-second boom? Man. I’ve seen so many squirters run and be missed.”

Squirter is a favorite term in the drone operator’s lexicon. It refers to anyone who runs—who “squirts”—from the scene of an explosion. Patrick moved on to his final point before the group headed out to the drone trailers to practice.

“Okay,” he said, “where are you going to aim for a dude? One single dude?”

Justin was quick in his response: “You aim the crosshairs right at the bottom of his feet.”

Patrick directed another question to Justin: “How about a group of people—where would you aim? No high value individuals, just a group of bad dudes.”

“Dead center,” Justin replied, “for that 360 effect, for the fragmentation.”

“The frag goes in every direction,” Patrick affirmed. “Very low probability of someone escaping.”

“Or you could drop a GBU [a large laser-guided bomb],” Justin said.

Patrick liked this. “Or a GBU. Absolutely.”  He smiled for a moment, then turned serious.

“Unless there’s a baby milk factory right next to it,” he added.

Justin chuckled. “Yeah, we don’t want that.”

* * *

“I remember the very first time I struck in a B-1 bomber,” Patrick said one afternoon, when I asked him if there was any difference between dropping bombs from a regular airplane and dropping bombs from a drone. “It was my first combat sortie. This was in Afghanistan. We checked in, and they said, ‘Hey, we have troops in contact and you’ve got to fly over here.’ So we flew, like, 500 knots, and by the time we got there, they were ready for the bombs and they had given us clearance. And we dropped bombs right away. And it was like, ‘Whoa! Super exciting! Oh my gosh, my first combat sortie!’

“And then my very first strike in a [drone]—I had nearly identical feelings inside.”

For that first drone strike, like every other drone mission he flew, Patrick was at Creech Air Force Base in Nevada, thousands of miles from the people he was targeting.

The truth about flying drones in the Air Force, Patrick said, was it that “it can be relatively boring on the normal day-to-day grind.” About 97 percent of a drone operator’s job—the boring part—involves what the Air Force calls ISR: intelligence, surveillance, and reconnaissance. The other 3 percent of the job involves what Patrick called the “cool” or “exciting” part: dropping bombs and firing missiles. This is the job that drone operators wait for, and that wakes them up no matter how sleepy or dulled they are from the surveillance work on their shift.

Imagine, Patrick says, you’re working the midnight shift, and that you’ve just come off your weekend. “It’s a lot harder to engage and get woken up” if you’re transitioning from a day off to your reconnaissance duties, he admitted. But if you arrive on your shift, he said, and “they’re about to strike, no one has any trouble waking up. Because everyone’s excited to be there. Even if it’s the middle of the night and you’re in the middle of nowhere.”

At the end of a shift in which a strike has been performed, there is an official debrief, which can last for 30 minutes to an hour-and-a-half. In the debrief, the pilot and sensor analyze everything that went wrong and everything that could have been done better. Patrick, who is working on his master’s degree in entrepreneurship, adopted the language of corporate America in describing the reason for the debriefs: “We try to get better every single time so you’re not just on cruise control and giving the same product. You’re trying to give better and better solutions to the ground commanders every time.”

After the morning class on dropping bombs and firing missiles, Patrick, Paul, and Justin headed out into the searing New Mexico sunlight for a quick walk across the parking lot. To the right was a massive wood-ceilinged hangar where scenes of Transformers: Revenge of the Fallen were filmed. This was the part of the day that everyone had waited for: the chance to fly, to pilot a $14.7 million MQ-9 Reaper over the acres of scrubby desert between Holloman and White Sands. 

To do so, they crowded inside a dim, chilly, cramped trailer where they sat haloed in the queasy light of 14 computer screens. To an outsider, the next several hours were a mind numbing series of practice runs as Patrick guided Paul and Justin through every possible missile- and bomb-strike permutation. To the students, however, this time was not only exciting, but it utterly high-stakes. However well they performed in their classrooms, their ability to operate a drone was the only skill that would ultimately matter. For both Paul and Justin, that ability improved markedly over the course of their time in the trailer. When it was time to head back to squadron headquarters for a debriefing of their performance, they felt confident that, after a rough start, they finally got the hang of things. And indeed, that’s what Patrick told them in the debrief.

Problems remained with killing squirters, however. And those problems highlight the inherent tension that comes with drone training. For all the precision that such technology allows, its fundamental purpose is the same as with any weapon: Drones are designed to kill. But throughout Paul's and Justin’s practice runs, many of the squirters, Patrick said, were able to escape. The fault lay with Justin: he was moving the sensor too slowly when tracking the squirters, so Paul’s shots missed their marks. Justin was immediately defensive. Maybe these were particularly slow-running squirters, he said. Patrick looked dubious. Eager to prove his point, Justin doubled down, saying, “They might’ve been fat!” 

This time it was Paul who shot him a doubtful look. Almost apologetically, he told Justin, “There aren’t a lot of fat Afghans.”

Patrick broke into a big smile. “That’s true!” he says. “These guys can go 10 to 12 miles per hour in a sprint—remember, they’re running for their lives.”

Running isn’t the only thing squirters can do to avoid being killed, Patrick added. Depending on the trajectory of the missile, a squirter can simply crouch or lie down.

If that happens, he said with a frown, “You might just blow out their eardrums.”