Predator: The Secret Origins of the Drone Revolution

PROLOGUE

Late in the afternoon of Wednesday, July 12, 2000, a bus carrying about a dozen high political rollers, as thirty-six-year-old Air Force Captain Scott Swanson viewed them, pulled up at Indian Springs Air Force Auxiliary Field, a broiling desert outpost northwest of Las Vegas. The visitors included the National Security Council’s deputy counterterrorism chief, a senior official from the CIA’s Counterterrorist Center, officials from other intelligence and military agencies, and the director of a shadowy Air Force technology shop known as Big Safari. Indian Springs was home to the 11th and 15th Reconnaissance Squadrons, units assigned to fly a relatively new unmanned, remote-control aircraft. Though the drone was equipped with nothing deadlier than daylight and infrared video cameras, it bore a menacing name: Predator.

Swanson, a former special operations helicopter pilot, had flown Predators for the past two years, mostly in regular intelligence, surveillance, and reconnaissance missions over the Balkans. But soon he was to join Big Safari as that outfit’s sole Predator pilot, which was why he had received a phone call from the Pentagon a few days earlier asking him to organize a show-and-tell session for those on the bus.

There’s going to be a bunch of people in suits, Swanson was told. Can you do a briefing on Predator and show them some capabilities in flight, but keep it quiet?

On the other end of the call was Air Force Colonel James G. Clark, whose official title was technical director, simulation and integration, Office of the Assistant Vice Chief of Staff, Headquarters, U.S. Air Force. In reality, Clark worked for the service’s two top leaders, the four-star chief of staff and the civilian secretary of the Air Force. He was their favorite fixer, an inside operator who was canny about how to bypass bureaucracy and who relished getting things done, as he liked to put it, quick and dirty. This was partly why Clark encouraged everyone to call him by his nickname, Snake—or, if regulations required they salute him, Colonel Snake. He consciously cultivated the image of a shrewd and slippery operator who might be dangerous if stepped on, a reputation he found useful in intimidating real or potential opponents. For the past three years, the Predator had been Snake Clark’s pet project, and for the past two years, Big Safari’s as well.

On the phone, Clark told Swanson the suits were coming to Indian Springs to get a better understanding of what they might expect from the Predator if it were used in a rugged part of the world. Clark slithered around the obvious question of what precise mission the government had in mind, but Swanson got the message. He read daily intelligence briefs and worked with intelligence officers, and he could guess which rugged part of the world, and even which country, was of such keen interest to these particular visitors.

From their air-conditioned bus, the delegation climbed down into the shimmering desert heat at Indian Springs and filtered into an air-conditioned briefing room, where Swanson gave them a PowerPoint presentation on the Predator and what it could do. Introduced six years earlier under a new type of rapid Pentagon procurement program, the Predator was the military’s first Medium Altitude Endurance Unmanned Aerial Vehicle—a drone that could linger in the air well beyond twenty-four hours, pointing cameras at the earth and transmitting live video images back to its operators. The little aircraft owed its phenomenal endurance to its unique configuration, a design informed by its inventor’s childhood hobby. The Predator had thin, tapered wings stretching fifty-five feet from tip to tip and a slender fuselage just under half that length, about five feet longer than a Piper Cub’s. Fashioned from lightweight composite materials, the fuselage was flat on the bottom, rounded on top, and bulged into a dome at the nose. With its flimsy wings and skin, the craft resembled a weekend hobbyist’s glider and couldn’t fly much faster. Powered by a four-cylinder engine akin to those used on snowmobiles, the small propeller on the plane’s tail could push it through the air at a sluggish top speed of just over eighty miles an hour.

Another reason the Predator could stay airborne for so long was that there was no pilot inside. Instead of a cockpit, the dome at the aircraft’s nose housed a satellite dish; its two-member flight crew remained on the ground. Seated before an array of computers and color video screens in a metal box resembling a freight container, they flew the drone by remote control, using a satellite data link. As they did, one of the two crew members, designated the sensor operator, aimed and manipulated a turret under the Predator’s chin that held two video cameras—one to shoot color images in daylight, the other able to produce infrared images by detecting variations in temperature, whether by day or by night.

As Swanson briefed them on the basics, his visitors interrupted with questions. Some wanted to know specifics about the Predator’s speed, range, and endurance. Some wanted to know if the video from the Predator’s cameras was sharp enough to identify an individual, and from what altitudes. Some wanted to know whether people on the ground could see the Predator or hear its engine. The visitors also wanted to know what tactics he would use to find a specific person and how hard that might be to do.

Swanson’s responses almost all began with It depends. But in order to better answer some of their questions, he invited his guests to follow him back out into the heat and down the Indian Springs flight line, where they could examine one of the faux freight containers the Air Force used to fly its small fleet of Predators.

To tour the twenty-four-foot-long, eight-foot-wide, eight-foot-tall ground control station, known to insiders as a GCS, the members of the party had to take turns, entering two at a time. Before taking anyone inside, Swanson suggested that while his visitors waited, they might try to spot or hear a Predator; one was orbiting a mile or two from the airfield, he assured them, launched a bit earlier by a crew inside the GCS. The sky was clear and it was quiet around the desert airfield, but no one would detect the drone.

Two by two, Swanson’s guests entered the GCS through one of two large, meat-locker-style doors on one side of the container that were opened by swinging a big lever handle. It was chilly inside the dark compartment, where constant air-conditioning was necessary to keep the electronic equipment from overheating. Even in the desert, some of the Air Force people working in the GCS wore jackets.

To their right as the visitors entered was the Predator’s equivalent of a cockpit, a pair of identical consoles flush against the front wall, each faced by a brown, mock-leather chair for the flight crew that looked more comfortable than it was. In front of each chair was a keyboard; on a metal rack above the keyboard at each console were two nine-inch screens side by side that displayed data about the aircraft. Above the nine-inch screens were two nineteen-inch monitors for each console, one stacked above the other. The lower monitor normally displayed imagery, the upper one a moving map, although what they showed could be changed.

Each console also featured controls much like those of a normal airplane—or a video game. To the right of the keyboard at each console was a joystick with several buttons on it. To the keyboard’s left was a lever that moved forward and back. The joystick on the left console normally governed the Predator’s control surfaces; that console’s lever was the aircraft’s throttle. The joystick and lever on the right console normally served as the controls for the Predator’s cameras—the stick was used to aim them, the lever to zoom their view in or out. Below the two consoles were pedals that moved the drone’s inverted-V tail like a rudder to help change the Predator’s direction in flight. When the drone was on the ground, the pedals worked as brakes.

Either console could be used to fly the Predator, but only one controlled the drone’s flight at any given time. Normally that was the left console, and the Predator’s pilot would sit in the left-hand chair; the right console was usually used to control the cameras, and the drone’s sensor operator would sit in the right-hand chair. But with the push of a red button, the functions of the two consoles could be switched so that two pilots could sit at the flight consoles, the second officer serving as copilot to assist during takeoff or landing.

Behind the flight crew were work stations holding computers, other electronic gear, and telephones used by intelligence analysts who typically worked in the GCS. Along the walls—which were covered with thin synthetic paneling of the sort often seen in mobile homes—were fold-down vinyl seats similar to the canvas ones in military transport planes, but without seat belts or shoulder harnesses. Toward the back of the container was a second meat-locker-type door, which, like the first, had to be opened and closed gently to avoid making noise that might distract those working inside.

As each pair of guests stood behind the flight console chairs watching the Predator’s video, the crew showed them the view through the color TV camera, useful only in daylight, then switched to the black-and-white infrared camera’s view. The sensor operator also toggled the infrared sensor between its two modes, white hot and black hot, in which warmer objects looked either white or black in contrast to cooler objects. The crew focused and zoomed in on vehicles driving around the airfield, on individuals walking from place to place—and then on the visitors standing just outside the door, waiting to enter the GCS. Swanson got the feeling the delegation was impressed; there were hushed conversations, whispers back and forth, and a lot of notes taken.

Not long after the guests left, Swanson learned that his guess about their reaction to the demonstration must have been correct, as was his hunch about the mission they had in mind. The Predator was going to play hide-and-seek in Afghanistan with one of the world’s most wanted men: the elusive leader of the Islamic fundamentalist terrorist group Al Qaeda, Osama bin Laden. And when it did, Scott Swanson would be the chief pilot.

Life with Big Safari was getting interesting in a hurry.

*   *   *

This is the story of the first armed drone ever to be flown by intercontinental remote control and used to kill human beings on the other side of the globe. The military has long had an interest in unmanned aircraft, but before the Predator, drones were at best a niche technology. The Predator itself was widely ignored at first, until a series of iconoclastic visionaries began transforming it from a simple eye in the sky into an exotic new weapon. Once the Predator became capable of firing laser-guided missiles at enemies half a world away, military and industry attitudes toward such unmanned aerial vehicles changed nearly overnight. The drone revolution began.

How and why that happened is a tale previously told only in dribs and drabs, and often inaccurately. This account is based on five years of reporting and hundreds of interviews with the insiders who made the Predator what it became—an invention that not only changed the military, the CIA, and warfare itself, but also led the way into a new technological age. Drones of all kinds are now poised to transform civilian aviation, law enforcement, agriculture, and dozens of other human endeavors.

This is the drone revolution’s book of genesis, and like another creation story it opens near the confluence of the rivers Tigris and Euphrates. It begins with a boy in Baghdad.

1

THE GENIUS OF THE GENESIS

He was a born engineer. From the time Abraham Karem was a toddler, he was always drawing things and making things and taking things apart to see how they worked. At age two he pulled the back off a large standing radio, the sort owned by every middle-class Jewish family in Baghdad in 1939, and pulled out the big glass vacuum tubes one by one, looking for the talking man inside. Abe cried when a technician came and put the tubes back. Not long afterward, he became intrigued by the magic of electrical switches—click, the light went on; click, the light went off! One day he climbed on his uncle Ezra’s bed, found a round, brown light switch on a cord beneath the pillow, and took it apart, getting a 220-volt sting he seemed to regard as more interesting than painful.

When not engaged in this sort of basic research, little Abe was usually building a toy of his own design out of cardboard, or drawing something with a pencil or crayon on a piece of paper. His parents proudly encouraged their little prodigy. When Abe was eight, they gave him and his oldest brother, Isaac, each a set of Meccano, a British construction kit much like the Erector sets popular in America. Meccano provided perforated strips and plates of sheet metal; miniature girders of different sizes; wheels, pulleys, gears, axles, nuts, bolts; and instructions for using the parts and pieces to make models of all sorts. You could put together little buildings—or ships, bridges, cars, and trucks—or you could design and construct your own creations. In 1937, the year Abe was born, Meccano offered eleven different sets, numbered 0 through 10, according to rising levels of difficulty. Their parents got Abe a No. 2 and Isaac, who was seven years older, a No. 7, but Isaac quickly grew frustrated and abandoned his kit. Before long, Abe appropriated the No. 7 and combined it with his No. 2 to create toys as complex as those in Meccano No. 10.

By the time he got the Meccano kit, Abe was sure he was going to be a mechanical engineer. But as a teenager, he acquired a new ambition. His family had immigrated to Israel in 1951 as part of an exodus of roughly 120,000 Jews from Iraq, which, with other Arab League nations, had tried three years earlier to wipe the newly declared Jewish state off the map before it could take root. Abe’s father, a prosperous textile merchant named Moshe Kiflawi, already owned land in Israel. He had taken his wife, Flora, and their boys to Jerusalem during World War II, when the British still governed what was then Palestine, seeking a safe haven for his family after a June 1941 pogrom in Iraq that left 130 Jews dead and hundreds injured. Forced by the British to return to Iraq at war’s end, the family was relieved to get out again, even at the cost of forfeiting all their Iraqi property to the regime in Baghdad.

Once back in Jerusalem, Abe—who as an adult would change his surname from Kiflawi to the Hebrew for vineyard, to make it sound Israeli—was eager to fit in. Already fluent in Hebrew, he found schoolwork easy and had plenty of time for the nonprofit youth clubs that abounded in newborn Israel, which was heavily influenced by socialist attitudes of collectivism. Diminutive, intellectual, and cursed with flat feet that made it painful to run, Abe initially joined a chess club and an electrotechnical club. Then he discovered the Aero Club of Israel, where a young adult counselor was teaching members to make model gliders that could fly. When Abe built his first glider and saw it rise into the air, his heart soared with it. Within a year, he was flying models in competitions. Within two, he was the instructor for his Aero Club chapter. He also now knew what he was truly going to do with his life. Mechanical engineering wasn’t it after all. He was going to be an aeronautical engineer. He was going to spend his life designing aircraft.

*   *   *

On October 26, 1973, a Friday, Abe Karem was clearing his desk for the weekend when an unexpected visitor burst into his office at Israel Aircraft Industries, whose manufacturing and modification facilities lay on the north side of Lod International Airport, near Tel Aviv. Colonel Ezra Beban Dotan didn’t need to introduce himself, and not because he was a famous fighter pilot. He and Abe had known each other for years, first as Aero Club members, then working together as young majors in the Israeli Air Force. They had a lot in common, and they were entirely different.

Now a small, pale, baby-faced thirty-six-year-old, Karem had made his mark in the Air Force by leading teams of engineers in quick-reaction fighter plane modifications that three times won the Israel Defense Prize, the military’s highest honor for technical achievement. Joining the Air Force in 1961 after earning his aeronautical engineering degree at Technion, the Israel Institute of Technology at Haifa, Karem served until 1969. He rose quickly to the rank of major and was so respected as an engineer he could get the Air Force commander in chief on the phone—a privilege that may have spoiled the young genius, who later in life, preferring to deal with those at the top, developed a hazardous habit of going over other people’s heads.

At government-owned Israel Aircraft Industries, the most important aerospace company in the country, Karem was director of preliminary design—IAI’s director of innovation, in effect. He was known as a brilliant engineer, but he was also known for knowing he was a brilliant engineer—and for being impatient with those who weren’t also above average. Precocious subordinates often became acolytes. Those who weren’t, or who just weren’t team players, often found themselves quickly out the door. Abe Karem simply refused to work with anyone he didn’t respect. He also didn’t mind telling subordinates exactly what he thought of them, and he could let them go with as little apparent regret as Beban Dotan downing an enemy in a dogfight.

Physically, Dotan dwarfed Karem. He was big, muscular, tanned, and known not only as an audacious fighter pilot but also, within the Air Force, as a congenial commander. Nicknamed Mr. Skyhawk, from the moniker of the U.S.-made Douglas A-4 fighter planes he flew, Dotan was celebrated for downing five Syrian MiGs—including two in a single May 12, 1970, air battle in which he killed one MiG-17 by the unorthodox means of firing an antitank version of the unguided Zuni rocket. He was also a fan of Karem’s, for Dotan appreciated his friend Abe’s brilliance as an engineer; indeed, that was why he had come to see him this Sabbath eve.

Dotan dispensed with pleasantries.

Abe, why don’t you give me a Zuni rocket with a dihedral wing? he demanded without even sitting down.

Ezra, Karem softly scolded, motioning to a chair in front of his desk, don’t tell me what I need to give you. Tell me how you’ll use it.

I want to fire it from my aircraft so they’ll turn the radars on and shoot missiles at it, Dotan said.

He didn’t need to explain further why he wanted to add a wing that was dihedral, a common shape on airplanes, to a Zuni, a tubelike munition just five inches in diameter. Nor did he need to explain why he was in a hurry to get it. For the past twenty days, and for the third time since the creation of the Jewish state twenty-five years earlier, Israel had been at war with its neighbors Egypt and Syria, but with a frightening difference. Israel’s Air Force, which six years earlier had dominated the skies and been a key to victory in the so-called Six-Day War, had been rendered nearly impotent this time by surprising new enemy air defenses.

The Egyptians and Syrians had invaded from two sides on October 6, choosing the most important holiday in Judaism, Yom Kippur, to launch their surprise attack. Egyptian tanks and troops flooded across the Suez Canal. Syrian forces joined by Iraqi and Jordanian troops pushed Israeli forces back to the edge of the Golan Heights, which Israel had captured from Syria in the Six-Day War. After initial setbacks, the ground troops of the Israel Defense Forces repelled the invaders, establishing a bridgehead on the Egyptian side of the Suez and holding the Golan Heights. Under pressure from their respective superpower allies, the United States and the Soviet Union, Israel and its foes had signed cease-fire agreements two days before Dotan came to Karem. Whether or not those bargains held, though, the Israeli Air Force was going to need a way to counter those Arab air defenses in the future.

In the four short weeks of what history would dub the Yom Kippur War, Israeli air losses had been devastating. Mobile batteries of Soviet-built SA-2, SA-3, and SA-6 surface-to-air missiles, or SAMs—supplied to Egypt and Syria by Moscow, operated with the help of Soviet advisers, and supplemented by thousands of advanced antiaircraft guns—had cost Israel not just one hundred warplanes but also its far more precious pilots. The SAMs had also inflicted untold Israeli casualties on the ground by making it difficult for the Air Force to provide Israeli ground troops effective close air support with attacks on enemy land forces. Using the Soviet-led Warsaw Pact’s tactics and equipment, the Arabs had erected a five-layered air defense umbrella consisting of long-range radars and missiles, short-range radars and missiles, and fighter aircraft. The long-range radars, positioned beyond Israel’s reach, would paint an Israeli fighter as it approached Arab forces and transmit the aircraft’s position to a short-range SAM battery lying in ambush. The short-range battery would then turn its radar on just long enough to target the Israeli plane and fire a SAM. Within seconds, a missile traveling nearly three times the speed of sound would be homing in on the Israeli aircraft from a short and deadly range. As its missile flew, the SAM battery would douse its radar and scurry away to avoid being targeted by Israeli planes armed with missiles that homed in on radiation.

Dotan had just spent days flying risky observation missions over the battle lines in a desperate search for ideas on countering the enemy tactics, which was why he was in Karem’s office that evening. What Israel needed, Dotan told Karem, was a decoy to fool those SAM radars, and the radar image of a Zuni rocket with a dihedral wing ought to look enough like an airplane to do the trick. As an Israeli fighter jet neared the enemy air defense umbrella, its pilot could launch a winged Zuni. The enemy’s long-range radar would see the rocket as a manned aircraft and signal a short-range SAM battery to attack it. When the SAM battery turned its radar on to target the Zuni, Israeli planes with antiradar missiles would detect the SAM radar and fire at the enemy missile battery while the SAM battery pointlessly fired on the decoy.

Karem silently pondered Dotan’s idea for a moment, the roar and whine of jets on the runways outside the only noise in the room. Then he waved his friend off.

You don’t need a rocket and you can’t afford the dihedral, Karem said. You do need a decoy, but it can’t be a Zuni. Karem knew the Zuni’s radar signature was too small to mimic that of a fighter plane. Invite me for lunch tomorrow, he told Dotan. By then, Karem promised, he’d have a solution.

After Dotan left, Karem called his wife, Dina, to tell her he wouldn’t be home for dinner that evening after all, news she was used to after seven years of marriage. Karem stayed at the office through the night, feverishly working the engineering challenge Dotan had given him. The next day, promptly at noon, he was on the doorstep of Dotan’s home in a suburb north of Tel Aviv. Under his arm was a set of drawings produced during his all-nighter: a design for a winged decoy about the size of a small target drone, but with special radar-reflecting spheres on its sides to make it look big to SAM radars. Light enough to hang under a manned jet’s wing, Karem’s decoy would be unpowered but aerodynamically shaped so that when released at thirty thousand feet it would glide at a plausible fighter plane speed of Mach 0.85—about 575 miles an hour at that altitude—into the enemy air-defense umbrella.

The next morning, Dotan took Karem’s glide decoy idea to the commander in chief of the Israeli Air Force, Major General Benjamin Benny Peled. By midday, IAI had Air Force approval to build some prototypes as a quick-reaction project. The ranks of IAI’s engineers had been thinned by the war, as many were reservists called up by the military, but Karem assembled a small team from those available and started work immediately, pushing his group to keep at it nearly around the clock.

A week into the project, someone came back from a trip abroad with a copy of a magazine article describing a new U.S. Air Force radar decoy with folding wings that had just been flight-tested successfully. Karem urged the Israeli Air Force to consider trying to get the U.S. decoy instead of having IAI produce his, but he was instructed to carry on with his work. Four weeks later, his decoy made its first test flight. Over three more weeks, seven further tests were flown. Not all were successful, but the prototypes flew well enough that Karem’s preliminary design work was largely done, which allowed him to turn his attention to other things.

Soon after his work on the decoy, Karem decided to take a major gamble with his career. For most of his decade and a half as an aeronautical engineer, he had focused on the problems and possibilities of manned fighter planes and transports. But for the last couple of months, he had spent twenty hours a day thinking about what amounted to an aircraft with no pilot inside. There was nothing new about pilotless planes; inventors had designed them in a cornucopia of configurations since the First World War. By the 1970s, Jane’s, the authoritative military publishing company, listed 120 separate types of pilotless planes in a Robot Aircraft Today pocket guide to what experts of the day usually called remotely piloted vehicles, or RPVs. Yet except for radio-controlled target drones, few unmanned aircraft had been adopted by the world’s militaries, in large part due to poor reliability. As Karem knew, aircraft without pilots on board tended to crash a lot more than those with pilots inside them. But this decoy project had taken him down a novel mental path, and before long that path would lead him on a journey of discovery, a pioneering exploration of an aviation frontier.

*   *   *

Only two months after finishing work on the decoy for the Israeli Air Force, Karem stunned his bosses and most of his coworkers at IAI by quitting his job as director of preliminary design. He left to start his own company, a move he had been mulling over since before the Yom Kippur War. After only three years at IAI, Karem had become deeply disillusioned with the way things were done at the massive aerospace company. All IAI’s stock was held by the Israeli government, and in Karem’s view government officials and many IAI executives treated the company more like a jobs program than a corporation. They hired far more people than necessary; the way Karem saw it, IAI’s employees did less work in more time at higher cost to the taxpayers than was warranted. He believed in his bones that in most fields, especially aeronautics, the best work and the best ideas were produced not by large organizations—especially those influenced by politics—but by small teams of talented people working hand in hand toward common goals.

Karem’s belief in teams stemmed from his youth, when Israel was still a vibrant work in progress, a magnet for Jewish idealists, and a dynamic dream coming true for Jews the world over. He fondly remembered how, in those days, younger members of the fresh and still fragile Jewish state teemed with plans and hopes for building a new society, a fundamentally good civilization, and a sanctuary for the Jewish people after centuries of persecution and a Holocaust whose extent was still being revealed. His high school home room teacher for four years, a tail gunner in a British bomber during World War II, preached that the Jews would survive and secure their freedom only if they lived one for all and all for one. The young counselor who advised Abe’s chapter of the Aero Club stressed the importance of members helping one another with their designs. At IAI, Karem had been a rapidly rising star, but after proving several times that he could do major work faster and better with fewer people and save the taxpayers money at the same time, he found his efficiency rewarded mainly by resistance. One executive even reproached and threatened him for telling the Air Force that the modification of a particular fighter plane could be done in one year instead of the three IAI had estimated, saving two million out of the three million man-hours of labor the company was planning to commit to the project.

By January 1974, Karem was fed up, and after running the idea past friends in the Air Force he announced that he was leaving IAI in May to start his own company. After he invited a handful of his favorite engineers to come along, an IAI vice president called Karem into his office.

Abe, we all love you, the man said. We owe you a lot. But we had a meeting about you forming your own company, and we decided we are going to crush you. So don’t do it.

Why? Karem asked. You are seventeen thousand people, and I’m going to be what, five hundred? At the time, Karem hoped to build a company that would be about that size.

Size didn’t matter, the vice president said, but every time we’re going to propose something, they will say, ‘But Abe will do it faster and cheaper,’ and we’re not going to have that.

As Karem later learned, the warning was friendly, but the threat wasn’t idle.

*   *   *

The clash between Karem’s brash ways and the realities of IAI kindled his decision to leave the company, but something else was at work as well. The task of designing that radar-tricking glide decoy had led Karem to look at things in a new light. When he did, he had an epiphany.

The decoy he designed for IAI was primitive; in fact, it would have been less capable than the remote-control target drones used since the 1930s to train fighter pilots and antiaircraft units. But while thinking about the problem the decoy was meant to solve, Karem realized that an unmanned aircraft with the right capabilities could do far more than merely trick SAM batteries into revealing their locations. A remotely controlled drone armed with antitank missiles and designed to loiter in the sky for hours at a time could be one way to defeat—or, better yet, deter—another invasion of Israel.

Geography had forced Egypt’s tanks to mass as they funneled through holes blasted in defensive embankments to cross the Suez Canal and enter the Sinai Peninsula. Looking at that, Karem recalled years later, I said, if we are right on top of this high concentration of forces, you can throw some missiles at them. They will say, ‘This is not a good day,’ and back off, and we are not killing that many people. You let them spread, they throw their armor against your armor, their air force against your air force, and all of a sudden you have thousands and tens of thousands of casualties.

Pilots add weight to an aircraft, and more weight requires carrying more fuel; besides, pilots need to land every few hours to rest. But a fleet of pilotless aircraft such as Karem was imagining would have the persistence—the flight endurance—to provide an air-to-ground defensive missile system guarding Israel’s borders day and night. Moreover, if the enemy shot one down, no pilot would be lost. Achieving the necessary flight endurance would be one of the hardest parts of his challenge, but Karem was sure he could design such a drone. For one thing, he believed himself not only the best engineer in aeronautics but probably the best engineer of any kind in all of Israel. Beyond that, he had been a model aircraft hobbyist since his teens, and as a modeler his specialty was a type of aircraft whose sole objective was endurance.

Thrown or towed into the air like a kite on a fifty-meter-long string and then released, free-flight gliders are the oldest form of model aircraft and among the trickiest to build. A schoolchild’s paper airplane technically fits the definition, but in competitions the models are far larger—wingspans of six feet are common—and more sophisticated. Model gliders are built of lightweight material, such as balsa and tissue, and usually weigh a pound or less. The goal is to make a plane that is capable of staying airborne long after its launch or its release from the tow line, but without any remote or automated controls to help keep it aloft.

Like the materials used to build it, a tow-line glider’s flight characteristics are the opposite of a jet fighter’s. A model glider typically floats through the air at two or three miles an hour or less, gently circling until aerodynamic drag and the law of gravity return the aircraft to earth. In international competitions, the winning aircraft is the one that accumulates the most flight time over a series of launches within a strict limit of three to three and a half minutes per flight. The time limit and multiple rounds prevent winning or losing by the luck of the wind, though catching a thermal of the sort birds often ride is a favorite technique for success.

A good free-flight glider is rugged enough to withstand turbulence and return to stable flight or take a hard landing, yet light and aerodynamic enough to fly as close to three minutes as possible even in calm air. Most carry a dethermalizer, a device operated by a timer or activated by radio signal, which moves a control surface to put the plane into a deep stall—a sudden dive—and bring it to a gentle landing. Deep stall is used after a round’s maximum flight time has been achieved, or to escape a thermal that threatens to carry the model away.

The best free-flight modelers employ all sorts of tricks and techniques to improve endurance, and Karem was a world-class free-flight modeler. Free-flight gliders were what Abe had learned to build in the Aero Club, and he pursued the hobby into adulthood. In August 1963, as a twenty-six-year-old Air Force officer, he placed tenth in his category representing Israel at the free-flight World Championships in Wiener Neustadt, Austria.

Two years later, Abe met Dina Schleiffer, a petite, lively, smart conscript who showed up one day in the Air Force engineering spaces, assigned to fulfill her obligatory two years of military service drafting technical drawings for engineers. Only eighteen, she was ten years younger than Abe, but he found her much more mature and confident than older girls he knew. They started dating, and soon Dina was traveling with Abe to free-flight competitions. She loved the little planes he designed, especially one he put together just as a spare and called his ugly duckling. She loved helping him launch his gliders when he competed. Most of all, Dina loved Abe, and in 1966 they married. For the rest of their lives, she would sometimes accuse him of tilting at windmills, but she always helped him in his work and backed him when he took major risks.

A month or two after leaving IAI, Karem set up a company of his own called Matos (Hebrew for aircraft) and began work on his idea for an armed drone to patrol Israel’s borders. Friends in the Air Force counseled him to be less ambitious in his design; he should just make a reconnaissance drone with a TV camera that would send imagery to forward air controllers (officers on the ground who direct air strikes) rather than arm the RPV itself. The Air Force commander, Benny Peled, himself an aeronautical engineer, was even less encouraging. Peled told Karem that he wouldn’t buy such a drone for the Air Force because its mission was fundamentally an Army job. Appalled, Karem pointed out that if he created a pilotless aircraft for the Army, it would have to be a helicopter, so that it could operate from wherever the Army was stationed. But a helicopter flies on a column of thrust created by rotors whose blades turn hundreds of revolutions per minute to keep the craft airborne, a fuel-thirsty and violent way to fly compared to airplanes. As Karem reminded Peled, helicopters shake like hell and have lousy endurance. Karem was also aware that others had taken on this challenge with unsatisfying results. In the late 1950s, the U.S. Navy built and deployed an unmanned helicopter to carry antisubmarine torpedoes; in the 1960s, the Navy added a TV camera and various weapons, and used the aircraft in Vietnam for reconnaissance and attacking enemy convoys moving at night. But in 1971 the Navy cancelled the program after half of the 810 unmanned helicopters that had been built crashed. Most of those that remained were used as target drones.

Unhappy with Peled’s response, Karem told the Air Force commander that designing a drone helicopter for the mission he envisioned would be excruciatingly difficult.

That’s why you’re Abe! Peled cheerfully replied.

Karem came up with a preliminary design, but the Army declined to buy it. So Karem went back to his drawing board, as he would do nine times more over the next three years, without ever making a sale. He ultimately concluded that because of IAI’s political influence, the Israeli military was never going to buy anything from his new company. Former colleagues confirmed that IAI was indeed out to kill his company; his friends, meanwhile, told him he might as well go to the United States. If he did, no one could accuse him of betraying Israel by leaving—not after all he’d done for the