The Coming Revolution of Drone Warfare

New technologies will allow many states—and nonstate actors—to make low-cost but highly credible threats

Imagine an aircraft carrier—in the sky, not on sea. From its bay, it deploys swarms of armed drones that can fly, spy and kill, all guided by the touch of a computer keyboard thousands of miles away. This isn’t a scene from a science-fiction movie. It’s part of a recent proposal from the Defense Advanced Research Projects Agency, the Pentagon skunk works that brought us the Internet, videoconferencing and GPS. Now Darpa is soliciting ideas from companies on how to bring this technology to life.

Equally important are the questions about how drones will be used strategically. Drones do not only offer new ways to kill. They can prevent war. The cumulative U.S. and Soviet nuclear stockpile peaked at 70,000 weapons in 1986. None of them was fired, but all kept the peace by threatening mutually assured destruction.

Pentagon planners and defense intellectuals have spent decades analyzing the functions of nuclear weapons, but they have never considered seriously how drones could change the face of combat and coercion, whether by threat or with deterrence. Meanwhile, more than 20 nations, including China, are developing lethal drone technologies. In December, Iran said it was deploying an aerial drone replicated from Boeing’s ScanEagle surveillance drone. But Iran’s version is fashioned to crash into designated targets, earning it the nickname “suicide drone.”

Drones are going to revolutionize how nations and nonstate actors threaten the use of violence. First, they will make low-cost, high-credibility threats possible. Military planners have long assumed that high-cost actions risking blood, treasure and national reputation make the most credible threats. The classic example is U.S. Cold War “tripwire” forces in Germany. Risking 200,000 American lives signaled to the Soviets and to NATO allies that any Soviet invasion would kill many Americans, inevitably drawing the U.S. and its nuclear forces into war. Putting lives on the line proved that U.S. leaders meant it when they said the nuclear umbrella covered Europe.

Lethal drones, by comparison, are low-cost weapons They are remotely piloted (U.S. drones in Afghanistan have been piloted from Nevada), so they pose no risk of a pilot being shot down over enemy airspace. Each MQ-9 Reaper, one of the mainstays of the U.S. unmanned arsenal, costs about $14 million. By contrast, the Air Force’s newest manned aircraft, the F-35 Joint Strike Fighter, is expected to cost between $148 million and $337 million per jet.

Boots on the ground aren’t cheap, either: According to the Center for Strategic and Budgetary Assessments, the estimated all-in cost of a single deployed service member in Afghanistan in 2014 was $2.1 million.

The political costs of using drones are much lower, too. President Obama’s lethal drone strikes in counterterrorism operations have been controversial. But a December Rasmussen poll found that 71% of the public still favors using them. Such low political risks could change the game. Effective threat messaging used to mean taking actions that conveyed: “You know I mean business because I’ve put so much on the line.” A future effective threat could be: “You know I mean business because I can send swarms of cheap, lethal, stealthy drones at you all day long with no risk to me.”

Convincing the enemy that you have the domestic political support to do what you threaten is more important than ever because America’s wars since the 20th century have generally grown longer and more inconclusive. U.S. involvement in World War II lasted almost four years, and in Korea a full three years. Yet U.S. combat in Vietnam and the second Iraq war lasted nearly nine years, and the war in Afghanistan has lasted 13. When warfare was nasty, brutish and short, credible threats entailed convincing the enemy to “do this, or I’ll start shooting.” Today, the true test of political resolve is not initiating combat but sustaining it. Adversaries used to be sure that, over time, pressure would mount in the U.S. to bring troops home. The drones of future combat won’t have families or come back in coffins.

The current generation of drones also has capabilities we could not have imagined 20 years ago. Artificial intelligence and autonomous aerial refueling could remove human limitations even more, enabling drones to keep other drones flying and keep the pressure on for as long as victory takes.

Finally, lethal drones may make possible a new form of high-tech coercion: targeted hurting. Targeted terrorist-killing operations are designed to take an enemy off the battlefield. Targeted hurting could be designed to change any enemy’s behavior—by destroying selectively the family members, friends, associates, villages or capabilities that the enemy holds most dear.

Targeted hurting once was nearly impossible, because intelligence demands of precision targeting were too great, the lapse between identifying and hitting a target was too long, and the penetration of enemy territory required to succeed was too risky and difficult. It took 269 days to find Saddam Hussein in his spider hole even after U.S. forces invaded Iraq, and even though many of his countrymen wanted him caught. Now needle-in-haystack precision operations are growing far more feasible by the day. Drones already have the ability to hover over a target for up to 14 hours without being refueled, and to combine real-time imagery with real-time strike capabilities.

As robotic warfare technologies proliferate and evolve, the U.S. is in a strategy race with other countries engaged in drone programs. If we do not develop innovative ideas about how these weapons can be used for coercion as well as combat, others will.

Ms. Zegart is a senior fellow at the Hoover Institution and co-director of Stanford University’s Center for International Security and Cooperation. She is on the board of directors of Kratos Defense & Security Solutions, which manufactures military equipment, including equipment for drones.