A new report from the Center for a New American Security suggests that the U.S. Navy—and indeed the other military services—could save tens of billions of dollars by adopting semi-autonomous unmanned combat aircraft. Money would be saved because the Pentagon would not need to buy as many planes, nor would unmanned systems have to fly as many flight hours for training purposes. Moreover, it would afford the Navy a way to extend the reach of its carrier air wing so that naval aviation could strike deep into the heart of heavily defended enemy territory.

“The cost savings are just tremendous,” said Paul Scharre, one of the authors of the study and director of the 20YY Future of Warfare Initiative at the Center for a New American Security.

According to the study, the cost savings estimates top $100 billion. In fact, under the most optimistic scenario, the savings could range between $95 billion and $170 billion while a more moderate case it might mean savings of between $80 billion and $140 billion. Under the most modest case scenario, the savings range between $30 billion and $54 billion.

Scharre makes the argument that when a “pilot” or drone operator “flies” an unmanned aerial vehicle, the human is only really making macro-level decisions rather than directly flying the machine—particularly in more advanced drones. Basically, even if the pilots were to maintain their proficiency in simulators only, it would make little difference to that operator. As far as that operator were concerned, the input or data coming into the simulator would be identical whether the drone as simulated, flying over Southern California or flying over Iraq. “It may be indistinguishable to the operator,” Scharre said. “They might not even know.”

That means that the Pentagon could save huge sums of money because pilots would not need to fly to maintain currency or fly peacetime training missions. Only a few aircraft would be needed to retain the skills of maintenance crews for operations at home station. The rest of the aircraft could be deployed for operational usage.

“The report is really about cost,” Scharre said. “The cost of the airframe is identical. If you’re building a combat aircraft it’s going to have certain requirements in terms of speed, range, low visibility, weapons and everything else—those drive the cost of the aircraft. The cockpit costs are negligible. All of the saving come out of the training hours and the hours to keep up pilot currency. But then that has follow-on effects on the number of aircraft you buy, and that’s where there are huge savings.”

Right now, the Navy has 10 air wings for its fleet of 10 carriers, however only a few of those carriers are deployed at sea with an air wing onboard. The remainder of the air wings and carriers are either in training or maintenance. Even during a surge, the Navy might at best deploy six carriers and their air wings, Scharre said. With a fleet of unmanned combat aircraft, the Navy would only have to buy enough unmanned combat aircraft to fill six air wings and its maintenance pipeline—which would significantly reduce the number of jets that the service would need. “You’re not buying 10 air wings,” Scharre said—but nonetheless, each deployed carrier would retain the same—if not greater—combat capability.

However, while Scharre said that he hopes the CNAS study will influence the Navy’s on-going F/A-XX or Next Generation Air Dominance analysis of alternatives (AOA), he said he recognizes the carrier air wing will never be fully unmanned. More realistically, the future carrier air wing will be a mix of manned and unmanned systems. And, indeed, Scharre said that it would be preferable to retain a combination of human and autonomous control. “I think there is a role for unmanned combat aircraft on the carrier deck from an operational standpoint,” Scharre said. “But it’s unreasonable to think—and I don’t know that anyone is suggesting—that the Navy go to an all-unmanned carrier air wing.”

The CNAS report is not about what the mission of a future Navy unmanned combat aircraft might be. It’s simply about the potential savings the service could achieve with increased numbers of unmanned aircraft. “We would like to see the Navy incorporate cost analysis into their AOA,” Scharre said. “In particular as they evaluate manned, unmanned and optionally manned variants of an F/A-XX—that they think about cost.”

Dave Majumdar is the defense editor for the National Interest. You can follow him on Twitter: @davemajumdar.