Watch, Read: The Future of Unmanned Military Aircraft Starts Now

Retired Maj. Gen. Larry Stutzriem, director of research for the Air Force Association’s Mitchell Institute for Aerospace Studies hosts Steve Fendley, president, Kratos Unmanned Systems Division; David R. Alexander, president, General Atomics Aeronautical Systems; and Lt. Gen. Joseph T. “Gus” Guastella Jr., Air Force deputy chief of staff for operations in a discussion on “The Future of Unmanned Military Airpower” at the AFA Warfare Symposium on March 4, 2022. This transcript is made possible through the sponsorship of JobsOhio.

Speaker: Ladies and gentlemen, please welcome the director of research at AFA’s Mitchell Institute for Aerospace Studies, Maj. Gen. Larry Stutzriem.

Retired Maj. Gen. Larry Stutzriem: Thank you. Thank you. It’s not often a director of research gets an applause. Thank you. I’m particularly pleased to host this panel, where we talk about the future of airpower, with a focus on unmanned in part, and we’re exploring a key intersection between operational requirements and technology. And it’s no secret that demand for airpower is off the charts. As many Air Force leaders have explained, it’s a common denominator across the theaters. You know, we fight with our Soldiers and Sailors, of course, but the same can’t be said for surface forces. And to that point, a carrier battle group has a difficult time navigating through Central Europe, and an infantry division faces challenges bringing forces to bear in the Pacific. Their domain, however, covers all these regions. And the same holds true for space, and the demand reflects this. As part of that airpower future, unmanned aircraft are going to be crucial parts of it. And we’ve seen this over the past 20 years with aircraft types like Global Hawk and Predator and Reaper, they fundamentally reshaped how airpower can deliver key effects throughout the battlespace. Mission activities once thought to be impossible are commonplace. I’m a witness to that, having seen the first Predators back 22 years ago and having seen them just recently, the Airmen have just taken it through the roof, what they can do. We owe a lot to those Airmen, and we owe a lot to our industry partners who made this happen.

So today we find ourselves at a crossroads. Operation realities are demanding a new generation of technologies and capabilities. Remotely piloted systems are going to increasingly transition to from being automated to the autonomous realm. But those are difficult technology hurdles to overcome. And we also know money’s going to be tight. So we’re going to have to bridge with capabilities on the flight line today.

So that’s what we’re here to discuss: to understand the future operational requirements and then discuss how unmanned aircraft are going to meet these new demands as part of the overall airpower inventory. What parts of this equation will be wholly revolutionary, and what pieces will be built off proven success—we’ll get into that.

With that background, I’d like to welcome our panel of distinguished gentlemen. First, we have Lt. Gen. Gus Guastella, deputy chief of staff for operations for the U.S. Air Force. We have Steve, excuse me, we have Dave Alexander, president of General Atomics Aeronautical Systems. And then we have President of Kratos Unmanned Systems Division Steve Fendley. Welcome, gentlemen. Gen. Guastella, I’d like to—Gen. Guastella, I’d like to start by giving you the floor and talk a little bit about the bigger vision of the future operating environment and where airpower fits in. And then I’ll let the two other gentlemen comment after you.”

Lt. Gen. Joseph T. Gus Guastella Jr.: Well, thank you very much, Larry, and thank you to the Mitchell Institute and the AFA and our industry partners that are up here with me. And thanks to all of you out there—in and out of uniform, past or present—that support the generation and employment of airpower for our nation and our coalition, because you’ve done so much. And we all owe the Air Force a tremendous debt of gratitude for what you’ve done.

“And so, hey, I tell you what, as someone with my perspective, as the A3 or the COOs, if you will, of operations in the Air Force, we are an Air Force in demand, just like you said, Larry. We’re in demand, I would say disproportionately to other services. And how can I back that up? Well, we’re in demand geographically across every combatant command. There is not a single combatant command that isn’t asking for airpower.

“We’re also in demand in another way: in the vertical spectrum of warfare, from counterterrorism, counterviolent extremists, regional actors deterring them, dealing with peer competitors, as we see happening right now. The full spectrum of conflict demands airpower. With that horizontal demand, combined with the vertical demand, has levied a heavy toll upon the Air Force—a toll in terms of readiness and modernization. We can’t move forward in both of those areas without some support. We need to be resourced to the levels commensurate with our demand or take something off the plate. All right? And that’s just a fact.

But today, the focus is on our RPA community and the manned-unmanned aspect of it. And, I tell you what, just a shout out right up front to our RPA community that’s out there—community that has seen incredible growth over the last two decades and also a community that’s done incredible work for this nation and our coalition. You know, just as I speak, we have, they’re flying over 1,000 hours a day, well over 1,000 hours a day, seven days a week, 365, out there, providing that unblinking eye, providing ability to find, fix, track and sometimes finish targets all over the globe. They also provide—it’s an important aspect to think about—is the unblinking eye in the deterrent value that it provides, the ability to watch someone and attribute bad actors and then make that attribution known to the world. And we’ve seen that with the condemnation of Russia’s invasion in Ukraine. That happened in large part to seeing what they’re doing, predicting it, demonstrating it well beforehand and then watching it unfold. That’s what our RPA community does. And, and so, the contribution has been in invaluable.

There’s over 3,000 RPA operators—pilots, if you will—1,500 sensors, 1,200 maintainers and 600-plus on the intelligence side that support the enterprise. They’re doing amazing work. You know, just later this month, I’m going to Ellsworth, South Dakota, to visit the 89th Attack Squadron—the Marauders—because they’ve earned the Global Atomics Trophy for this year, for their incredible work in supporting Al Asad Airbase and the defense of that base during the Iranian missile attacks on and in Iraq a year and a half ago. And so—two years ago—and so the bottom line, though, is we’re in high demand in that RPA community. But interestingly, unlike other force elements, there is no dwell for them. In other words, most force elements that we’re used to deploy, and then they come home, and dwell and reset and train. There’s no relief for that community, because they’re in such high demand. They’re all in, all the time. But we’re really fighting hard—this is my message to the community, and this is from Gen. Brown, as well—we’re fighting hard to the community to take some segment of the force, keep it back at home, so you can stay home and train. Train against, not the uncontested environment that we see today, but train in a contested environment, where you’re tested kinetically and you’re also tested in terms of your electronic warfare aspect and ability to maintain links.

So I think that’s an important aspect that we’re fighting for. But regardless, as I finish up here, there’s an imperative for change. Our modeling and our simulation, the analysis that we’ve done—and we need to partner with industry on this—shows that, without a doubt, artificial intelligence, human machine learning and taking it all the way to manned-unmanned teaming, is going to provide a degree of lethality, survivability and effectiveness that we have not seen. We have to do this. We see tremendous value in here. But I’ll say that we don’t have all the answers right now. We’re going to need to experiment. We’re going to need to see team with our industry partners, see what’s possible, look at the price points, where is the best value. And, out of that experimentation and iteration, we will see improvements to this force that we’ve that we’ve never seen.

But I just want to say one last thing, and that is, throughout all of this, as we increase automation, it is our policy—and it’s ethically correct—to always have human in the loop before anything is employed kinetically or nonkinetically. If before effects are employed, that consent needs to come from a human. OK? And that’s something we’re gonna bake in as we move forward. But the sky’s the limit on terms of what the future is for the community. So again, thank you, Larry. Appreciate it.

Stutzriem: You betcha.

Guastella: I look forward to questions.

Stutzriem: Great words. Great words. Dave? Please.

David R. Alexander: Well, again, I’d like to say thanks to the AFA and Gen. Stutzriem and Guastella. We’re really proud to be on your team. And we’re especially proud to have 1,000 people forward-deployed and be actually part of the real fight. So, we’re just glad to be here and be part of what you do.

You know, the strengths for unmanned aircraft are, you know, are pretty obvious. Persistence is really a key one there, and you’re not limited by having crew on board your aircraft, so I think that frees up the airplane to do, you know, super long endurance and persistence, which, you know, we really see as key to this unblinking eye that the general was talking about. You don’t have a pilot in harm’s way. You know, that’s a big plus. Something happens, a pop up comes up and shoots one of your airplanes down, you know, you don’t have a bad story to go home, you know, to his family.

The weaknesses, you know, come with that, and you know, through the years, the first one that comes to mind on a weakness is you need a data link to talk to these things. And I remember we thought we’d died and gone to heaven when we got 6.4 megabits of, you know, bandwidth and we could actually go to high def. You know, we thought that was revolutionary at that point. And so the data links, you know, are a limitation, although that’s getting better with time. But I would say in maybe, by 2025, when you have thousands of LEO satellites up there that can do up to, you know, 100 megabits, and more each, I think the connectivity that’s coming in the next couple years—and it’s not that far away—will be very game changing, and maybe some of that weakness will go away.

And then, of course, autonomy and automation, you know, and artificial intelligence. You know, we’re missing that real intelligence, which is, you know, the pilot in command that’s inside the airplane, we’re missing that. So we got to re-create it. And so there’s the kind of the strengths and weaknesses that I see. So how do we on-ramp, you know, these new capabilities with unmanned aircraft that are meshed together and taking advantage of all these new data link solutions? You know, I think what we’ve really got to focus on is early operational experience and get out there and get out there quick. If we take 10, 12 years to go to this next stage of, you know, autonomous collaborative platforms, it’s too late. And we need to take what we’ve got, we need to get out there. And I’ll just throw a date out. It’s my opinion that we need to be out there by 2025. And we can do that today. We can do it today with long-range sensors and just get a handle on AMTI and GMTI right now. We don’t need to wait. And if you study it forever, you’re going to miss the boat. And so my last say on on-ramping any kind of capability is, let’s move out and get operational experience on the way. Thank you.

Stutzriem: Very good. Steve? Please.

Steve Fendley: Yes, sir. Thank you, General. And thank you, AFA, for the opportunity here. This is obviously, this is my passion. I’m thrilled to be a part of this, thrilled to be able to speak with you today, and with Gus and Dave Alexander. I think it’s interesting, over the past 20 years, what we’ve gotten to witness is the effectiveness and the capability that exists in employing uncrewed systems, particularly unmanned aerial systems. So if we think about that, and we think about what the threat environment has been over those 20 years, we’ve been very fortunate. Because it’s been not entirely a contested environment, and we typically have owned the airspace. And both gentlemen talked a little bit about that in their openings.

I think if we look forward, that’s going to be the key, right? The key is going to be we’ve seen what these systems can do. What we need to do now is recognize the emerging threats, take technologies that are evolving on a daily basis, integrate those technologies and achieve the things that we’re going to need to be able to be successful and be effective, in now the contested environment in the near-peer threat that exists. So mass is going to be critical. I think all the analyses, the operational analyses, the wargames have all shown that mass is critical. Uncrewed aircraft systems provide the opportunity for us to achieve mass.

We have to do a couple of things to be able to realize that. One thing is it needs to be affordable. If every unmanned or uncrewed aircraft cost the same as the exquisite systems today—the exquisite manned systems today, for example, the F-35—we won’t get there. What we want to be able to do is augment those systems with very affordable uncrewed systems, consider distributing the capabilities that you need that increases the capability of the crewed aircraft systems that are out there. And I’m going to take a bold step here, and I’m going to say, I think you can achieve something equivalent to a generational advancement in your capability set, if you couple, for example, with manned-unmanned teaming, if you couple that capability with the crewed aircraft systems that exist today—and with the uncrewed aircraft systems that exist today. It’s a whole new dimension that hasn’t been fully evaluated at this point.

Finally, what I would say—and I would agree with both gentlemen—we have to start today. The time is here. The aircraft exists that can start to enter into this space. They’re flying today. What we need to do is start the experimentation, evolve and determine the tactics, techniques and procedures to take advantage of them, incorporate the technology advances over time, and what we’ll find is we will stay ahead of all of the threats, and we’ll be effective in this new threat environment.

Stutzriem: Very good, Steve. Great comments. Great opening comments. Let me begin with you, General. Kind of step back. We’re curious, you know, we basically got an Air Force, the entire enterprise is high demand, low density, you might say. And I’m curious if you could talk about a few of the more stretched-thin areas, mission areas, today that might need a little more resource focus.

Guastella: I had an opportunity to look at the questions beforehand, and I was like, ‘I don’t know, of a weapons system in the Air Force that it isn’t in demand and stretched thin.’ And you know, when you think about it, our airlift community, which did incredible work in the evacuation of Afghanistan: always in demand. Our tanker fleet: always in demand. Our rescue, our PR capability: in constant demand. One of the most stressed elements is our command and control—our E3s—in need globally. Certainly we’ve already discussed what the RPA community, what’s facing them. But the same thing exists for our fighter and bomber community as a deterrent, both in the Pacific theater, against the China threat, and certainly what we see with Russia. And, you know, but what underpins all of that—all of those force elements—is all the Airmen. You out there that are also involved in the generation of airpower, to man those airfields, to provide the agile combat support—that force element is also in stress. And so I don’t really have ones in particular, because they all are. And it’s good to be in demand.

Stutzriem: Very good. Very good, General. Well, let’s get back to the man-and-machine discussion. And I’m curious how, General, you see the strengths and weaknesses of what you might see in the future, with manned teaming and unmanned teaming. But, in particular, maybe talk about how you see us—our Airmen—getting comfortable with this. And I just harken back, you know, when I started my career, we started to have the combining glasses within the aircraft and the fighters with all the green stuff that gave us all this information and you could have your head out of the cockpit. And I remember my wing commander telling me, ‘Don’t trust the green stuff.’ You know, it’s a ridiculous thing today, but we may face that same thing with manned-unmanned teaming. I’m curious what you think about that.

Guastella: That’s a great question, you know. And I think Secretary Kendall addressed yesterday a potential right bookend of the future with you have a fifth-gen aircraft team with multiple unmanned wing men out there operating together. What struck me is like the debrief’s gonna be really lonely. It’s just you and the software, you know. And the bar afterwards is going to be really bad. But, no, the potential is there. When you think about the cultural shift, if you do have those kinds of teaming events, there’s going to be human machine learning that’s going to have to happen. And there’s going to be learning and what kind of what kind of different coding we’re going to need to to fully enable the to get the most out of the the unmanned elements of the flight. And then how can how can we better have orchestrated—how can the human better have orchestrated—that kind of capability? So the iteration experimentation, I think, is going to be key there.

But the one key advantage to the teaming as we see which is—even if we don’t team—is the accepted the potential for the acceptance of risk with an unmanned aircraft that doesn’t exist with the manned aircraft. We had multiple platforms shut down in the Middle East, by the way, but it shows … we accept some additional risks with the platforms, but it also shows an adversary’s willingness to engage an unmanned platform where they may not have engaged a manned platform. So that is a dialogue and that is a discussion we’re gonna have to have, especially when you consider the cost of the platforms and … what are we willing to risk?

Stutzriem: Yeah, really good point. Dave? Steve? I wonder if you have any thoughts on that?

Alexander: Well, I just, you know, I’m thinking back 20, 25 years, you know, when we started this whole mad rush on the war on terrorism that we had some lessons that we learned along the way. And, you know, we rushed a lot of product out to the field. We, and basically, just multiplied it and produced it quick. And, you know, it was very successful. And it was combat-air-patrol driven. And we got the caps out there, and we hit the 60 caps and then had to chase ourselves to maintain it. And, you know, I think what suffered along the way was things like automation, things like autonomy and things like open-mission systems. And, you know, for me, that’s a big lesson learned, you know, going forward that when we do rush out to the field—and that’s what I believe is the right way to go do it—but we also need—if it does get big—we also need to be able a way to insert technology as we go, so we don’t end up with a big, you know, pilot crisis, like we had, you know, just a few years back.

You know, an example of that, if you just embrace automatic takeoff and landing over SATCOM, you eliminate the forward-deployed launch and recovery unit, which is a whole other set of crew and a whole other, you know, ground control station and data links, that whole thing. And so you can eliminate all that. If you eliminate manual takeoff and landing, you will also reduce your training substantially. And then if, you know, things like single-seat ops and multi-aircraft control for your long transits. But if you combine all that together, you can reduce your crew by 60%. OK? And, you know, I think we were so busy putting those caps out in the field that we kind of left that piece behind. Yeah, and I think it’s a lesson learned going forward. And, you know, as we get into this new, you know, autonomous, collaborative-type platform, we need to keep that in mind, so we don’t, you know, don’t get the focus on just having the, you know, the mass out there, but maybe have a few combat air patrols set aside, you know, for keeping the technology up as we go. So anyway, that’s a key thing, I think, going forward that we need to keep in mind.

Stutzriem: Thank you. Steve?

Fendley: Yes, sir. For, to me, for manned and unmanned teaming to be successful, there’s really one fundamental element that has to be resolved and has to become true. And it is, the pilot needs to trust the unmanned or uncrewed system. Absolutely fundamental. If we don’t achieve that, this will never be successful, it’ll never be effective, no matter how much technology and capability is there. So think about that fundamental. You’ve got to be able to trust the system.

How do we get to be able to trust the system? I think there are a couple of couple of ways to approach that. One is very, very important to start, and I’m going to say, ‘start simple.’ Don’t start with a highly complex, highly integrated, uncrewed system that represents—and I’m going to make up a number—that represents 11 different sensors and weapons that you want to control remotely while you’re trying to fly an aircraft. Think about one sensor on an uncrewed aircraft. Think about one uncrewed aircraft to start with. Maybe over time, we’ve all that and it’s a number, but think about one pilot controlling one uncrewed aircraft, with one sensor, maybe one weapon system. Think about the interface to that. Don’t load that pilot up; don’t put him at risk, right? The uncrewed system is supposed to make him more survivable and more effective. Don’t do something that takes those capabilities away.

So think about the most simple interface that you can offer. I would suggest if you use interfaces that exist in the airplane already, for example, one an interface maybe to a wing station pylon. Think about having that same interface be able now to control your uncrewed aircraft, which effectively becomes an off-board sensor or an off-board weapon. So trust the system, start with a simple model, gain the trust of the pilot, implement something that will be successful early, and what we’ll find is it’ll start the momentum to allow us to increase the complexity without increasing the risk and without reducing the effectiveness of the overall battlespace.

Stutzriem: Yeah, and when you say one sensor, sensor technology’s moving so quickly, and integration of different sensors can do some amazing things. So it can still be a very powerful platform, even limited in its capable hardware. I would think. Dave, Dave’s kind of a historical figure in this, and we were actually, General, Dave and I were up at Ellsworth a couple years ago when the squadron won the award the first time—the Marauders. And, Dave, you go way back, you know, at the start of this year, and I’m curious if you’ve gleaned some key lessons that can inform the future when we talk about manned and unmanned capabilities.

Alexander: Yeah, I think it’s like I was mentioning before is just to make sure that, you know, you don’t leave big aspects of the system behind, while you’re, you know, fighting a war. You know, I just, you know, when I looked back, there are some things that we could have done, you know, a whole lot better, that would have prevented having, you know, a pilot crisis or, you know, would made software releases more efficient with open-mission systems. And so, if we could get that blend of getting in the, you know, getting into the war fight early with not totally the, you know, the 100% solution, but don’t forget about, you know, updating as we go. That’s really key.

Stutzriem: Very good. Steve, a question for you, please. Kratos has really come to the table with several new technologies. And I think everybody’s familiar in the UAV realm about the Valkyrie. And that’s just one of your efforts, of course. How do you size up the macro requirements for UAVs? You just talked about those fundamentals just now. But how’s it shaped your thinking in terms of what we’re innovating out there on the horizon?

Fendley: Sure. At Kratos, we’re incredibly proud of the Valkyrie. And I think that’s probably what we’re most known for at this time. We obviously have several unmanned aircraft systems that we’ve developed over time. And maybe not everybody’s familiar. Our background, our original background, really was in unmanned aerial target systems. And the concept that was applied to the Valkyrie—and now our other tactical systems—was to take the approach that we take on the target systems. Of course, target systems are used for crew training and for weapons development. Necessarily, they have fighter-like performance; they can replicate fighters; they can replicate cruise missiles, for example; they carry different threat systems. So consider that mission systems. Fundamentally, they have to be very inexpensive. Because ultimately, they’re designed to be shot down—sometimes in their first mission, sometimes after a number of flights. But this by our own forces, right? Our own forces use these target systems.

Fendley: So we looked at that and said, ‘OK. How does that apply to the tactical arena?’ Interfacing mission systems: It’s already something we do. Having a very, very inexpensive to develop and operate aircraft system and produce is something that already exists in the targets realm. So apply those to the tactical mission. Now, if we think about where we need to go, and how we need to consider evolution of those tactical systems, we imagine that—back to the simplicity—there needs to be a very simple interface.

Oh, and I know something, something very key to point out, our first contract related to the Valkyrie was with AFRL. And very interestingly, it was sort of focused on the technology. But really what it was focused on was validating the cost model. The cost prediction said, even though there was almost 100 years of data on military aircraft that said you can estimate the cost of a military aircraft based on a dollars-per-pound parametric. The fundamental element of that program was to validate the fact that we could get off of that cost curve and develop and produce an uncrewed aircraft system that was effective against the mission types that were being considered but nowhere near the cost per pound of those of the legacy aircraft. We were able to do that. That program was successful. That data was turned over. That, to me, really opened the floodgates on what’s possible. We talked earlier about the need for mass. We keep talking about affordability and the budget environment that we’re in today. And I think that really has paved the way for what the possibilities could be.

Stutzriem: Hey, General, I’d like to follow that up with maybe some thoughts about how these are going to be used—these unmanned systems in the near future.

Guastella: Oh, thank you. I say what, I think, it’s the how they can be used is wide open. If you look at every mission set that the that we have in the air domain, air command, air domain awareness, C2, air lift. Obviously, Secretary Kendall talked about the, you know, the combat air forces and how they could team. Rescue. Which mission area out there wouldn’t possibly benefit from manned-unmanned teaming? I don’t think there is one. You know, as the general mentioned before, the need to do, how can the RPA or unmanned systems integrate in with the agile combat employment, which we’re already doing now, with traditional aircraft? How can unmanned systems leverage that concept for maximum effect? You know, we know our main operating bases are going to be threatened. And by using agile combat employment in a myriad of different landing surfaces and capabilities, both manned and unmanned, we’re going to have a very resilient Air Force. And it’s very difficult to catch us or target us on the ground, which provides a tremendous deterrent.

So that technology and capability, I think, that we get we need to think about upfront as we development system, unmanned systems. When you think about it, you know, if you have a air moving-target—an AMTI—airborne moving-target indicator, need basically air command and control, air domain awareness, air battle management. Think about how an E3 or maybe an E7 platform could be augmented with unmanned systems that are farther in, that provide that more distant sensing, that all allows it to integrate and provide better air battle management. Think about could you use unmanned aircraft to protect high-value assets that need to be up and in potential threat environment? So those are all areas I think that are worthy of exploration.

Stutzriem: Yeah. Let me follow that up, General, with, you know, pilots, the number of pilots we have. You continue to have around a 2,000-plus pilot shortage. And we’ve cut back the training complex, you know, the capacity to surge. And we’re in a place where we’re pivoting to pure conflict. So I’m curious how you see the unmanned aircraft, actually helping alleviate that shortage, but also being able to surge into combat.

Guastella: Well, I think, if I can first address the pilot shortage, we absolutely have a pilot shortfall in the United States Air Force. We underproduced pilots for over a decade, and as a result, it’s going to take a decade to get out of this situation. And we have a shortfall in two different areas. No. 1, we have a shortfall in, you know, new pilots that have just been produced. And we also have a potential shortfall in more experienced pilots that have a chance to later in their career to get out. And the only way to fix all of that is to increase production. And so that’s something that we’re definitely looking at and going to pursue in the upcoming budgets.

But yes, there will be a growth in the unmanned community, but it will not be a growth that allows us to offset the pilot shortfall. That is today a problem that we’re going to have to get after. Sometime in the future, a decade plus more, I could definitely see that shifting a little bit. But for now, we need to address this problem, because it’s very acute for the United States Air Force.

Stutzriem: “ou know, we see this in a lot of the research we do at Mitchell that it’s so important to have ops capability, war fighter capability in a lot of places, not just in the squadrons or in this, but in the staff’s program offices working with AFRL. So the human piece of it certainly needs to be be adequately sized for all. Thank you for that.

I’ll pivot here a little bit to Dave and Steve. You know, we recently released a paper at Mitchell looking at future UAV capabilities and a major portion centered on this notion of automation, which is, you know, basically the machines are following a script and autonomy, which sees, you know, this independent machine-driven decision-making response to dynamic real, you know, real-world, real-time events. Where are we on that development path? How do you see that in your industry endeavors? Our research shows that we’re more in the automated place than we are in the autonomous place. Where are we at? And Dave, I’ll start with you.

Alexander: Sure. Thank you. So, you know, I think automation, you know, is here and we’re using it, and I think the future is, of course, autonomy, and we’ve got a long way to go there in my opinion. And then we also need to separate automation and autonomy. You know, when you’re taxiing and … taking off of the runway in a real crowded airspace and then on the return mission coming back, and you’re under air traffic control, and there’s a lot of jabberjawing going on the radio, you know, that kind of automation—where a midair collision or runway incursion could occur—that kind of automation, I think, is really, really hard and really, really risky. And I really, you know, I think that piece we oughta stick with automation for a while.

So what I’m going to talk about is mission autonomy, and meaning there, you know, the platform’s in place and now you’ve handed it over, you know, into the mission. And I think that, you know, that’s the future’s bright there. And, you know, it’s really more driven around AI on the sensors, and then let the sensors tell the aircraft where to go. And I think if we take that approach, and a lot of these new AMTI, GMTI and some strategics, these kinds of sensors lend themselves to big data, they lend themselves to AI, and in the automation that can or autonomy that can drive the aircraft to be in the right position.

So I think the future’s bright there, and I think the one key piece of that is bright, and it’s, you know, there’s excellent programs going on right now with Skyborg to develop that and nurture that and, you know, a common code base that everybody, you know, deposits their code into. And so, I think, you know, you’re gonna have everybody working together instead of everybody working apart with this new approach, using code, and I’m talking code now—C O D E—not software. But anyway. So that, I think is right. Let’s separate it, though, and not get too carried away. We still need a pilot talking on the radio, when it’s in heavy traffic. So let’s keep that part safe please.

Stutzriem: Steve, please.

Fendley: So this is a fascinating topic, fascinating element of the technology trade space, and I’m going to call it kind of the ‘geopolitical trade space,’ which enters in as well.

Going back to something we talked about earlier with trust, autonomy—let’s say automation. Automation and sequencing is coupled pretty tightly to trust, because they’re all predictable, pre-established responses that you can expect. I can tell you from our test and demonstration flights, pilots don’t like to be surprised. We never surprise them, of course, with our systems, but I’ve heard that it can happen. I’m joking. We do, obviously, have some surprises. And they’re not appreciate, and I can’t imagine, in a conflict environment how detrimental that could be. So back to the trust, it’s so important.

So deterministic solutions for, I think, manned-unmanned teaming in close proximity is is paramount. Now let’s talk about a scenario where you have a UAV off operating by itself performing a mission. In that scenario, and in the, and—truly, in the overall mission scenario, David, that you talked about—the artificial intelligence … the technology is here, to start to use it, and the opportunity that presents is boundless. Because if you can have real-time decision-making that occurred, that occurs based on responses and factors that you couldn’t predict all of before you were in Scenario X, then your capability, your potential effectiveness goes up exponentially.

So I would think about close-in operations, manned-unmanned teaming, be careful with the introduction of AI. You want to do it very slowly—back to the trust—that’s so critical. As you’re in a more remote scenario with a one-off—maybe even with a swarm—then AI can apply. In a swarm, what I would say is you want the higher-level hierarchy using the artificial intelligence, and then you want the followers, if you will, to be following an automation script. And I think in both those cases, you get the trust you need to make it successful.

Can’t stress enough: We need to be pushing the edge of the envelope on these technologies today. We need to be practicing with them. We need to be getting them out there in the field and letting the user evaluate it and determine exactly how we want to incorporate these. Because the technology exists to go very fast once we’re ready.

Stutzriem: Well, we’ve got about one minute left. So I just offer any closing comments you might have, briefly. General, I’ll start with you.

Guastella: Well, I’m just glad I flew fighters when I did, because if I look into the future, I think the unmanned systems out there are gonna make up for all the shortfalls that I brought to the cockpit. But seriously, though, it’s definitely a growth industry. I think it’s critical that we partner with industry moving forward. And I think this type of work is going to give us the edge that we’ve had before—that we need to regain—against the peer competitors that we’re going to face. So we’ll seize the opportunity.

Stutzriem: Well said. Thank you, General. Dave?

Alexander: Well, I for unmanned aircraft, I’m just going to say that, you know, the future is super bright. I mean, just like you know, in the late ‘90s, 1990s, when we exploded into counterterrorism, what I see going forward is an even more big change. And it’s going to be coming with satellite connectivity that is just unbelievable with, you know, if you think about all the LEO satellites are going to be floating around the world by 2025. And the data rate that you can get through these in megabits is mind boggling. It’s orders of magnitude more than what we’ve been experiencing in the past. And you connect that up with persistent aircraft, but multiple battle management nodes, and, you know, this is the beginning of ABMS.

And so I really think the future’s bright there. And then when you bring in automation and autonomy with automatic, you know, agile basing, I really think the future is gonna change for unmanned aircraft going forward. But you know, the markers will be, are we going to talk about it for 12 years? Or we going to go get something done by 2025? And I really think we all oughta take that to heart. How do we get something going now? There’s airplanes right now that can go do a lot of these missions, that are basically standoff missions with AMTI, GMTIs, strategics. Again, they can all be done today. We don’t need new starts. Let’s go get it done. We need new starts in some of these, you know, combat-patrol-type aircraft. But you know, in the meantime, let’s go get this done.

Stutzriem: Yeah, true.

Alexander: Let’s not, let’s not—12 years from now, there’ll be a different problem. And we’ll have solved the wrong problem by that time. So my last word is, ‘let’s move.’ Let’s make it, let’s make it awake.

Stutzriem: Thank you. Steve?

Fendley: I’ll double down on the ‘let’s move.’ All the building blocks are there. We’ve talked about them today. We’ve proven the cost curve. We’ve proven that you can produce an affordable system. An affordable system is required to be able to get to the mass that we know we have to get to solve the peer-to-peer threat. We know what’s happening in the technology associated with artificial intelligence. We know we can apply those things in a progressive manner. We have airplanes flying today that can solve some of these problems. We know that—based on what we’ve learned with unmanned aircraft to date, based on what we’ve demonstrated with the technology building blocks, we can start to integrate these systems—we need to get them out in the field. We’re all unanimously saying that up here. It’s time to move.

We can be ready for the next conflict. Most importantly, we can probably prevent it. If we evolve these technologies, get our military trained and the best in the world with the application of these technologies, and have the rest of the world recognizing that we show the dominance in this area. And particularly, we started off, I think, by talking about how critical the ‘owning the airspace’ is, basically. And I think this is the next generation that gets us there.

Stutzriem: Very good. Very good. Thank you. Well, gentlemen, thank you so much for participating. I’m sure folks are gonna mob you after this. It was a great discussion about the future of airpower and especially the unmanned piece of that. I’d like to ask everybody to do something. Go to and sign up, and you’ll get announcements and notices of our activities and publications first before anybody else. And we’re not going to give our panelists any gifts. As we’ve been doing in the rest of the activities, we donated a sizable amount of money so that the Airmen and Guardians could have a great time at the pool last night. In fact, I think the president of AFA underestimated how much they’re going to eat and drink. It was a great time. Really good time. Thank you very much. Next activity’s at 10:35.