2019 Air Space and Cyber Conference: Demystifying Multi-Domain Command and Control panel.
Moderator: Lt Gen (ret) Deptula
Participants:
Brig. Gen. Kumashiro (Dir, Jt Force Integration & Lead, MDC2 CFT)
Mr. Preston Dunlap (Chief Architect, AT&L)
Mr. Chris Brose (Head of Strategy, Anduril- Industry)
Lt. Gen. O’Donohue (JS/J7)
Lt. Gen. BJ Shwedo (JS/J6)
Transcript
Well thank you all, and welcome ladies and gentlemen. Normally it’s, you don’t wanna clap until after you’ve heard the speech. But that’s okay, we appreciate it. The term Multi-Domain Command and Control stands as one of the Department of Defense’s buzz words these days, however, far from being just a slogan, this concept describes one of the most central concepts to attaining success in modern warfare. Mainly, how do actors secure desired effects at the right time and place to be net desired objectives regardless of the domain service or system type from which the effect was initiated? Manifesting this vision demands a dynamic, connected, resilient command and control architecture that’s a far cry from today’s highly centralized, fragile lineal system. It all comes down to empowering decision making at the edge with timely, high-quality information, while also allowing real time teaming between assets in a multi-domain fashion. Now many new concepts have been postulated that demand a new approach to command and control and are driving the need for this area of focus. You’ve heard ’em before, Combat Cloud, the Third Offset, Mosaic Warfare. Those are just a few. While the details behind these concepts may differ when viewed specifically, it’s really encouraging to see that thinking at the macro level is migrating toward the need for a new command and control paradigm. A necessary shift, or evolution if you will, from the notion of centralized control, decentralized execution, to one of unified command, distributed control, decentralized execution. Now this is gonna be a challenging task, as the proliferation of information impacts across the strategic, operational, and tactical levels of execution. Command and control spans all services and all commands. It cannot be found on a budget spreadsheet as a specific program line item, however. Nor does it have a single champion, which is part of the challenge. But it is, in fact, the linchpin of what will bring success in future wars. For the potential inherent within military personnel and hardware demands competent execution and ability to do so. A tool is of little use unless harnessed properly in alignment with greater aims, and in many ways this is the essence of what Multi-Domain Command and Control is all about. Today, we’re fortunate to have a selection of leaders who are at the cutting edge of working this challenge. Preston Dunlap, who is Chief Architect for the Office of Air Force’s Assistant Secretary for acquisition, technology, and logistics. Chris Brose The former staff director at the Center of Armed Services Committee and now Head of Strategy at Anduril Industries. The Brigadier General Dave Kumashiro, who is the Director of Joint-Force Integration and the lead of the Multi-domain Command and Control cross functional team. So let’s get started, and we’ll here from Mister Dunlap first on his views of MDC2 an the challenges he faces. Preston?
Thank you. I wonder if you’re here this morning and have ever used Uber, Lyft, or a rideshare app. Raise your hand. Okay, congratulations, you have participated in Multi-Domain Command and Control. If you’re like me, yesterday trying to get out of the parking garage, you might have wished you’d taken an Uber to get here. I certainly did. Let me walk you through why I think you participated in that. Well you have an app that rides on a phone or a tablet that knows your position, position of those that might help you. It sends that information seamlessly over a network, cellular, wifi, Bluetooth, 3G, 4G, 5G one day. It knows whether you have high connectivity or low connectivity. It knows where you are and where people that can drive you are, and it matches through artificial intelligence and machine learning algorithms, the right person to the right car. And it does so much more than that. You can look at the screen and you see the situational awareness of where you are and what’s around you and you can watch the car appear to you. You can even coordinate responses and get multiple people riding in the same car. So just like that works for us in our private lives, we, too, are trying to get our capabilities within seconds, to the right place at the right time to the right target to the right effect. I’m gonna argue that it’s intuitive and it makes sense, and secondly I wanna be able to explain that I didn’t give you an example from the military, but I gave you an exampled from a booming commercial sector that is trying to solve many of these challenges. If we can harness that, and take advantage of that. Not just with cars, but with aircraft, satellites, submarines, and people and tanks on the ground. I’d like to show you a short video that our AFWERX, who you just heard and saw have partnered with this to be able to explain what Multi-Domain Operations is and get some energy behind the components that are available in the commercial world that can help us, even if you don’t know anything about DOD or the military, but can get after this problem with this. So let’s go ahead and cue that video, please.
[Spokesman] Imagine at any given moment, and in a matter of minutes, the five branches of the US Military, along with relevant allies and other organizations, merged together as one united team on a mission. They respond to a threat, stage an offensive, save lives in a disaster, and more. Leveraging the latest technology, they have instant and secure access to the most relevant data and information to optimize rapid, effective, AI-assisted decision making, And they have seamless and instant communication to deploy the right resources to take action. Building a multi-domain operation system requires solutions and automations for each activity that makes up the whole. We need to access data from a wide range of sources, and then prepare and present data so we can use it. We need AI-assisted human decision making, and then we need to instantly communicate the decision to all stakeholders. Along with the required actions and resources, we need to simultaneously act to neutralize the threat or avert the crisis. While all this is happening, we need to manage multi-level security requirements and keep the data secure. Finally, we need to integrate these activities into one seamless and resilient system that accomplish all of this in a matter of minutes.
So this is one way that we’re trying to explain what it is that we’re trying to do to a world that may or may not be familiar with DOD and challenges. It’s applicable in a variety of different operations. Humanitarian assistance, strike operations, and it’s that integration of capabilities that we see as the power of the effect, working together as a family of capabilities across the US Military Intelligence Community, our allies, and partners. So that’s kind of what it is and framing it. I’d like to end with a few minutes describing how we’re getting after the problem. What are the Lego blocks that we’re trying to build and shape to be able to snap together to achieve effect for multi-domain operations. Let’s go to the next slide, please. When we think of multi-domain operations, we sort of see three basic pillars or components. Number one, we need a concept of operations. We need ConOps, how do we fight together as a joint, multi-domain, all domain capability with our allies and partners? We’re two next, please. We wanna be ready, we need to be trained, equipped, manned, and ready to be able to take on the challenges that we have in a way that we’re training to be able to do those ConOps that are truly integrated, multi-domain capabilities. And number three, next, we need to build the technology that supports those new ConOps and the readiness that we need to be able to do that. And then technology is driving at such a rate that that may also change, and should necessarily change the way we have our ConOps and our readiness. And so it’s a tie left, up, and down, and to the middle. And General Kumashiro is gonna be talking more about ConOps and readiness. So let me focus on technology. Next, please. You got six lines of effort that are being pursued rapidly and aggressively in this area. These sort of follow the kill chain thinking, but they’re all done in parallel to be able to achieve effects as rapidly as possible. The first category is sensors. We need to tie in the publicly available information from the lowest classification up to the highest classification capability that we or our partners have. Number two, oh, if you could go back, please. Number two, we get, Paul just go back one. Number two we need to be able to get data off of the sensors. We need to do that, number three, securely and be able to manage that and push that to the right place at the right time. Whether it’s SAP, SCI, Secret, unclassified, it shouldn’t matter. We need to move that seamlessly just like the commercial world moves secure data back and forth. And number four, we need to be able to get that data to the right place and follow what we’ve been doing about a decade now, and separate the way our thinking. We’re software-defined network, software-defined radios, physical apertures, and be able to move that in a meshed construct back and forth to the right place and respond and heel as necessary. Number five, that data and information needs to be useful, so we need to push it to the exploitation and artificial intelligence algorithms that can take advantage of the technology to be able to not only get the data to the right place, but also siphon through the data to be able to assist human decision makers in their situational awareness, through space, and the air, and land, and sea, and be able to execute those decisions quickly machine to machine, either re-tasking sensors. Because I care about a place, I can automatically re-task space, air, and land sensors to go after something. Or put, as number six says, some kind of munition or electronic warfare or cyber effect on target. And finally at that number six category of capabilities, we’re looking for smart network weapons. It has a sensor on it and it sees something in its path, we wanna be able to re-see that kill chain quickly back to start number one again as we go through. And so these are the six rough categories with initiatives that we have begun. And next, we always are thinking of this as an arc to fieldable capability. And that’s how we’re gonna get after speed and deliver capability quickly. What that means is every time we do a demonstration or build a technology or do a prototype, we wanna see how that is clearly on the path either in and of itself or next to something that a warfighter can use. But how do we do that? Open architecture, open standards. The commercial world does this all the time. That’s how things work together. That’s how we can get to webpages regardless of the browser that you’re on. We wanna be cumulative in nature so that when we go do a demonstration and pull together technologies that each time, when something is working well, we just build and build and build and shift where we need to if things aren’t going the way we’d like. And that’s good. We should fail quickly and cheaply so that we can achieve capability faster. And the new wanna integrate that capability. So we’re gonna do demonstrations within each line of effort, roughly every four months. We wanna bring those together as an integrated package every six to eight months. That’s how we’re gonna get after this. We’re gonna compete. We’re gonna charge ahead. And we’re gonna do that together with our innovation partners. So last slide. We’re happy to talk to you. General Kumashiro and I have about an hour in a room tomorrow morning for any of the vendors or warfighters that are here that wanna chat more about how to get after this. When we set up partners, we don’t think we have the market cornered on all the great ideas, and so we’re looking for help for the four categories. Partners, you all in uniform and civilians, University affiliated research centers, federally funded research centers, primes, folks that we work with regularly to help us build our military capability. Pioneers, those that have sort of broken new ground in this area in the commercial world. And we’d like to bring you guys together to be able to talk and be able to achieve this capability quickly. So with that, I turn it to Chris.
Thank you very much, Preston. And thank you AFA for the opportunity to be with y’all this morning. I’m a little lazier, so I’m just gonna keep sitting here if that’s all right with you guys. My perspective on this was shaped largely by the time that I spent, nearly a decade, on the Services Committee. But really more in the past year that I’ve spent at Anduril Industries. And for those of you who aren’t tracking, Anduril is a two year old, venture-backed technology startup. We are a defense company. We were founded with the idea of bringing together people who are deeply expert in defense and military issues, and people who are deeply expert in the kinds of emerging technologies that Preston was talking about, largely have been working on these technologies for the past decade, decade and a half, in the commercial world. So I bring a bit of a non-traditional perspective to this and I really just wanna hit three points before we get to the Q&A. First and foremost, I think, when we really boil down trying to do what we’re talking about. What we’re trying to do is accelerate the speed of the kill chain, period. I think what we’re trying to do is enable our men to have better understanding, make better decisions, take better actions much faster and much more often than our competitors. That’s something that I think for the last three decades we’ve gotten used to a pace and an efficiency of operations that is just not gonna hold up in the future. Where the Air Force, I think, is going to have to close thousands upon thousands of kill chains a day in a computer type conflict. So then the type of approach that you’re going to have to bring into this is systems that are, and processes that are highly autonomous and highly intelligent. Which kind of brings me to my second point, which I think flows directly out of what I heard Acting Secretary Donovan say yesterday. Multi-Domain Commandment Control, I think, is primarily a software problem far more than it is a hardware problem. You know, when you look at it that way, I think it gives you a very different perspective on the problem that we’re talking about. You know from the perspective of software, there are no domains, there are no platforms. These are just ultimately nodes in a network and you want each of those nodes to be able to sense, shoot, share information. That is something that is very doable. Something, as Preston said, happens every single day in the commercial world. And the amount of information that is going to be generated by those nodes is exponentially larger than what we’re dealing with today. As many of you know far better than I, we’re drowning in that information now. So the only way you’re going to deal with and make sense of that information is through increasing automation. Now there are many things that artificial intelligence cannot do well yet. Processing sensor data is not one of them. AI, well trained algorithms can eat through large quantities of sensor data and help human beings interpret it, structure it, fuse it, make sense of what’s going on in the world, and generate understanding from it. Much of the processing of that information can be done right at the edge of the network because of advances in distributed computing. So you can essentially have sensors and systems that are running those sensors that are pending their own information and interpreting the world themselves. This is something that is doable right now. And because you have these systems that are capable of making sense of the world, you can have high degrees of autonomy and high degrees of intelligence in these systems. They can do far more by themselves without direct human control, which allows you to fundamentally invert the way we’ve built battle networks for a very long time. Which is rather than having large amounts of human beings required to make one machine operationally relevant, you can now have a single human being commanding, leading large numbers of machines single-handedly. And that fundamentally changes the way that you build networks. Rather than sort of centralized, concentrated way that General Deptula Mentioned, you can now create networks that are physically distributed, physically reconfigurable, self healing, far more resilient, far more secure than what we have today. Which is how we’re going to need to operate in the future. And that sort of brings me to my last point, which I think very much follows from what Preston just said. Which is, this is doable now. These technologies, underlying technologies are here now. We’re not talking about a photon torpedoes and magic. We are talking about technologies that, as Preston mentioned, most of us use every single day. It’s technologies that Anduril is using to build defense capabilities every single day. These are things that exist. This is how artificial intelligence is helping us to make sense of the information that we collect and obtain on mobile devices, the kinds of edge computer processing that CPUs and GPUs that I mentioned are running inside of self-driving cars. They’re in the Tesla’s that I passed in the parking lot that have computers onboard them that are hundreds of times more powerful than the most powerful computer and the most capable weapons system inside of the Department of Defense. So this is something that exists, and you can build defense capabilities and networks differently as a result of it. I think the point that Preston mentioned, which I was going to make ’cause he’s exactly right, the way you link sensors to shooters and targets to weapons is the exact same way that I got here this morning, which is pairing riders to drivers through Lyft and Uber and apps like that. So this is something that’s doable now. And just to kinda close on a bit of a personal note. I am asked a lot why when I left government I joined a very small technology company that nobody’s ever heard of. And it’s largely because these technologies can be brought to bear to help solve defense problems like the one we’re talking about here. I think there are many technology companies that aren’t interested in trying to help you solve this problem or aren’t interested in trying to help you. Anduril is an exception in that regard. We were founded to help solve problems like this, or help solve problems like this to help make the jobs and missions that all of you have to do better or more successful. And that’s something that we’re very excited to try to be a part of. And I’m honored to be here this morning, thank you.
Thanks, Chris. General Deptula, thank you number one for again, just the intellectual and academic rigor that you continue to bring to the forefront for our Air Force to move forward. And I’d also like to say thank you to the teaming that we have up here, because it really is representative of some of the challenges and the requirements that we need to get at this very, very challenging problem. So I work hand in hand with Preston and we go to a lot of meetings that kind of discuss all of these things that are facing our Air Force and our joint force and how we get at it. And again, I could not have a better partner, so thank you. And then also with Chris Brose, and this really extends to everyone who is out in industry there. This is so incredible, the teaming that we need to move forward and to move quickly to get at this problem with Multi-Domain Command and Control. So as many of you know, I am the cross functional team lead for Multi-Domain Command and Control. And so what does that mean? And so what I wanna do very quickly is just to bring us back to where General Goldfein, our Chief of Staff, started this journey for our Air Force back in 2016 and 2017. Many of you in this room know or are aware of the big rocks that some of these focus areas, that he really wanted to get at when he took over as our Chief of Staff. First and foremost was revitalizing squadrons. Second was strengthening joint team’s leaders. And third was advancing Multi-Domain Command and Control. This was 2016 and 2017 that our Chief has that vision to get at a problem that only now our joint community is really trying to tackle aggressively. Alongside that, many of the leaders in this room, General Holmes included, General Fay, General Fantini. The stand up of the Air Force were fighting integration capability. That organization to really do the integration of the horizontal networking that we need to deliver the capabilities that warfighters. So from these three big rocks, we stood up a enterprise capability collaboration team, an ECC team that my predecessor, Brigadier General Salzman, now Major General Salzman helped lead. And from that three big outputs came from that team of which Preston mentioned. Really, it’s the need for operating concepts for Multi-Domain Command and Control. Next was the readiness or the people that we need to enable that, and then the advanced technologies. And so we look at the advanced technologies. Preston has done an extraordinary job along with his team to encapsulate that and this concept of the advanced battle management family of systems. Platform agnostic to get to the technology that we need to network and understand and truly deliver Multi-Domain Command and Control to warfighters. So on our side, in AFWIC, it’s really about how we deliver those operating concepts and the readiness to the warfighter. And that’s difficult, and that’s been the extraordinary challenge that we’ve been facing. Because if there’s one way that I could describe it with what Preston and Chris had mentioned before, I just say #Ditto. I’d say #ThatwasEay because it seems real easy to do that. But the devil is in the details and how we actually get that out to the warfighter. And so within the operating concepts, we have several lines of effort that we’re working towards. And I have to say, this isn’t the perfect solution, that Multi-Domain Command and Control is organic. Although we can be deliberate in some areas, the network, the advanced network and technologies that we’re talking about, requires to be adaptive and agile at all times. But we are doing lines of experimentation. We are doing war games that really tackle some of the challenges that our combat commanders and our warfighters need addressed. We are looking at operating concepts like Mosaic that AFA and DARPA have presented. And then really recognizing that mission command on top of all of the technologies that we’re talking about are critical, particularly when we’re in the joint environment. And then moving on to the readiness, making sure that our organizations are postured and prepped and prepared to move forward. Whether it being with the Shadow Ops Center, which has just been designated as the joint of the Main Command and Control Battle Lab, of the Joint Requirements Oversight Council, or whether it’s the innovation hubs that we have spread out through our Air Force and through DOD. Whether it’s the software factories that have been established to get at this problem, this network architecture, the software enabled networks and equipment that we need. All of this encapsulated to move us forward, to get what this vision of Multi-Domain Command and Control is. We’re not there yet, and we need everyone in this room to help us get there. And we don’t have all the good ideas. And so that’s what my challenge is to you, to our airmen out there, to our Air Force, to our industry partners, to help us to get to that vision of Multi-Domain Command and Control.
Well, thank you gentlemen for those insights. The audience has forwarded a whole series of, if I may say so, excellent questions. So we’ll jump right in. We’ve got plenty of time, so those of you who still have questions, please send ’em in. Here’s a really good one to kickoff. Uber, Lyft, et cetera are susceptible to single point failure limited to GPS capability. How do we create a redundant capable system of systems that’s an open architecture construct that isn’t vulnerable to physical, cyber, or electronic attack? Also, many active duty military members experience the typical com network outages for multiple hours multiples days per week. (audience applauding) I’ll just add, if you notice me go off stage, the reason I went off stage is ’cause this thing stopped working. (audience laughing) How will the MDC2 effort build in contact safe guards to ensure confidence in reliability? Anybody?
Ah, I feel your pain. I work at the Pentagon. We don’t have it much better. So the, I think what you need to be thinking about is designing the networks to be robust and resilient and redundant, but also assume that it won’t be that way when you need it. So the commercial world, not to kind of overuse the examples, but we tend to build things and ensure first that they work in a sort of pristine environment with complexities. That’s not always the case when, say a commercial telethon company is working through their system, they tend to start with, where’s the degraded environment? And then build additional. So start with the one bar on your cell phone, not the five bars on your cell phone. And I think we’ve got to have that mentality. Well, we’re in a world that you cannot have 100% assured perimeter cyber defense, and so we have to be able to ensure that as we design the system, we can do that. So that means that we need to be humble and interested and accepting of practices that work across the community and across the commercial world, not just our own usual practices. And then be able to pull in the ability to make rapid change, Chris mentioned this. And so did many folks yesterday at the talks talking about the Air Force is, we’re the coolest tech company on the planet, and moving to a digital software company. And if you can move at the speed of digits, then you can also respond quickly to what’s happening. And so that’s a fundamental technology shift and a cure that can get after that. And I think that’s going to be some criticals for components.
And just real quickly, obviously this doesn’t happen in a stovepipe. So again, I see Mr. Marion out there in the audience. Enterprise IT is a service our digital Air Force. How again from a large Air Force perspective do we get at this ability to ensure that that network is resilient, is redundant, and has that capability that the individual asked the question, that we have that capability for the whole thing.
Okay, here’s another one. We’re hearing so much about MDC2 this week as a SecDev priority and of course the Air Force focus, how are we coordinating with other branches to make this happen? Are they using the same language, acquiring systems with the Mosaic concept in mind, how can our systems be better anyway? Chris, you have the opportunity to work with all the services. What are your observations on that topic?
I was just about to defer to General Kumashiro on that one. I would step back and make an observation from the outside. I think that the way that you’re ultimately going to get there, there are different languages in the services, I think they’re speaking about it differently, they are approaching the problem somewhat differently. But I think the good news from the outside is that each of the services realizes that this is a priority. They are seeking to get after it. I tend to think that the best way to get after problems like that is largely more from the bottom up. I think clear prioritization and top cover from the top is necessary, but really allowing the services to collaborate and experiment together to sort of feel their way to the future incrementally, but nonetheless pretty decisively I think is ultimately the way we’re gonna get where we all wanna go.
And I’ll just add on, obviously, we have some challenges in terms of that integration piece. Alongside the standup of the Shadow Ops Center at Nellis has this Joint All-Domain Command and Control Battle Lab, the same memo that is the Joint Requirements Oversight Council approved was the initial standup of the Joint All-Domain Command and Control cross function team. So obviously from an organizational perspective, that will be led by the joint staff, with all services participating. But as Chris mentioned, where we have to be careful is the bureaucracy doesn’t get in the way of what we need to do to make this happen quickly and rapidly. I can tell you just from experience, we have lots of joint service integration that’s occurring at much, much lower levels. So you take Shadow Ops Center itself. The team, the Air Force team is out there already working with Army and Navy and Marine Corps. And it’s happening in other innovation hubs, as well. So we can’t let all of the. Just the exquisite architectures that we would like designed into it. We have to understand that’s going to be a bit more organic as we go through and find our way, our path to a Multi-Domain Command and Control universe.
I just think I’d like to be really positive on this one. So I’ve watched this over a number of years, inside and outside of government, and the exciting thing that we have now is that each of the service Chiefs, though the language may shift left or right in the exact vocabulary, they’re all prioritizing ability to work as a team and integrate. Whether that’s called OODA or Multi-Domain Operations or Multi-Domain Command and Control, or Joint All-Domain, they’re all talking the same language. And so the fact that the senior most military and civilian members across the Pentagon are all pushing and driving their enterprises to be able to achieve this integrated effect. You know for us doing the technology development, that’s just extremely helpful because there’s a unity of effort and focus there to be able to do that. On a technology level, the relationships that we have with those service partners so that we can move messages and information back and forth across the intelligence community and DOD and the services. And there are elements in the Air Force, and I’m sure the other services, as well. But certainly we have actually more resources to be able to ensure that as part of these four to eight month demonstration cycles, that we’re resourcing integration capabilities, not just Air Force capabilities, but other service capabilities. So we’re putting our money where our mouth is.
Great deal. Here’s another one, accumulation of data sources, normalization of data across the spectrum, as well as proper implementation of artificial intelligence and machine learning requires a large increase of data scientists and analysts across the services. Is there a strategy? If so, what is the strategy for elevating the resources required to make this a reality?
So I’ll take that on. Just from my former role as Director of General Officer Management and Executive Talent Management, and again I’ll have to say that we’re not there yet and we need to get there. And our former secretary obviously recognized that as a capability gap that we have in our Air Force. And so I don’t have the best answers for you, but I know our Airmen have the answers out there. And so working together would really get to those types of skill sets that we need, whether it’s in software development, whether it’s in the technical pieces to understand all of these networks and how they interact. With that being said, we also continue to need the warfighters. The warfighters that are out there that understand how to deliver effects across all domains to hold our adversary at risk. And so that’s one of the challenges we’ll face in terms of balancing that attention.
You know I don’t have the institutional answer, but I will just say that as we’ve seen both at headquarters as well as across the force develop positions like myself, Chief Architect of Staff at AQ. I’ve seen that across our program executive offices and across the major commands, chief data officers. So the mentality shift from sort of procedural acquisition to not only doing that sort of well and better than anybody else, but also wake up in the morning and think as an acquisition individual civilian or officer how I can propel capability faster and smarter and leaner every single day, that’s just an exciting mission to get to wake up in the morning to. So I’m actually seeing a positive shift towards people getting more excited and energized about how to do that as we support it. I appreciate General Kumashiro and others that are trying to get the personnel enterprise that we have being able to onboard, attract, and retain, and support that talent to be able to get after that. But these kinds of challenges that we’re talking about this week are those that I think really excite those kind of individuals who have to get after the data. We have to get after AI and machine learning. And we’re seeing, I think I’m seeing a lot of that sort of begin to happen. And that’s also created full pathways to things like Chris’ company and others that are out there trying to make the way and become a capability that can actually sort of deliver large effects. And that partnership with that small business community or the sort of larger community that’s pushing on this allows us to not only have our own talent, but have better relationships with those on the outside that have the skillsets.
Okay, we got talked a bit about personnel and the way folks approach these issues. Here’s another one that’s related. How are you, we, going to encourage commanders to accept the risk following an artificial intelligence recommended course of actions despite the fact that they can’t ask why it’s making that recommendation?
I’ll jump on that one and I’ll use actually the Uber/Lyft example. I mean we do that all the time with Waze. And I used Waze on the way here, my brother was driving with me. It tells you where the speed traps are. It tells you where the pot holes are. It tells you where the stopped vehicle is. It tells you what direction to go if there is traffic or if there’s some type of construction zone. So we actually use it right now. And as Preston had mentioned earlier, our Airmen, they’re ready. They’re ready to acknowledge and take this risk. We just as an organization have to let them take that risk. Let them fail fast if necessary, and then get up and try again.
If I could, just a brief word on that. I think we, too often, jump to the assumption that a lot of these technologies are so new that the way we had thought about developing technology, fielding technology in the past somehow doesn’t apply anymore. I think the way that those Airmen are going to learn to trust these new capabilities is the exact same way that they’ve learned to trust every capability that they use. They train on it, they test it to ensure that it does the thing that they need it to do reliably and predictably and repeatedly. And in the process of that they build trust that when they that something needs to happen or that they need to send a system out into the world to create an effect, it’s going to do what’s it done in training. I don’t think that this is as confusing as so often dialogue makes it out to be. It’s much more of an extension and an outcurve of the way that Airmen build trust in their capabilities and understand its capabilities and limitations every single day. And as General Kumashiro mentioned, many of these are technologies that we’ve been using and have become very familiar with, and it’s less about why the systems are making decisions and what decisions they’re recommending and whether those decisions that we have tested and seem to work repeatedly and predictably in training.
I’ll just say, too, on the technology side, so the three plus, right? ConOps, Readiness, and Technology, and technology we in the Air Force call Advanced Battle Management Family of Systems. But what General Kumashiro’s leading in partnership with the joint staff and the other services is that sandbox to be able to test out those capabilities and gain comfort to be able to do that. And that tagline, as we think of Multi-Domain Operations, we think of that secure network, AI-enabled advanced battle management capability. So that enabled by AI is something where we’re gonna be training and testing and trying every four to six months as we work through the system. We’ll learn what works and what doesn’t and what our comfort levels are and what circumstances we go through.
I’m just gonna add real quick, I apologize. So my background, so in 2002 and 2004, I was in kind of the early stages of Predator, where we did remote split operations, meaning we flew Predator from far off locations over Afghanistan and Iraq. And then we also strapped on a Hellfire Missile to that. And then we also strapped on other technologies as those technologies became relevant. So again, that level of risk taking and innovation really actually inspired me to be where I am today. I can tell you right now I would not be here if I didn’t have that freedom of movement that the Predator technology has allowed us. Everyday it was something different, and we were making an impact. And this is what MDC2 is. MDC2 is about making that impact to the warfighter. Again, we just have to get out of the way of our Airmen and let them do it.
I think that’s great and listening to all of your answers, one of the things that struck me in developing trust, you mentioned Waze and Uber. You know there were a couple times driving into work I decided Waze wasn’t as smart as I was, and so I took an alternate route. And I’ve done that three or four times and figured out that Waze is actually smarter than I am. So I began to trust it. And the next time, okay I’m not gonna go on Route 50. I will take the alternate. So I think developing trust is a way to build that confidence with what AI can contribute. All right, what is the Air Force and/or the Department of Defense doing to modify proprietary contracts and systems that limit our ability to share and move information freely between networks, platforms, and weapons systems?
There’s actually a whole panel, I think, going on right now talking about that down the hall. So it is one of the main issues when talking about partnering with industry. So I think this is one of those areas that the companies have learned over time, depending on your strength or not strength in your market space is how open or not you are to others being able to operate within your environment. What we’re talking about here is we’re gonna need to partner to be able to move seamlessly between capabilities and integrate capabilities. You know the commercial world has currently kind of been working through this with 5G, we did it with Blu-ray technologies a long time ago. What’s the standard that we go with? How do we interoperate, make that capabilities interoperable? And so we’re making strong initiatives in the acquisition community to be able to get our capabilities to be not only open and modular, but also have the ability to be able to move the data back and forth, which means that the government needs to be able to have the ability to do that in a useful, strong manner. But we also wanna be able to protect the profitability and support the companies that make them viable as commercial institutions, and so the team is working through that sort of very steadily. The main point is that, if you wanna partner with the Air Force and other services in this way, you’ve gotta come to the table saying the whole point is to integrate and work together as a set of capabilities. So how can you as a commercial company, whether you’re a small business prospect you’re one of those sort of game changing pioneers in the world or a traditional prime or partner, you’re gonna come to the table and say, my goal here is to be able to share information and move back. Because I need to be part of a technology ecosystem and I need to set up my business and my relationships with the government and services in a way that does that, because that is the point.
Okay, you all mentioned to one degree or another that MDC2 is largely a software problem. I might argue that our processes and rules of engagement also need to be tackled. How are we looking beyond the technical part and getting the other services, agencies, and our own Air Force A-staff onboard with what you’re envisioning for the technical future?
So I’ll take that on. I’ve only been in the position for about four months and during that period of time one of the things that I’ve understood is the great challenges with Multi-Domain Command and Control is almost less about the war fighting piece than it’s about the decision advantage that we have within the boardroom and how we make decisions. And understand the complexity of the technologies that we’re about to employ in how we integrate and synchronize across all domains. Key with that is also the decision science piece of it. And I know our Vice Chief, you might be actually talking about this tomorrow, but one of the pieces is how do we look at decision making as it integrates with these technologies? How are GenXers different from Millennials in terms of how GenZ might make decisions in a VR environment, in a mIRC chatroom? And so these are the things that we need to get at in terms of the decisions making piece of it. But we also have to understand the policies that interfere, again, with our ability to realize what Multi-Domain Command and Control is. Whether it’s multi-level security across domain solutions, or whether it’s how we actually incorporate and synchronize mission command with those technologies.
Okay, well I tell you what, I think that that’s a great way to wrap this up. We’re coming to the end of our time. For those of you who are interested in more information on this subject, please pick up one of the Mitchell Institute publications out there on Command and Control Combat Cloud and Mosaic Warfare. Please join me in thanking our panelists this morning for their insights. (audience applauding) (inspiring music)