 Good morning everybody. If people could take their seats and we'll get started. I'm Maaren Leid at the Center for Strategic and International Studies and I'm very honored to have, you can't hear me, I'm sorry, are the mics on? No. No. Okay. Somebody gets sitting in the front and said she couldn't, okay, there we go. Maaren Leid, I'm a senior advisor here at CSIS and I am honored to be here today with Dr. Arthie Prabhakar and General Rooster Schmiddle, Lieutenant General Rooster Schmiddle to have a conversation about innovation and future military operations. You have their bios in front of you so I won't delve into them in depth but the short story is they both have long histories both in assignments and in practice in leading and managing innovation in a variety of contexts. Dr. Prabhakar, principally from the technological side but also as a venture catalyst trying to seed innovation and identify it and now as the Director of DARPA laying out the future course for I think what most people see as the organization that is the, who has the most direct mission for thinking hard about technology innovation for the department as a whole. General Schmiddle has sort of come from the other side as an operator, a marine aviator who has been deeply involved in a number of both organizational and operational experiences where he's led and participated in attempts for the operational force to try to conceive of new ways of doing business and employ technologies as part of that effort. So I think an exceptionally well balanced panel and hopefully will foster a very rich conversation this morning. I want to make a quick note about the most important thing which is food. At the end of this panel it will be time to grab lunch before continuing on in your day and in an attempt to better manage people flow, our group is supposed to take the elevators down to the concourse level where we will be getting our food and others will go elsewhere but so anyway if you can just, when you leave here go down to the concourse to grab lunch before heading into the afternoon sessions. And as always if you could turn off your ringers that would be much appreciated and when we get to Q&A I'm going to ask Dr. Blocker to make some opening remarks and then General Schmiddle will follow her and then we'll open it up to everybody so if you could briefly identify yourself and then ask your question as succinctly as possible it would be much appreciated. So Dr. Blocker over to you. Great, thank you so much for the chance to do this, this is Rooster and I get to do a duet which will be a delight. I think many folks know DARPA but to review quickly our mission is breakthrough technologies for national security that is unchanged since we were founded in 1958 in the wake of Sputnik quickly realized that the best way to pursue our mission of preventing technological surprise was to create some surprises of our own and so that's still very much why we come to work in the morning. In 2014 what that means is preparing for a world in which we have a plethora of different types of threats and challenges. It's no longer the Cold War where there's one monolithic adversary that focuses all of our attention so today we are of course concerned about the emergence of peer competitors and the actions of nation states but of course we are also still daily confronted with terrorism and it's links to criminal activity around the world and linkages from those activities back to nation states we're confronted with the emergence of new diseases like Ebola and other humanitarian challenges that occur around the globe and that bubbling pot of chronic but also acute challenges that the department faces those are the things that we worry about at DARPA as well because we're a technology agency a lot of our focus is first thinking about the importance of global technology in empowering those threats that we face and that environment today is one that is radically different than the world in which DARPA was created we no longer can assume that the technologies that we have access to give us decades of advantage over everyone else on the planet and in fact we see from insurgents to sophisticated nation state capabilities we see the use of very powerful globally available technology moving at a really rapid clip and I think that creates an environment in which the the technologies that we work on need to deal with that that new reality in a very different pace and so that's the context for our work let me share with you three major themes for the work that we're doing at DARPA today much of our work today is about fundamentally rethinking complex military systems there we recognize that in many many aspects of what we do with military technologies today we're reaching the point of diminishing returns and many of our highly sophisticated highly capable systems are massive monolithic platforms that yes they're powerful yes they're capable but they become point sources of vulnerability they are extremely costly they're very slow to build they're very inflexible to operate and over time they are hostile to advances in the underlying technology they're nearly impossible it feels like to upgrade that's not going to cut it for the world that we're in and the pace at which technology is moving around the world today so in areas like rethinking air dominance in rethinking how we use the space domain in rethinking our control of the electromagnetic spectrum in many different areas also in the maritime domain we're fundamentally rethinking the kinds of systems and architectures that that we're building today to create a future where we have far more robust capabilities but importantly capabilities that can be built for fewer dollars built more quickly built to upgrade built to work in cooperative distributed often as a way to achieve those objectives that's the character for a lot of the work that we're doing across the agency a second major theme at DARPA today is about dealing with information at massive scale the information explosion that we all experience in our civilian lives is paralleled with the explosion of information taking place with the sensors and systems that we have in the military context and two key things emerge out of that that are important from our perspective one is a recognition that we are so reliant on information and information systems that if unless we create a way to have a foundation of trust in our information and information systems we're simply going to be we simply are going to continue to live with what I think is a an unreasonable vulnerability again because of the reliance that we have on information systems so cyber security and the use of the cyber domain fundamentally changing our capabilities in that area is one major area focus for us at DARPA a second is to realize that in that explosion of big data are places where bad actors hide but also enormous opportunities if we can start harnessing that information rather than just drowning in all of those bits and so that the opportunities for cloud analytics and big data I think are another important part of mastering the information explosion the third area for DARPA is one that is a perpetual part of our mission which is to scour the research horizon and look for the places where we see the bubbling pot of research that's that creates the foundations for the next generation of technological surprise we're working on a variety of things in that area but one that I would particularly highlight today is the opportunities that emerge as biology is intersecting with the physical sciences and engineering and with information technology in the work that we're doing today on neuro technologies on synthetic biology on working to outpace the spread of infectious disease examples of some of the ways that we see our ability to harness these new technologies but also our ability to understand how their power could be used against us those are some of the aspects of this business of always looking for the next seeds of technological surprise so I hope that gives you a little bit of a sense of the way we're thinking about the future of technology and its role in changing our national security capabilities we very much look at the world from the technology lens and I think you know very complimentary to our colleagues in uniform and turning it into real operational capability which is when all of this gets real so I'll stop with that when you pursue any of that and great discussion thank you that's a great lead and so what what what Martin asked me to share with you a little bit is is to talk a little bit about the sort of the culture of innovation and how we get from technology if you will from some good ideas how do we actually incorporate that into into an organization so the first thing that occurs to me is that we need to be I think suitably informed by sort of some historical examples right so if you remember back at the in the beginning of the the first world war the doctrine about offensive tactics was ingrained in every Western army that there was in spite of the fact that this thing called a machine gun had been invented that and it took a half a million casualties at the psalm before we changed the tactics so just think about that I mean the the scale it's it's sometimes difficult for hierarchical organizations to incorporate that kind of change and why is that well because the people that get to the tops of hierarchical organizations with the exceptions of occasionally I mean I'm living proof the Marine Corps has a sense of humor but the in order to do that everything that you've done to be successful to get to that higher level is actually an impediment to the innovation that's required for that organization to continue to grow and it and it and it's that kind of dichotomy that is that I think is something that we have to just have to be aware of my experiences in in running the special purpose marine air ground task force which did all of the experimentation back in the late 90s and early 2000s in the Marine Corps it taught me a number of things and one of them was that we could use technology to actually change culture interestingly enough there's two ways to do that you could have a theological conversation that could go on for years and years and years about the reason why this particular organization should do things differently or you can very cleverly insert technologies that are going to cause changes in the way that people do things and allow them to see things in a different manner and I'll give you an example we were looking at distributed tactics right in the battle space and the thought process was for instance what I came to it from where I came to it was looking at the at the evolution of land warfare from the failings is in the in the Greek days where they all fought close together to the fact that organizations and and and units began to move further and further apart I then looked at my experience in aviation where you know way back in the day in the First World War they used to fly very very close to each other because they wanted to be able to see them they use hand and arm signals to talk to each other and then as technology allowed us to we got further and further and further apart and I noticed even going from certain type model series airplanes that I flew from the F4 into the F18 and beyond that we started getting further and further and now we're so far apart that we sometimes don't even see each other out there we couldn't conceive of that doctrinal change without the technology being there to drive the change so that's just a just an example and the trying to change the culture by saying we're going to change this is I mean that's the hardest thing to do the other thing that's difficult in organizations is to create in in hierarchical organizations like in a service a a culture of innovation that allows you to protect the creative so to speak that allows the folks that are actually really nugget away and trying to come up with these kinds of ideas that that to pride and to create an atmosphere that is more sort of Google like than it is Marine Corps like if you know what I mean that's it's it's more you know that doesn't mean that you have to I mean I try this and unfortunately people I work for didn't think it was really funny to have people come into work and flip flops and Hawaiian shirts and stuff just to try to at any rate but but there are ways that you can try to create and to try to create an atmosphere in which people are inclined to take some of the technologies that are he just talked about and to sit there and actually think through well what could I do if I had that so so it's it's a little bit of the what if what could I do and and what could that technology do to change the way I look at the world for example we shouldn't be surprised and Arthur and I talked about this the other day we should not be surprised at the way I still has used social media that shouldn't surprise us at all as a matter of fact we should have anticipated that another example is the just prior to the first world war with the advent of rail in and and the train system in in Europe allowed people to mobilize and move forces to the front lines at a rate that they've never been able to do before and what we didn't understand is that that same capability was going to box in the decision makers because they felt like they had to make a decision about going to war because if country X mobilizes they'll be at the border before we can do anything about it so there's just an example of a technology that of an unanticipated outcome of a particular technology so that's you know kind of in a nutshell I think that the concepts in some cases the technology follows the concepts you know in the inner war years the which people love to talk about but there's actually a lot of really fruitful things that we can learn from that that the you know everybody makes a has made a big deal out of this concept of blitzkrieg but the notion was really came from a lot of the constraints that were put on the Germans after the Treaty of Versailles but what they couldn't couldn't do and so they're driving around in the Schwabian plane there with cars that had plywood on the side that said tank on it and trying to fly airplanes around which they couldn't for many years because of the treaty that didn't have radios in them and they were dropping things and putting white crosses on the tops of the vehicles but when the wireless radio became practical and worked they had a concept for doing combined arms training that allowed them to use that radio in a way that that Europe hadn't seen before so there's an example of a concept a way of doing things that you start to experiment with with surrogates when we first started these distributed operation experiments in a Marine Corps we didn't have tablets like we do now we called them end user devices we didn't know any better we we were trying to develop a wireless internet which I carried out to 29 palms and had these little things about size of my hand and we set them up on six foot platforms and pointed them at the formations of Marines so that we could get some wireless connectivity and when we first did that and it didn't work I backed up and I gave every one of the Marines out there a five by eight card and a pencil and I said draw on this card what you think you need to know from this system and it was different than what we thought they needed to know by a they didn't want all the information we were producing they wanted to know where they were where their buddies were and where we thought the enemy was that's all they wanted and they wanted to be able to rely on that information so I took those cards and we conducted experiments and we would change the cards out and give them a card say okay here's what you know now what could you do about it and that helped us design if you will this wireless system that we used eventually that we eventually used now it took many years for that to become part of the Marine Corps because it wasn't the way we'd been doing things in the past and thanks to visionaries like general crew like it was the commandant who provided the top cover that allowed this the institution to go forward and we were able to do that so so that's just an example we had a I'll give you one more example and we'll throw it open to questions we had a command control system in this operation center and I had some folks under contract to me from I think it was Cal Poly to build algorithmic learning agents to help us make decisions and what I was trying to do is to get the agents to make sort of the housekeeping decisions so that we as humans could apply our intuition if you will and we used to bring the programmers the people that built these agents out to Quantico once every six or seven weeks and we'd run a two-day CPX command post exercise and they'd watch us operate and we would tell them what the agent would tell us and that we liked or didn't like you know when we had a for instance if you tried to call a fire mission in on a friendly force that was within a couple hundred meters you got these two big blue arrows on that and they just started pointing at each other and flashing there's a blue on blue kind of thing we had agents that looked at particular parts of the battle space for information kind of things and at the time there was people would come in and look at this and say well that's just great but we command and control better than anybody else in the world why are we spending all of our time doing this and so and the folks in my staff would look at this and say well this isn't the way we do things back in the battalion we do things this way and you're now causing this thing to be much flatter and it took many years before we institutionally came to the point that realized that hey this way of operating kind of distributed manner like this you know we sort of were forced into it in Iraq and Afghanistan and excelled at it so so I guess that message is I'm an internal optimist because it is hard and you get a lot of scar tissue trying to do this but that's not a reason not to so thanks very much to both of you I want to ask you both secretary work talk this morning about the the sort of the mess that we're in and that one of the primary ways that that we're going to try to get out of it and or that is necessary or respective of what happens to the mess is this new innovation initiative the soft set strategy that is as of yet not not well defined so I wanted a see if I can get a little bit out of you on that but assuming that that's not going to succeed we're probably gonna let the secretary tell you but I don't think I can do that so but one of the things I think that was a theme in your comments general Schmidel was the importance of experimentation and the role that that plays in bridging the what what I would characterize as sort of the technical technical cultural divide so let me try here do you think that that that ought to be a feature of this innovation initiative that is being started to and what how critical is that do we have the structure in the department to facilitate the kind of activities that are more like what outline so so I think that for openers there's a there's a number of different layers at which we look at innovation and experimentation right so it starts at the at the highest levels and OSD creating an atmosphere and helping to resource these kinds of things at the lowest level though where the actual experimentation really occurs is in many of the service labs it's in many of the other laboratories the private labs that are out there and we need to ensure that that continues and that they have the kinds of resources that they need to be able to experiment with that I mean I'm very very confident that at the tactical level the services are excited about this the private and public labs that I've talked to are excited about this kind of experimentation it's just that we need to I think ensure that they are resourced and again I get back to this issue about you know you got to kind of protect the creatives inside of a given organization and there's some of you in this room here that have experiences that are similar to mine and when you are sort of in it on a different wavelength and a lot of the folks in your service that can be good or bad but it's necessary to protect and encourage people to do that and to think that way and I think that you know I mean I applaud what secretary work is trying to do I think it's a it's it's innovation looking at a technology offset strategy whatever we're going to call it we that needs to just be it's a core competency it's not something we ought to layer on top of everything else that we ought to build a budget at the end of it say hey we ought to throw some money at R&D no I mean this is this is the core competency of what we do as a nation and what keeps us moving forward and and I applaud the fact that we're looking at it that way I think that getting from that level all the way down to the to the level at which the services are experimenting or the labs are experimenting we just got to push that and we got to push it hard and we got to when we have to encourage it but but I'm encouraged by the stuff that I see happening out there whether it's in the Air Force or Red Flags or in the Marine Corps or the Navy or the Army they're doing a lot of that at that level and it's you know we just need to encourage more of it to the truth yeah just to follow on you know when I I came to DARPA I returned as director in July of 2012 and what I found I had been at DARPA very early in my career 86 to 93 what I found was that by 2012 having come through a number of years of getting in theater working side-by-side with operational commanders bringing some of the technologies that we thought might be able to help solve real problems but also listening and finding out what the real problems were and then finding in some cases finding real solutions to those problems what I found was an agency that was actually much better in 2012 at bridging between crazy technology ideas and something that would make an operational impact than the agency I had been at you know two decades prior which I found very encouraging my concern was that you know really you have to be able to do this without being in an active war because otherwise that's not a good way to solve the problem and my concern was and remains that we I think we don't really have the ability that we need to be healthy in the department to bridge between what is created in the 616263 science and tech enterprise and actual operational capability when we're not in an active war when we're preparing or when we want to deter appear adversary and we're working on a generational shift in those technologies the the the question of experimentation I think it's very much it's something that's on our minds at DARPA it's something I think that the department's grappling with how do you start doing the field trials the early operational capability how do you start showing that things work not just in a lab but in a in a relevant operational context so that we can start moving where the dollars get spent within the services away from the way that we're doing business to the things that are going to be far more effective for the next generation so I think that's you know that it that's just got to be core to anything that we do going forward I can ask one other question before opening up to everyone so there is a at least from the ground forces in particular a strong emphasis on trying to better understand what the army calls a human dimension what at however you want to package it the improved ability to leverage use technology to leverage enhanced human performance and so I wondered if both of you could talk about from your respective positions what sort of where we are in that in that endeavor what the potential implications are and how they relate to a sort of more pure technology platform kinds of things and I just and I did the F-35 might come up I don't know but anyway so if you could address that you want to start off on some of the stuff and I'll get the implications you'll do the F-35 for sure I mentioned biology and the you know the most interesting species in biology is is us right so well certainly that's our view I think we're actually right one branch of the work that we've been doing at DARPA is in neuro technologies actually began with with questions about restoring function our revolutionizing prosthetics program work that we're doing to repair the kind of memory loss that often comes with PTSD some of the other programs but in as we set off down that path to restore function and some of the things that are coming out of that are I think going to be very powerful for the restoration of function but in in the work that we're doing as we learn more about the motor cortex as we understand the we hope to understand the the sources of neuropsychological psychological disease as we understand some aspects of how memory works in the brain all of those things that we're understanding from a restoration perspective the learning also opens very interesting new doors into what might be possible for in some cases the literal integration of human beings with technology and you know if you step back and you think about how capable human beings are and how amazing human intelligence is you think about the advances we've made in the last 40 or 50 years in the information realm and these incredibly complex systems that we that war fighters today their core job is is interacting with and controlling and mastering and using these incredibly complex systems but the interface between these two amazing capabilities is still a touch screen and a keyboard and a mouse and how how how that unfolds as we start understanding how the brain interacts with complexity I think the potential is is vast that's why we pursue it it absolutely raises all kinds of interesting questions that we as a society are going to need to deal with and and I think you know that's the nature of very powerful advanced technologies I want I think it's important also to just offer a little bit of a caution one thing I find is our military partners and customers for technology and in particular the soft community I think the appetite for anything that gives our human beings a little bit of an edge it is such a big deal it is so powerful I think the appetite is enormous and I think that in fact we that makes me in this case it makes me slightly more cautious rather than more aggressive because I think the burden falls on us to be completely clear about what works and what doesn't work and I think that this is also an area that is rife with the potential for getting carried away with ourselves because we want it so badly so I would just offer a note of caution about that after caffeine I think we should sort of take a deep breath and think hard about what the next human performance enhancement is really going to be so so two things occur to me one of them is that along with the with a caution that Arthur just articulated things that we understand about the brain are not necessarily things that we understand about the person about us I mean we are an embodied entity right and we are very suspect very subject to context and culture and the things that occur around us in determining what our behavior is and determining how we react to particular stimuli if you will so as we look at enhancements form enhancements one of the things you could look at is the enhancements that have occurred for example in the world of professional sports over the last 30 years if you look at films from the 1960s or 70s of NBA basketball players and then you look at films today you'll see a striking difference in technique and speed and everything that that is about that same thing with any other with any other sport so if that's the case and we have the ability to into to increase our performance enhancement vis-a-vis those kinds of motor skills then the question remains so how does that translate does it translate how can it translate to our cognitive processes and our ability to be able to understand more of the environment to understand it quicker you know we've always wanted as humans to be able to make better decisions faster and be smarter about it I parachuted into an experiment years ago and they were looking at trying to provide the commander of this organization with all this information and the hypothesis was if he knew ten times more than he knew before he would make a better decision so I was an observer in this thing for about three days and what I discovered is that in fact what was happening is he was making decisions he was taking longer and longer and longer to make decisions as opposed to doing it quicker because once he became accustomed to all this information he said well shoot if I wait another five minutes another three minutes I'll know this much more so as opposed to making a large number of smaller decisions we found ourselves in a situation where we were making one big war-winning decision at the end of this which was the exact opposite of what the hypothesis was for the experiment to start with so as we look at this it's important I think for us to keep in mind that at the end of the day it is all about us and it's about the way we make decisions and it's about the understanding the cognitive and social psychology that surrounds our decision-making processes and what we react to and how we react to those decisions and and so the physical enhancement of this is one piece of it so we can carry more we can run faster we can jump higher maybe we can even stay up maybe we don't have to sleep for days at a time and still retain some cognitive capability they don't know what all that could entail but at the end of the day it's our understanding how we make decisions one of the things I was monkeying around with years ago was trying to get these algorithmic agents to make all these housekeeping decisions about numbers of things I didn't want the any of my staff to be involved in trying to calculate the number of tons of fuel that had that we could find other things to do that what I wanted them to calculate was why is that unit X using so much more fuel in unit Y is there a reason for that and if there is let's understand it and see if there's something there that we need to change or not change so I mean I'm absolutely with you're at the I the technology revolution and human enhancement is something that's staring us in the face right now but it's been kind of sneaking up on us for a couple of years but I think that if we if you look if you're looking for a place to use as a model if you look at the enhancement of in the world of professional sports and then just see where I mean just just watch somebody dribble in a 1960s Lakers team as opposed to what you see today it's really quite impressive the difference and so we can get there the question is how do we get the decision-making capabilities that cognitive abilities because ultimately we are social creatures I believe and our culture has in context has a great deal to do with what we decide and how we decide to do it so okay thanks okay so I've got a hand up over here and then we'll come here people again could raise their hands they'll bring the mic and if you could briefly identify yourself before asking a question great great thank you wonderful discussion Doug Brooks from the International Stability Operations Association first of all I always pictured DARPA as a bunch of 20-something year-old sitting around dropping heroin and coming up with ideas please don't dispel that my question is you have to work with non-traditional contracting contractors and the non-traditional contracting community with Silicon Valley a lot more than than you know the the old contractors and so how is that going to impact a Opsec how's it going to impact on on how the development is done when you're working with people who have never had to work with the Department of Defense before and coming up with all sorts of wacky ideas that maybe don't want to leak out so you know there's I find nothing really new about the communities that we need to work within our working with it's all the way back through DARPA's history we've had work that focused on things that were directly tied to clear military need and systems and that would transition to the services and almost always that is done working with the traditional defense contractor base because they're the ones that are going to carry it forward but again throughout many decades of history we've also had these enabling technology investments that ARPANET that led to internet the microelectronics and RF semiconductors and MEMS revolutions all the groundwork that got laid we made those investments because we needed it for the national security capabilities but of course many of those technologies have also built massive industries and changed how we live and work and the nature of that work has has historically and continues to be that we work with universities and companies of all sorts including those who don't consider themselves part of the defense contractor community so I don't really see a big change but your question is still completely relevant I will tell you that the way I think about it is that our work is a portfolio of programs at the moment for example we've got about 200 active programs at DARPA some of those in the portfolio need to be classified and need to be done in a very controlled fashion but many of them actually need to create this foment in and generate new technology revolutions that will happen in a very public way and then many of them actually will need to get commercialized before we can harvest them for DOD that was certainly true of the information revolution I think it's true today in robotics as an example so not only do we not worry about keeping them classified we actually want them out in the world we recently established an open catalog as an example where much of our published research is widely available where a lot of the software tools that we're building that are open source are available and I was pleasantly surprised by the reaction that we got from some of our military customers and partners I thought they might be worried about OPSEC but in fact they said thank God we can never actually get our hands on anything now I can just go to this public website and get this piece that I need and then I can go build the protected capability that I want to build so it's an ongoing question but that's how we think about it I think here and then we've got over there and then there and that so here first Colin Clark breaking defense it sounds a great deal as if we're looking at capabilities that will merge data fusion efforts cyberkinetics or whatever we're going to call them does this lead are we finally going to be abandoning these sort of industrial warfare electronic you know older electronic warfare and moving to at least sort of commando versions of I don't know if we want to call them boosted people but augmented by you know things like the data fusion engine the F-35 is gonna have where humans aren't as important in terms of lugging enormous amounts of equipment but can make decisions and redirect other weapons well so let me let me tackle it the fusion piece first so I think that the issues about data fusion are this fundamentally the same kinds of things that I was trying to get at years ago which is that if you confuse data if you can have data fusion capabilities that are able to that are that are technologies that are able to fuse the kinds of data that you would want to present to a decision maker to make a decision important thing is that the decision maker has some say over what that information is and how it's presented when you talked about the industrial age piece of that I mean like anything else I mean that's going to take years and years of transition but I would suggest that one of the things that we can see today in the in the world of cyber for example in another venues is the blurring of offense and defense and and I think that we have historically looked at those two things as you know conceptually is two different things we don't operate that way I mean you don't have an offense and a defense inside of an operation center but you can sexually think about for example for traditionally in the offense we would command and control very much hands-off and you would allow units to go and do things that they needed to do mission type orders I mean in that kind of thing whereas on the defensive side it was a tendency to have a more centralized command and control because you were trying to influence if you will a smaller piece of space or time but I think that that is blurred and I think that the the notion that you can ever be on the offense without being on the defense in at least the same sense of effort is is gone I mean in the world of cyber I think it's naive to think that you can do things that you can do anything without having both of those things in mind and being giving equal shrift to both of them and realizing that there's a real fuzzy line between offense and defense and not only conceptually but in terms of execution so so I think that that's ultimately where this is going that that line is going to be it's going to break down but like anything else it's going to take years this the industrial layered mind I mean that developed because it's easy for us to understand that kind of hierarchical sort of you know this you know the Newtonian linear based the way things are even though we instinctively and intuitively know that that's generally not the case even our buddy cause which there's accused of being that way he says that he when he talks about Napoleon he starts talking about genius because he can't figure out any other way to explain how he did what he did at Jenna or Auschwitz or ours that so that's great. There sorry. Is that yes hi thank you it's really interesting to listen to you my name is Hannah Roslin I'm a research professor at the Norwegian Cyber Defense Academy and a visiting fellow here at CSIS and you talk a lot about you know that how technology changes people it obviously changes how we are what it means to be human and I see with our students as well the young cyber officers and those training to be cyber professionals they cognitively relate to technology in a very different way than their superiors do they just use it differently they think about it differently and they don't necessarily even understand when their superiors try to explain technology the language they use we lack some we have a challenge of finding a common language actually that we can at their superiors and say 50s 60s try to explain multirate technology and you see that the 19 20 year olds they don't necessarily understand so there is and also inside the military it's a potential friction between the you know the tech savvy skill set that they have and the military hierarchy because they at the same time lack multi-experience military know-how we have obviously our challenges and how to deal with this I would just it's so interesting to listening to you so I was hoping you could comment on how the US military will relate to the challenge of incorporating that skill set of the youngsters of today I would just say that's how revolution happens sometimes it's a generational shift and we we're seeing it exactly the way you described one simple example is we've been working on it's robotic essentially a robotic mule the question is can we help Marines with the load that they have to carry it just to be very clear this is not something that's going to transition in the Marine Corps in the next 20 minutes really what we're trying to do is give them an example of a robotic capability so that we and they can explore together how real human beings will interact with these machines and you know we have all the PowerPoint conversations and and we all have these sort of theoretical conversations and then we get it in the hands of actually young Marines and they are all over it and they are interacting with this technology in a way that's completely natural to them and so they're going to tell us or we're not going to tell them and I think that's the way it should be and it's it's good and constructive so so one of the things that that I noticed as a 40-year-old that used to try to explain this that to youngsters how do you do well thank you so one of the things that I noticed when we first started doing a number of these experiments this is back you know 10-15 years ago with what was then very nascent sort of computerization and interfaces if you will the human and GUI they used to call it the interfaces is that we folks in my generation would get very frustrated when the thing didn't work the way it was supposed to and what I noticed is that the youngsters didn't get frustrated at all they would just sit there and say okay I guess that didn't work and they'd go do something else and I'm ready to take this device and like throw it as hard as I can against the wall so there clearly is a difference in that and and you're absolutely right there also was a difference in the way the technology is presented to you the information is presented we have noticed that there are different pathways neural pathways that that we are using as in trying to understand that so we there's a couple of ways that we can go about doing that somebody asked a question before about some different kinds of of contra research folks and so a number of years ago I hired this guy who was a brilliant brilliant man he added two degrees one in philosophy and one in neuroscience and I brought him into the command center and I wanted him to talk to us about information displays and show us how this is how other ways to think about this and he had a full-on ponytail I mean doing his best Jerry Garcia imitation he wore tight-eyed shirts all the time and my captain's I was a colonel of time used to call him dr. tie-dye and and he was out there he was way out there and he was awesome because he came in and he looked at our way we were displaying this information and it was all the way that we'd done it for many years on maps you know with this cold 2507 whatever he'll did technology that we used to display this is what a tank looks like it's a square box with an X through it and something squiggly on the top and you know that look like a tank it's like a foreign so he had another way to display this information and it was a picture literally it looked like the Starship Enterprise and the engines would would would heat up and and the colors would change based on what the enemy was doing and we never did get there but it was a phenomenal way to look at this and where we did get to was picturing things as opposed to having a symbol that required another okay I know what that is now I'm gonna do something else as opposed to a picture of a thing what I think we're seeing today with a lot of the technologies is that because they're using different pathways because they are taking this in differently or these absolutely right this has got to come from the bottom up again and I think that the hardest thing for the the for other generations to do is to let go I mean we just talked about hierarchies right and it's just so hard to let go and it's hard to let go in a way that that not only do you put up with that behavior but you encourage it because you know the institution has got to have that kind of blood in it if it's going to continue and if it's going to get better it's going to survive so I mean you've hit on on the the challenge for us in the leadership role at least the way I see it is is to understand enough about the technology and enough about the way they're using the technology I go back to what I said earlier why would we be surprised that Isis would use the why would why would that surprise us at all it shouldn't have even said we should have been surprised if they didn't so just you know okay I don't want to give short shrifted this side of the room but okay all the questions are right here and then we'll go in the back back there that's odd David Scruggs ran a song strategic advisors my question was just about one of the three main themes that Darb is working on today the first one you mentioned about fundamentally rethinking complex systems can you give a little more color on that can you give a couple of sub themes and are we talking about just platforms a lot of it is platforms but it's not limited to platforms let me give you maybe two examples that are a little bit different space we are everything that we do militarily that gives us overwhelming advantage is critically dependent on space whether it's the position navigation and timing with GPS or communications or ISR everything and I think it is become you know just something that comes to you from above and people sort of take it for granted but we're dead without it we don't control space the way that we used to and not only because of other nations aspirations and actions but also because of what's happening commercially on orbit our paradigm for space for all for good reasons we have now arrived at a place where the way we access space is we build a phenomenally massive expensive satellite we put it on a phenomenally expensive rocket it takes a couple of years sometime I take longer than that to schedule a launch and then then you got what you got and and it we are incredibly inflexible and there are things that we need to be able to do from orbit that we simply can't even do that way so our question is can we completely rethink that can we change the way that we do launch a couple of our programs aim to dramatically reduce the cost of launch to a low earth orbit the cost is I think important but actually even more impactful with these programs would be achieving 24 hour call up so that you could go from any airfield around the world on a day's notice and be able to get something on orbit that would be a major change and that coupled with what is already happening with small sats including commercial activity you can start to see how those that's already I think there's a lot of momentum there if we can add the launch piece that starts changing it in a fundamental way when you start having those new technology capabilities then you can start thinking about new architectures to do the jobs that you thought about doing on orbit and think about what's possible now from the cooperative actions of multiple satellites so that that's one example a very different example might be thinking about how we operate in the electromagnetic spectrum again we have these massive is arrays they are the envy of the world there they're really pretty awesome what we can do but because they take so long to develop and because their capabilities are fixed when the adversary uses commercially available technology literally that they can buy with a few clicks from a keyboard to come up with missiles that might threaten our capabilities we you know when we go back to the drawing board it can be a 15-year block upgrade cycle it you know we might get lucky and it's just something that takes a few months to go you know diddle around with the algorithms but neither of those is adequate for the world in which we need to be able to know in real time in the battlefield what's going on and adapt and and and achieve the control of the electromagnetic spectrum that we want so finding ways to do that that the underlying technology is so powerful and the advances that are still that we that we see in the lab are so far beyond even our impressive capabilities in the field our question is instead of just continuing to build big monolithic systems with those lab capabilities can that next generation be distributed coherent systems that would give us the ability to you know have not one monolithic target but have this very graceful degradation have the inner have made it much harder for the adversary to know where the electromagnetic radiation is coming from a capability that can be upgraded much more gradually much more gracefully and quickly over time so those are so I think it's a it's a general concept but how you really implement it is going to be different depending on the specific domain and the challenges in the back over there and then we'll go there and then right here in the back row right there yeah hi thank you very much extremely informative I have two part question one what do you want the American public to know about this remarkable description that you've given us today and how would you do that we've gone the news hour we get interviewed by Charlie Rose it's supposed to be an open-ended question second part relates to our allies Japan Germany England whom UK rather whomever is are there DARPA equivalents out there that you'd like to tell us about that we partner with to the extent that you can discuss that let's see so I think we try very hard to get the story out about what we're trying to create in the future for a couple reasons one is simply because you know DARPA's 200 government employees in a in an office building in Arlington Virginia so we don't do anything without engaging 200 government employees right because all of the work gets done in this you know we tap the the intellectual smarts in this vast technological community which means we have to be able to tell them where we're going if we're going to get their engagement so we try very hard to get the word out but also because at the end of the day it doesn't matter unless we get it transitioned to operational capability and you know that's why I I'm particularly enthusiastic about the fact that the department is working on how we're going to get better at moving technology from the innovation stage to actual operational capability so I think that's important other you know a lot of our focus tends to be on working with the technical community we occasionally will will do something in other countries or with other agencies in other countries but that's not our primary focus virtually every major defense actor around the world has a has an R&D capability of some sort there been some of them have some resemblance to DARPA but there's not really a one-to-one map I would say can I follow up on that and ask a little bit about your relationships with IARPA and DHS I don't know what the DHS version of but I know they have one as well so the interagency ARPA's how you interface with them yeah just whenever there's an opportunity that we're working on together I you know of the of the organizations that have ARPA and the name meant they're doing a variety of different things H.S. ARPA is not you know doesn't particularly have a mission that resembles what we're doing ARPA E and the energy department is actually the one that is probably the closest to having the ARPA-like model of you know a deep focus on breakthrough technologies and understanding that they're in the revolution business and you know being tolerant of risk and being willing to fail in order to reach far ahead but I think that's actually a great thing to see different parts of government nurturing but I think it's implemented in very different ways. If I could respond to your comment about the allies before it's it's been my experience that when we look at how some of these technologies and concepts would transition into the services or into the way that we actually do the war fighting part of this there is a great deal of interest on the part of the allies and then many of them are very involved in this for a number of reasons one is technical they want to keep up with us so they can stay connected and plugged into what we do and the other is conceptual they want to know why we're thinking about fighting wars differently and what we think is is important and is so there there actually is a very rich dialogue that goes on here and and it's made richer because the allies all bring a different perspective to this than we do and we sometimes forget that the way we do things is not necessarily the way the rest of the world does things and people view us differently obviously than we view ourselves so at least you know at the tactical level from what I've seen it's been a very healthy relationship and I think that you know that that as already said there's other connections that that that also rich in that relationship so I think I had I had I promise it some in this on this side and now I've lost them there and then in the back right there thanks thank you both for thank you all for this amazing discussion this I'm Matt Jones from the Boeing company and just wanted to ask first of all thank you Dr. Prabhakar for alluding to Alasa and XS-1 hope those are successful to make them work now that's the only thing ever since World War II I would argue the US has enjoyed technological superiority or at least parody with potential adversaries and we've been happy to invest heavily for that the world has changed a bit that's not several changes you've alluded to the rapid acceleration of the pace of technological advancement the rapid proliferation of technological advancement and know how around the world and closer to home as we discussed downstairs we're in a fiscal austerity period for defense spending that this could argue we last longer than typical downturns have given those how do you see DARPA and the department dealing with that cost imposition culture that we've accepted to get to that technological spirit I for so long we probably can't afford that we can't afford to spend whatever it costs you mean where we impose costs on ourselves yeah so how are we addressing that both technologically in your organization and general culturally in the services to start thinking about how can we do great stuff for a lot less money just as a just a way of not accepting that we're always going to be able to just spend our way out of any kind of technological threat yeah I'll just say I think that that is the one of the core underpinnings of the work that we're doing in rethinking complex systems because for a lot of reasons the cost issue that you mentioned but also simply because of the power that we're going to be able to achieve by going down this traditional path of big monolithic systems I think we're just out of steam on that coupled in with that I I think it what's topical today is to talk about defense downturn and sequestration step back one more step and if you look at the investment that this country in the United States that we make in research and development today one third of that comes from the federal government and two-thirds of that comes from the private sector it's not that long ago it's only a few decades ago that that that that that national investment was dominated by public sector spending now public sector investment R&D has plus or minus kept pace with GDP it sort of bubbled along and grown at about the rate of GDP declined a little bit private sector investment R&D has grown faster than GDP that I mean that is the innovation economy that we're all very aware of that's what that means and it to me there's a much more fundamental structural shift that has happened and it's about the fact that technology is you know happily right I mean in DoD we think oh this is really bad news because we're not in charge of this anymore in fact this is good for humanity that technology is being driven for these for these market reasons and to improve lives around the world but it means that we we is exactly what you said we don't not only do we not have technological parity but what's available to anyone around the world nation-state or individual at the top of the keyboard is phenomenal technological capability and I think we are self-impeding in in our lack of ability to grab it and use it and run harder and faster than anyone else that's that's the thing that I think is that's the piece of the puzzle that the technology community needs to take on many other issues about acquisition and Congress and the requirements process all of those are real but the part that the technology community has to take ownership of is can we come up with new architectures new way of new ways of thinking about complex systems that allow us to tap that very powerful commercial technology and the pace at which it moves I think I think if you look at the one of the the interesting things that you discover when you look at the history of say the last hundred years or so is that a lot of the military innovation that we celebrate occurred in periods of economic downturn and in periods of austerity and and I think that that's that that occurs I mean there's all kinds of examples but the I think a lot of that occurs because there's a natural tension that develops and tension is good tension is what breeds creativity is what breeds these kinds of thoughts and so I think that there there actually is something to be said for the opportunities that exist even even in a in an economic downturn with regard to the technologies one of the challenges north you mentioned at the end here is that once you as a service when you make a decision you're going to start to transition something into into a long term program the the the art in that is to ensure that there is still room for innovation in that program as it goes forward because the technology is advancing so much faster and leaps and bounds than it did even 20 or 25 years ago that for us to lock something in and say okay this is going to be what we're going to use to fight for the next x number of years whatever that happens to be we may find it to be outdated shortly after that or we may find that we don't have the ability to change it in a way that keeps up with the the way the technology in the way that the that the industry if you will is changing so you know I'm kind of a I am a glass half full sort of fellow in this regard because I do think there are some opportunities out there for us to think and and to leverage some of the commercial technologies that are and we spent many years back in the 90s with no money using cots equipment that we literally bought it at you know at the commercial marketplace down the street and took it out in the field saw see if it could work and we make this do something can we run can we do combined arms you know sorts of operations with this technology and the answer was yes we could so and it you know years later it found its way into the normal system so right here hi there scott augen boss CSIS I had a question sort of going back to a lot of the themes you've talked about you've talked about data you've talked about knowledge filtering and culture and so one of the things I wanted to ask is as you look out over the next 10 or 15 years and you look at an increase in data and thanks for releasing those ex data tools that's a great catalog project at DARPA the graphical interface hasn't really changed we've added new things to it like voice communication heads up displays and we've made improvements in some of these technologies but it doesn't really change the game enough and if you look at pilots and all sorts of military leaders going forward they're gonna have so much information coming at them I guess the question I have here in this space is culturally I'm not sure if the US and many other countries were ready for algorithms and robots to be sort of the leaders in the battlefield going forward and I was warning to that extent how much are you thinking about that cultural aspect of some of the changes by adding more algorithms and more computational models that provide a more autonomous fleet going forward and culturally what that will do in terms of the job losses that'll likely occur in that same time period yeah well I don't first of all I don't think about algorithms and robots as leaders right I mean that's actually that's the thing that human beings are gonna still be the best at so just to just to nitpick with your language a little bit you know I want to make a really important distinction between we use the term autonomy often to mean a whole host of things let me separate two pieces of it one is over time what machines and systems are capable of doing and in fact as we know technology is progressing at this phenomenal pace and yet capability is is is advancing quickly can continue to advance into the next decades the the separate question of autonomy is a question about rules of engagement and it's a question that applies to the autonomy that we give to the humans involved in warfighting as well as about the machines that are involved in warfighting and and I think it's important to make that distinction because I think it's gonna help us ask a higher quality question when we when we deal with those kinds of societal choices that you're talking about because I want us to have that debate but I want us to have a really smart debate about it and not a Hollywood debate about you know Terminator right so if you do that I think what I would tell you is my objective is on the one hand to build the technology capabilities that are increasingly powerful because I think that is the only way that we're going to get ahead of the threat that that we are dealing with enabled by global technology but with that it's going to be I want to build into the technology the ability for the human for the warfighter to choose the degree of autonomy that they that they confer to the technology depending on the situation and the needs of the moment and and that that that control structure I think is actually needs to be core to how we think about how we use the continuing performance capabilities of the technology so I think that first of all I finally agree with what you just said when you mentioned that thing about algorithms and autonomists this we need to remember that we're the ones that develop those algorithms we're the ones that empower those kind things to do the things that they do it actually brings up a larger ethical question if you start to talk about that and that is if you are if you are trying to give to some kind of autonomous agent if you will some ability to make decisions based on these algorithms you've given it on things in the battle space we probably need to understand to try to understand what we are basing what worldview we're using to base those decision that decision matrix on I'll give you an example if you believe for example that there is one answer to every question and that answer is out there it's out there it's an objective answer it's waiting for us to find it and determine that is that that's it if you believe that then you are going to program those algorithms to try to find that one answer in amongst all of that noise on the other hand if you believe that there's a more relativistic aspect of how you make decisions because you take into account the context the culture all the other things that are out there then you're going to build an algorithm that's going to try to understand all of those things so when you take and you have a pilot in a cockpit that's doing close-air support and everything about the target he's getting ready to hit he's met all the requirements all the objective requirements yet when he rolls in he looks at it and something is just not right and he doesn't drop and it's understanding what that just not right thing is do we want that to be part of a of a decision matrix and if we do how are we going to develop and program the things to do that so as as you think about that I think there's a lot more that we need to bring into account here other than just the technology to go back to trying to understand how we make moral choices ethical choices how we do those things as humans and understanding ourselves a little better so just a thought well I thank you both for what I knew would be a very rich conversation I to me my I guess my big takeaway at least at first blush is innovation is really hard because it starts with people and ends with people we're really hard but that's also the best part and that and also that there is a lot of promise and some amount of risk and how we as people find our way through this is perhaps the challenge of the day so very grateful to both of you for a wonderful discussion and good luck in reading us through all this thanks for having us take the elevators down to the concourse level you can get food there