AI and Drone Warfare: How New Technology is Transforming the Future of War

BIANNA GOLODRYGA, SENIOR GLOBAL AFFAIRS ANALYST: And now, it’s the stuff of movies. What will waging war look like in the future? Well, our next guest can tell you. Raj Shah and Christopher Kirchhoff launched an elite Pentagon unit aimed at integrating cutting edge technology into the U.S. military. Their book is called “Unit X: How the Pentagon and Silicon Valley are Transforming the Future of War.” And here they are with Walter Isaacson.

(BEGIN VIDEOTAPE)

WALTER ISAACSON, CO-HOST, AMANPOUR AND CO.: Thank you. And Raj Shah and Chris Kirchhoff, welcome to the show.

RAJ M. SHAH, CO-AUTHOR, “UNIT X”: Thanks for having us.

CHRISTOPHER KIRCHHOFF, CO-AUTHOR, “UNIT X”: Thank you, Walter.

ISAACSON: Raj, let me begin with you. There’s a wonderful scene of you piloting an F-16 fighter during the Iraq war. You’re on the border of Iraq and Iran. Tell me what epiphany you have about what happens then.

SHAH: It was a very interesting time, Walter. So, it was my first deployment in 2006, and it was my first glimpse of really understanding how divergent the Department of Defense’s way of buying technology and developing it was from the commercial world, particularly around software. So, I’m deployed there. Our missions are to support ground troops, often on the border, and we’re flying at 500 miles an hour in a circle. And while the F-16 is an amazing airplane, it at the time did not have a moving map GPS. So, I could look down the screen and tell you which side of the border I was on.

ISAACSON: Wait a minute. This is a type of thing that back then I had in my car, right?

SHAH: Exactly. So, that same year, while I didn’t have it in the jet, I could take the predecessor to an iPad and have that in your car or in a civilian airplane I was flying. And so, many of us there in the squadron had these, you know, iPads in the in the ready rooms for playing games. And so, we would strap them to our legs and fly with them because it helped us stay on the right side.

ISAACSON: Then you go, I think, to Qatar a decade later. And tell me how things had improved.

SHAH: So, a decade later, I had the chance to visit with you, which was a wonderful trip, the command center that ran air — all the air operations for Iraq, Afghanistan, and Syria. And candidly, it hadn’t really changed that much since when I was in uniform that — a decade prior. And we saw a lot of the different applications and processes that people doing. There was one that was most interesting, which was how do you plan where an air refueling tanker should be to give gas to a fighter jet? And that plan was being done on a whiteboard, with magnetic pucks and dry erase markers, and it would take this team of officers 60-man hours a day to make that plan. And so, this is one, on that visit, that with the help of the defense innovation board, we said, we’ll go solve. And so, the unit Chris and I helped run built software, to automate that system. And what we discovered at the end was by using this automated system, we could save three or four scrambles a day. And what that means is, because the system was so hard to transform, every time a soldier came under fire from an adversary, they would scramble fighters to protect him, then you would have to scramble tankers. And those tankers were sitting at bases around the region. It saved three or four a day. Each one of those costs about $250,000. The tool paid for itself in just a matter of a week, but more importantly, it started this revolution of software factories and an emphasis on how important software is for the military.

ISAACSON: And it was part of the Defense Innovation Unit Experimental. And, Chris, you got involved with it as well, not from a military background. Tell me how you got involved and what you thought when you’re on these trips with Raj.

CHRISTOPHER KIRCHHOFF, CO-AUTHOR, “UNIT X”: Well, you know, in 2016, it was such a moment in technology. You had companies like Amazon, you had companies like Apple and Microsoft. And, you know, Walter, each of those companies had a market capitalization that was itself bigger than the entire U.S. defense industry combined. And so, what that means for the Pentagon is that the locus of innovation, which, you know, used to live in labs like DARPA and other military R&D labs a generation ago, the locus of innovation had completely shifted to the commercial sector. And so, Secretary Ash Carter, who founded Defense Innovation Unit and appointed Raj and I to lead it, he recognized this trend, this generational shift, and he knew that that for the military to keep accessing the best technology in the world, it would have to become, in his words, a fast follower of the commercial sector. And so, Defense Innovation Unit was established to access that whole other technology economy that in 2016 was not doing business with the Department of Defense.

ISAACSON: Let me read something from the mission that you all had when you were helping to found this unit. Our mission was to disrupt and transform the culture of the largest and possibly most bureaucratic organization in the world by infusing its clogged arteries with the nimble, agile DNA of Silicon Valley. In other words, to hack the Pentagon. First, the Pentagon. Why were the arteries so clogged?

SHAH: I think, you know, these two worlds had just drifted given the advancements of technology. So, the systems that the Pentagon had and still have are really good for buying an aircraft carrier. Something you’re going to keep for 50 years. And you really want to make sure that all the planning and sustainment is done well. But when you think about the advancements of software and how fast it changes, right? Our iPhones get updated on a daily basis. Our systems weren’t — were not built for that. And, you know, the real reason behind that is if you think back into the ’70s and ’80s and when we won the Cold War, military and government R&D was the largest source of innovation in the world, right? It would trickle down from that to the commercial world. That’s completely flipped because of the huge consumer market for iPhones, for drones, for commercial technology. And so, all the great R&D is happening in the private sector. And so, our systems needed to be reformed to be able to take advantage of that. But, Walter, there’s one other really key thing, right, because at the end of the day, this is a human — these are human endeavors, right? There are people on both sides that need to work together. And so, when Secretary Carter came to talk about his desire to build DIU in the Valley, it was the first time a sitting secretary of defense had come in 20 years. This was also just after the Edward Snowden revelations of how, you know, our intelligence agencies were, you know, collecting information. And so, there was a massive amount of mistrust. And so, that was part of our mission as well, was how do we just bring these worlds together? And so, we had to really rebuild relationships.

ISAACSON: I remember reading in the book, it was so fascinating that Congress becomes a problem for you. You go to the Hill and they say, wait, you’re spending this money with these companies in Silicon Valley. Why don’t you spend it in Indiana, which is a District I. So, what’s the problem with Congress? Does it make it worse, Chris?

KIRCHHOFF: You know, it was a challenge, Walter. We’re not going to say otherwise. In fact, you know, so Secretary Carter flies out an Air Force Two, some of the most powerful people in Silicon Valley come to our unveiling when Secretary Carter names Raj and I. And two days later, we get a phone call from a friend of ours in the Hill that says, I have some news to share. Your budget’s been cut. And we said, well, what do you mean? How much has it been cut by? And he said, that’s the thing. You’ve been zeroized. Your entire budget has gone away. Your unit is being shut down by Congress. So, our first trip to Washington, it was supposed to be a victory lap. We were supposed to be meeting with the service chiefs, the service secretaries to figure out what their top priorities for us to work on were. And instead, we ended up meeting with two junior staffers on the House Appropriations Committee who had decided for petty reasons of their own that they were going to nix Secretary Carter’s high-profile initiative to get back at the guy. So, that’s just one illustration of how much DIU had to cut against the grain, not only with Congress, but also, frankly, with the Department of Defense. Many of which, many of whom in the department were skeptical that Silicon Valley could actually provide technology that they could rely on in war.

ISAACSON: Are you worried that too much innovation could lead to humans not being in the chain someday?

SHAH: It’s a really interesting debate, Walter, as to where does the line as A.I. and autonomy continue to grow? And the department’s been pretty clear that for a life-or-death decision, there’s always a human in the loop. But I think, Walter, you also highlight a rapid change in how warfare and how we’re going to deter war is occurring. Chris and I had the opportunity to visit Ukraine in October to see firsthand how drone warfare is just changing our ways of thinking about conflict. So, for example, just recently, it was revealed that we had sent 31 M1A1 battle tanks to Ukraine. These are the most sophisticated tanks in the world. And 25 percent of them had been disabled by Russian kamikaze drones. So, they’ve had to withdraw all of them from the front lines. So, here’s a situation where, you know, since World War I, you know, mechanized armored warfare has been our mainstay and now cheap, low-cost drones are challenging that. And so, we’re going to have to rethink our whole way of how we fight and what are the right tools that we’re going to equip our men and women in uniform with.

ISAACSON: Well, wait. Let me push you on the moral question. Suppose you were able to develop something that used visual and face recognition and pattern recognition, and it was 10 times more accurate than some, you know, human pushing a button, do you think we should go to all autonomous warfare?

SHAH: These are challenging questions, Walter, that we’re going to have to deal with. You know, if you think about it, where if autonomy is going to lower collateral damage, it’s going to make us more accurate. You know, those are all good things. So, how do we incorporate that yet still have humans ultimately making the decision? And I think, look, the other thing is our adversaries, right? They have a vote in this matter, too. So, if a potential adversary is using this technology and they’re using it differently than maybe we are, how do we respond to that? And how do we do it where we maintain our, you know, sort of American liberal western values yet ensure that we deter conflict? This is something that people that both understand technology as well as understand warfare need to come together. And I think, again, this is where things like the Defense Innovation Unit will have an important role to play.

ISAACSON: So, Chris, among other things that happened in Ukraine when the war begins is all of our defense satellites, all the commercial satellites get totally wiped out by the Russians, except for Starlink, by SpaceX. And suddenly, you’re depending in Ukraine on a private commercial company that does it better. Is that a good or a bad thing, Chris, to have to say, wait, we don’t need Boeing and Lockheed to do it. We can contract these things out to nimble companies who can do it better?

KIRCHHOFF: Well, Walter, it certainly is a thing. You know, and as you write in your biography of Elon Musk, the fact that a company like Starlink, controlled by a single individual, who, you know, is now making decisions about who to extend service to and who not to extend service to is a new reality in the world. And again, this is a part of the shift that Raj and I confronted firsthand of the technology ecosystem met to service the consumer market, just exploding in size. And so, increasingly it’s going to be private companies like SpaceX, like Palantir, like the new breed of Defense Unicorns, Andarol (ph), to Palace Space, Shield AI, that are going to be producing really powerful technology that will be central to how the U.S. military defends the country.

ISAACSON: Wait. Let me push back on that because I read in your book when you have all these grants going out, instead of going to all these startup and venture backed companies, most of the grants are going to the big bloated old defense contractors doing cost plus things like Lockheed and Northrop Grumman.

KIRCHHOFF: Well, Walter, you’re exactly right. You know, the Department of Defense has a long way to go. So, at one level, it’s made very impressive strides, even in the last 12 months, by elevating the unit we used to serve with, Defense Innovation Unit, to a direct report to the secretary of congress enshrined its mission and law. The budget of Defense Innovation Unit is now a billion dollars a year, with even more money being spent on it through the replicator initiative, the most important initiative started by Secretary Lloyd Austin to develop autonomous swarming drones. So, all that is good. But Walter, you are right. If you look at what percentage of the Department of Defense budget is actually going towards new systems rather than legacy systems, it’s still a fairly small fraction. And we are at risk as a country, because if we are spending most of our money on things like tanks, which now can be demonstrably defeated by low-cost drones, or for that matter, aircraft carriers, which as we know, are at increasing risk of being defeated by weapons like hypersonic missiles, we could end up with a military that might not be as survivable in a conflict of the future. And so, I think people like Raj and me and other people at Defense Innovation Unit are going to be looking to see how much of the defense budget continues to swing towards this new generation of companies and weapon systems and how much of it swings away from the traditional weapon systems that for years we have procured.

ISAACSON: You talk in the book about the civil military integration that happens in China. Are they doing it better?

SHAH: Well, they’re being more directive, right? So, China can go to any startup and say, you’re going to have a CCP member on your board. And in time of war, we’re going to push you, right? That’s not how we do business here in America. I think our version of civil military fusion is working closer with places like Silicon Valley, putting those companies under contract, and try to incorporate that technology. And really, you know, I think the American system of free flow of capital, free flow of talent is really how we win, and we double down on that. And I think we’re already seeing that with generative A.I. and the rise of those great companies here and not in Russia or China.

ISAACSON: What implications do these different approaches to defense innovation have in relationship to our competition with China, and especially the fact that so many chips are made in Taiwan?

KIRCHHOFF: Well, Walter, it used to be said that wars were fought with steel, and it’s very clear today they’re going to be fought with silicon. And so, it is unquestionably a major security vulnerability in the United States that all the advanced chips we use to power our automobiles, our phones, our computers are made essentially in a series of fabs that exist on the east side of Taiwan facing China, 70 miles off the Chinese coast. And so, that’s a great worry. It’s the reason why the U.S. has passed the Chips Act, and it’s something that we need to all focus on as a nation to make sure that it’s not just Taiwan that has the capacity to make the most advanced chips.

SHAH: And, Walter, maybe if I’ll add to that, you know, as we think about this competition and this supply chain and manufacturing, this is where having friends is important. We have a lot of allies in Europe, in Asia, and I think doubling down on those alliances, not just from a military standpoint, but also from a manufacturing standpoint, will allow us to meet this challenge and continue to prevail.

ISAACSON: Are we integrating generative A.I. into our weapon systems fast enough, Chris?

KIRCHHOFF: You know, Walter, we are beginning to. There’s a task force called Task Force Lima in the Pentagon that is working very hard to experiment with generative A.I. However, you know, the whole reason why Ash Carter created Defense Innovation Unit is because the U.S. military essentially missed every major wave of technology in the 2010s. It missed the rise of modern software development. It missed cloud computing. It missed the shift to mobile. It also missed the rise of artificial intelligence. When DIU is in operations, the Department of Defense’s computers and cloud computing infrastructure couldn’t even run an A.I. program because they weren’t robust enough. So, I think we all have to watch very closely at whether the department is going to take generative A.I. up aggressively. And the reason why they must, Walter, is that, you know, this is the one advantage our nation has. It’s — you know, a lot of people, China, Iran, Russia can take cheap hardware and build autonomous systems. And in fact, if you look at China’s manufacturing base, it’s many times the size of the U.S. manufacturing base combined with the European manufacturing base and that of our allies. So, what then is going to be our comparative advantage? It’s going to be generative A.I. and integrating that in better and more innovative ways in new weapon systems.

ISAACSON: Raj Shah, Chris Kirchhoff. Thank you all so much for joining us.

SHAH: Thank you.

KIRCHHOFF: Thanks Walter.