Defense Secretary Dr. Mark T. Esper and Michael Kratsios, acting undersecretary of defense for research and engineering, discuss artificial intelligence at the Defense Department’s Artificial Intelligence Symposium and Exposition, September 9, 2020. Their discussion focuses on the DOD’s current activities, future direction and challenges in artificial intelligence. The two-day, virtual event brings defense officials together with representatives from industry, academia and the government to discuss AI operations.
Transcript
Hello, everyone. My name is on multi nanny, and I’m the acting director and CTO of the Joint Artificial Intelligence Center here at the D. O D. And as you know it as the Jake. So I’m really excited to have Michael Cross Seals here today. Who’s joining us for a fireside chat? Although there is no fireplace here, but given how hot it’s been, I know we’re all feeling the heat. Um, one of the funny things was We’ve got one flag here. Somebody before when we started said we put two ferns here, but I guess we could put two flags and call it between two flags, so I didn’t know how well that would go. Uh, so, Michael, what an incredible background you have. You have degree from Princeton. You’re a visiting scholar at Qinghua University. You’re in the tech industry. The venture capital industry, with a great great career and then on to work in the White House, is the chief technology officer for the United States. And then, of course, now named is the acting undersecretary for defense for research and engineering and are also the d o d CTO. So I would almost start with the first question, which is with those many jobs. How do you have time to sleep? But more seriously, you know, it’s part of your new responsibilities. You’ve got a really large part of the deal. These technology development, right? You’ve got DARPA. You’ve got the missile Defense Agency, the space Development Agency. Dia you all of our research labs. So all between all these big areas, what are your top priorities? And how does a I fit into all of this? Well, first off, thank you so much for including me in this amazing event today. This is a big opportunity for the department to showcase so much the incredible What is going on the last the last couple of years for the role around around recent engineering. It’s something that, you know, when when I was at the White House, the D. O. D. Was always such a very critically in part of a critical part of the larger federal family and pushing forward recent development for the American people. And it’s incredible mission that is able t deliver results that not only helpful warfighters but often have incredible spillover effects for the country more broadly and as coming big in this role now for for a little time, I sort of started to shape a bit of what we think The biggest priorities are for the department. With regards to rich engineering. I think the first one and I think one that in some ways can often mean seeing or interpreted as a bit counterintuitive is this idea of leveraging the unique authorities that the Defense Department has to drive innovation for the department and for the country. And the example that I would like to give is incredible effort that the D. O. D. Is doing around five g and I’ve been working on this issue a Zus detail for some time and how we can spur innovation and drive next generation testing unemployment, five g networks And what we have found outside of the duty context is that it’s actually really hard and really challenging to do that, even if you’re a well resourced private firm and you have a a city or locality was excited to a test, the amount of red tape and permanent and everything else has to go through to establish or a bus test can make it very challenging and usually have to scope down your tests or doing a smaller scale. What, We’re able to do it the department with some of our deployments on some of our bases we have. We have a name or five and, uh, winners very soon and seven more to come. And at these installations you can have four more robust tests you could ever have in the private sector is an example of sometimes the D. O. D has gets a bad rap for being archaic and the way goes with processes. But the one thing I’m constantly challenging the team is to think about where the unique authorities that we do have and how can we leverage those two drivers for For the American people, the second issue are airier line of effort that I’m thinking very closely about and helping to drive is, is this this emphasis on leveraging the full resources of the American invasion ecosystem towards our hardest problems? You know, we have we have made progress over the years to kind of expand the scope of which technologists and scientists are part of the family working on these hard deal deep problems. But we could always doom or and the transformation that we’ve seen in the U. S. Innovation ecosystem of the last 66 almost 70 years now has been enormous. Coming out of World War Two, the vast majority of basic orangey was actually funded by the federal government. Now that has flipped entirely. Rather than being 70 30 federal private, it’s it’s flipped on its head and you have 30 70. And that’s not a bad thing. That’s actually amazing for the U. S. Government, and we’d be able to do is leverage the best R and D efforts from from the private sector, bring them and bring them through partnerships into into the into the department and kind of last effort, which I think is is very important, as we think, thematically, towards achieving our national defense strategy, his international partnerships. This is not something that we can do alone and a lot of efforts that we’ll chat about it a little bit relating to AI ethics and things like that. There are things that we cannot do alone, and we must bring our partners and allies in the mix with us on. I think that that’s something that I that I constantly pushed the team teething closely about Well, that’s fantastic. I mean, you’re bringing a perspective both. That’s sort of the, you know, international National and now the D o D levels. So blending those together what? What a great, uh, you know, sort of rolled a having in position. So going something more personal, which is wanted to discuss the irony and Jake relationship. You know, obviously, the Jake was created a couple of years ago. You’ve been in the eye, you know, sort of business and thinking about this even before the Jake was created. How do you see the relationship working out and sort of where? How? How can we optimize that for Teoh? Get the best out of both organizations. I think it’s a It’s a fantastic partnership that’s gonna continue to grow over the years. I think when I think of kind of the core competencies of recently engineering enterprises about that early stage basic research that needs to be done to drive the next great breakthroughs in artificial intelligence than taking those and deploying them and creating tools that are warfighters can use them in the battlefield is what what you guys work on and how we can optimize that hand over and that relationship so that were tight length, as breakthroughs were happening in the labs and across the across the Enterprise and and figure out the best way to move them into into deployment is kind of where I think about a lot. I think a great example is efforts that the DARPA did on deep fakes that have now been kind of pulled into the into the Jake process. And I think we’re also working on adversarial EI together. His idea. How can we, you know, once we’re actually deploying AI tools, how could we harden those and make sure that they’re not susceptible toe adversarial attack? That’s fantastic. Yes, absolutely. I think this idea of our any generating, the big ideas and investing, you know that you know, at scale and the Jake being a feeling organization focused on hopefully taking those and being able to get them field it out there, you know, really? Absolutely. This this given take would be fantastic. So So like, I mentioned, you know, your unique sort of rolled background and experience, you know, at the national level and then the d o d so moving up from the deity to the national level. Would you mind covering some of the bigger ideas around that you’ve been pushing, talking about as the CTO for the United States? Because at a national level and government wide level, I think AI has significance even beyond the d. O. D. So any thoughts on the and sort of initiative there? Absolutely. I think over the last three years we’ve seen a tremendous emphasis across interagency and all the way up to the White House on ensuring American leadership in artificial intelligence. Very early on in in this administration, there was this desire to begin to emphasize AI. And particularly, what are the things that we need to do? Is the federal government to drive leadership in this particular critical domains? Their adversaries are stepping up, and the first major step was a signing of artificial intelligence executive order and that laid out the national strategy for AI, which is, I think, I think, the next day or two days later of the d. O. D. Actually put out their EI strategy. So is a well. It was a well sequenced, sequenced rollout, but I think what was critical about that. That that executive order it laid out the thematic vision or lives for the four main lines of effort that we as a country need to drive to ensure our leadership, and the first is around research and development. It is absolutely critical that we’re making the next great breakthroughs and AI machine learning here in the United States, and that is something that requires a whole of government commitment to to the endeavor. It can’t just be a national science Foundation tinkering with their stuff and DARPA doing their thing and on ih doing their thing the deal deal deal e last old pipes Exactly. These agencies need to come together, and part of part of strategy was releasing an orderly strategic plan. And there’s there’s four core elements around that, and now there is a sort of vision and guiding document for the larger or in the enterprise across the across the department to not only increase funding and support financially tourist redevelopment, but actually coordinate that and make sure that’s done and used used effectively. The second line of effort is always right, no matter what technology working on is around workforce and I think it’s something that you guys are constantly thinking about how you can build Jake Workforce, and we are on our team over here as well. But but broadly speaking, it’s very challenging to build a large scale, robust pipeline of talent. Today I scientists that want to continue to make the next breakthroughs and incentivizing those and figuring out what leavers the federal government has to prioritize grants and fellowships and other types of funding opportunities. So these researchers continue to enter that field. It’s something that that executive order I’m really pushes on. The third is is around removing barriers to AI Nation. I generally tend to see the world of of emerging technologies and sort of two buckets, or either technologies that are born in captivity or technologies that are born free. And phone technologies that are born in captivity are often technologies that is the hardest to commercialize. Those are things like drones or autonomous vehicles or nuclear power. These are examples of technologies. Once you make a breakthrough, you can actually commercialize them as an entrepreneur. Scientists, unless the government says yes, it is OK. I give you see regulation exactly versus technologies are born free, you can say the Internet, for example. It is probably most commonly used example where this technology that people have have built on for decades and we have an extraordinarily vibrant digital economy because because of that so part of the mission of of the the Executive order was to identify the barriers for innovation around at power technologies and allow the US that kind of to be the lead and not allow regulations to kind of hamstring us. And one of the major help puts of that is something I think we’ll talk about a little bit, which is our AI regulatory principles and actually being able to sort of provide sort of confidence to the community and consistency, consistent sort of guidance to the industry, over kind of what the government is looking at when it comes to a I rags and, well, first in first in the world to do that. And in the last piece, which I think we mentioned a little bit early on as an Arnie priority is around international engagement. We we recognize that you know, our adversaries are pushing ahead with a very different view on on the way I should generally be used, and it’s critical for the US to join forces with with our allies to make sure that the West is the home for the next great technological discoveries. And one of bigness ships we did was go to the O. C. D. And lead a development of the O. C. Ai principles for the first ever ai principles that were agreed to by by governments Mrs Effort that that the U. S. Led there and served as a foundation for sort of the Western liberal democracies as they as they pursue a i A. Regulation. So long story. The main, the main strategy and theme, really centers around recent development workforce regulations in international engagement. That’s fascinating. I mean, one thing that I’m getting from from just this discussion is you’ve had a really, really focus on, you know, focusing on what government does great, you know, in terms of what are the things that government should do, things that private industries should do, having a framework and having these applying the best of government to enabling and opening it up for private industry. That’s a fantastic theme here and along those lines, I mean one thing that I think the symposium attendees would like to know is how are you putting together the sort of budgets you know, when it comes to both the d o. D. But also actually government wide. I mean, being part of the White House, having this pan government vision around what investments we need to make, I would like to hear kind of how you’re thinking through that in the scale of what we’re doing here. Yeah, with the the year CTO had, I think the first we were thinking about was, how can you How can you shift the larger federal government? Uh, I guess massive, sort of aircraft carrier real renting to a new to a new orangey priority. And it could be extraordinarily challenging. And I think that the best way to do it is to kind of use higher level administrative goal, sort of administration goals and guidance that can trickle down to all the agencies in the 2021 budget that the president set the set. The goal to double AI spending over the next two years. And this was a A really, I believe, a tremendous step forward for four of the country actually setting a goal in saying over two years, we can. We believe that our or an ecosystem can actually absorb those dollars and observing the way which will continue to drive high quality research across of across the ages? That’s kind of the first thing you can. You could turn the dial on the dollar exactly. They’re being wasted. It’s not. It’s not really good. The first assessment was done. Is is how many sort? Essentially high quality proposals are out there and unfunded and National Science Foundation has sort of done very interesting work on this because they have a very rigorous process and evaluating their grants, and they could definitely absorb this type of thing. This type of plus up and so that was announced in February of this year and the first full cross cut of the 21 budget actually came out this month, and what was very interesting about that was we are very much on our way to achieving that, that doubling no, we’re up over over 50% year over year and are gonna hit the doubling target as a asshole of government next next year. So we moved from a little shy of a billion dollars in unclad in non defense AI spending toe a little over a billion and a half. Also you can see is that in some ways I think that the the D. O. D. Was actually leading the charge before the White House guys act together to to realize that we need to push the push the ball here in 2018. DARPA’s many of you know, launch day I next. And that was a $2 billion commitment towards towards AI that Jake has obviously seems to have tremendous support, both from from the Hill and the department, wide budgets on ramping up in a in a way that can can allow it to build the teams and start start deploying its efforts. So I think a few years ago you guys, I think we’re under $100 million. Right now you’re you’re almost 300 so that there’s great progression there in showing that this commitment to dollars. But I think it’s not just the money alone and I want to stress that I think the dollars air just kind of one piece of the larger puzzle, and what’s critical now is is with this this incredible investment, a commitment by the federal government towards AI leadership. We need to make sure that those dollar skeletons spent in a very smart, ineffective, ineffective way. Yeah, I know. That’s that’s That’s amazing point. I mean, like you mentioned the dial’s right, this idea of the technology commoditization curb, right. We’ve seen this over and over again for dozens and dozen years. The idea of, you know, scaling the right pieces of AI with money, which you have to lean into and then bending the technology commoditization curve with investments. On the research side, it’s a it’s a little bit of an art, but, you know, that’s sort of the whole game here. Which is how do we actually investors correctly, as you’re saying with that emphasis, but focusing on the right side of the curves? No, absolutely. So it’s not an easy job. That’s why smart people like you know, um so, uh, one topic on everyone’s mind at the symposium and AI in general is around ethics and policy and regulation and all the other pieces, and you made a fantastic point, right? This idea of applying and using the government’s role the right way to create this. As he said, The captivity versus free. How do you open things up? So you know, commercial industry can flourish. Research can happen things like that, but within the bounds of what we as a society, want to see and the principles and values that way, like and cherish. So what? What thoughts you have on that and the work that you’ve done? What you’ve been very active in this. So I think the the audience would love to hear that. Well, I think first off, it’s sort of important to think of what the global landscape looks like. It’s it’s very clear we have an adversary in the CCP that is twisting this technology in a way that is in violently, uh, unaligned, with the way the United States and the West views technological innovation. AI is being used by the CCP to track ethic minorities, Teoh, suppress freedom of speech, do all sorts of things that way, definitely do not want artificial intelligence to be used for. So in that context, I think it becomes more imperative than ever to ensure that the U. S. Leads with a set of principles around AI, which are rooted in our American values. And the first step from a sort of government Weinstein standpoint, was to provide regulatory certainty around what that means. This is a This is an emphasis on on sort of, you know, trustworthy, safe, robust ai that that’s rooted in American values and ethics and what what I think was very important around this document and in the executive order, the president asked. The White House essentially produced a regulatory guidance stockman to the agency. So any agency that was gonna be promulgating any regs that affected AI technologies in the private sector what are the one of the considerations they should take and making that rag? So first and foremost, what we we argued, I think, is very much a line with kind of what happened here at the D. O. D. Is is that is the critical importance and understanding of sector specific approaches. And I think very much the conversation, fortunately moves to the super high level you know, 100,000 ft views on on what should be the right pathway towards towards AI ethics or a regulatory posture more broadly. But you quickly realize that the questions that the Department of Transportation and Nitze face when they’re attempting to create a regulation for a new car of its not gonna have a steering wheel or brake pedal. Very different in the types of considerations that the FAA is taking the play when you have a drone doing commercial operations or when you have the FDA trying to approve an air power medical diagnostic lower for the case of D. O D. How you actually implement AI into a kill chain or anything else that will be working on here. So for us and what was very critical that the White House laid out was it has to be sector specific and and that that guy has got put out in earlier this year is out for comment and hopefully be able to have a final final version before the end the end of the calendar year. And I think from that you will see individual agencies that are dealing with regulations start thinking very carefully around how they can taken to Berrigan toward posture to align with sort of the ethical in American values that are that are laid out in that memo but are specific to their sectors. That’s fantastic. Yeah, this idea of a single single regulation, you know, across all of these domains and also a itself is not a single market. It’s a collection of technologies applying different things to different things. So that approach sounds amazing. I know we’re coming to sort of the, you know, our time here, but, you know, we talk T o d. We talked national level and, you know, you obviously have had ah, you know, a role in sort of the international stage as well on the technology front. So anything to talk about sort of ai you touched upon our adversaries allies just going a little deeper into that. Any thoughts or comments but policy, but, you know, partnerships, a whole bunch of other things that I think it’s it’s actually critically important in the world of artificial intelligence. For a last couple of years, United States has been very thoughtful about the way that it approaches international engagement, and it’s something that we, as a country need to be very sure of its in the best sense of the American people. Before we sort of pursue international initiatives. And I think that was spelled out and asked defense strategy that it’s critically important. We actually do that are thoughtful about it. So I think the efforts that the United States made at the O. C. D. To say we’re willing to with arms and arms with our sort of developed economies around the world, we will lay down and agree to the first international AI ai principles was a huge step forward. And it shows how critically important engagement with our allies are on this issue because of the things we talked about of the wear adversaries are thinking about it and their adversaries try to influence the way that that AI is sort of impacting citizens around the entire world. We need to provide at the right positive, Western based alternative to that. And I think we can We can do that. I think we’re seeing it to the eye principles. And I think that is something that that sort of the the D. O. D. Is gonna continue to think about as well with the very critical. Um, it’s sort of a I ethics principles that we put out as a as a department, you know, this department has sort of led the world for for decades on the way that sort of the right way to conduct and engage sort of, ah, military department. And it sort of set an example for the rest of the world. And I think it was a an incredible step forward for us to kind of be the first out of the gate as a anywhere in the world to kind of put these out. I think it will serve as an example for many of our allies as they grapple through the very hard questions. So thought leadership at an international stage that then drives everybody, you know, to sort of in the same direction. Absolutely. What a great what? Almost great weight. And, um, I thought we’d end actually with just one small thing. Kind of what I call the crystal ball moment, right? You are the chief technology officer. You are the primary technological leader for the D. O. D. Driving the sport. You’ve been involved. Obviously. Ai quantum computing, five g. All these great technologies. How do you see these things coming together? I mean, it’s obviously are really, really hard question, but I thought, you know, just some ideas and how you how you’re approaching the fusion of these things coming together and what those capabilities might mean for us here. Yeah, I think that the most critical thing that everyone working on these issues across department needs to constantly keep in mind as we’re not benefiting the department or the American people by thinking about these technologies in stovepipes that has to be integration among them. And as a department, we have 11 modernization priorities you mentioned. Three of them is probably some are most important ones. But there is no way that the future deployment of these technologies for uses buyer warfighters, is going to benefit by people thinking about those technologies in isolation. And I think that’s something that I constantly challenged my team. And if you zoom up a level and think about sort of a larger national national picture, the P cast Presidential Council advisor of science and technology that advises the president in the White House on on science matters, they put out a report a couple months ago around around kind of this, this exact question and they truly believe that the greatest benefit could be replied. The American people, if you are able to do the R and D associated with some of these technologies in sort of co located spaces or in places. We already are thinking about the integration, and it’s great to have, you know, individual research programs or institutes on singular technologies. But you can get a lot more benefits if you can start working, working things together. So that’s something I know I’ll be working on for a long time. And I think department certainly will be focused on. Wow, this has been great. Well, thank you so much for your time. I see best of luck with all your jobs, various things. But having your leadership and vision here is gonna be really, really welcoming and exciting. Thank you. That means a lot. This has been incredible conversation. Thank you so much, Michael. Thank you. Hello. I’m Secretary of Defense Mark Esper, and it’s a privilege to join you today as part of the Defense Department’s AI symposium. I want to thank Man Mirchandani and his team at the Joint Artificial Intelligence Center, or Jake for organizing this important event and for bringing together the best and brightest minds in the department who are working to accelerate the adoption of critical emerging technologies. Today, I’d like to speak you about why the United States must continue to lead the way in the responsible and ethical development of artificial intelligence and, more specifically, why the department defense must remain at the forefront of the military application of this game. Changing technology history informs us that those who are first to harness once in a generation technologies often have a decisive advantage on the battlefield for years to come. I experienced this firsthand during Operation Desert Storm with the United States. Military smart bombs, stealth aircraft and satellite enabled GPS helped decimate Iraqi forces and their Soviet equipment. In today’s era of great power competition. As new technologies alter the character of warfare, we must stay ahead of our near pier rivals, namely China and Russia. Our national defense strategy guides us in doing just that as we modernize the force and protect our competitive advantage across all domains air, land, sea space and cyberspace. Thanks to our largest research and development budget in the department’s history, we are accelerating the development of critical technologies that will enable us to outpace our strategic competitors and maintain our military overmatch. But unlike advanced munitions or next generation platforms, artificial intelligence is in a league of its own, with the potential to transform nearly every aspect of the battlefield from the back office to the front lines. That is why we cannot afford to see the high ground to revisionist powers intent on bending, breaking or reshaping international rules and norms in their favor to the collective detriment of others. In 2017 Russian President Vladimir Putin declared that whichever nation leads and a I will be the ruler of the world. He ordered his government to develop a national AI strategy and called for increased investments. His intent is to employ any possible advantage to expand Russia’s influence and chip away at the sovereignty of others. We saw this with Russia’s incursion into Ukraine, when Russian forces used a sophisticated and well coordinated combination of unmanned aerial vehicles, cyber attacks and artillery barrages to inflict severe damage on Ukrainian forces. Since then, Moscow has announced the development of AI enabled autonomous systems across ground vehicles, aircraft, nuclear submarines and command control. We expect them to deploy these capabilities in future combat zones. Likewise, the Chinese Communist Party recognizes the transformational power of AI and openly declares its intentions to be the world leader by 2030. The People’s Liberation Army regards AI as a leap frog technology, which could enable low cost, long range autonomous vehicles and systems to counter America’s conventional power projection. At this moment, Chinese weapons manufacturers are selling autonomous drones they claim can conduct lethal targeted strikes. Meanwhile, the Chinese government is advancing the development next generation stealth waves, which they’re preparing to export internationally. If there is any doubt as to how the CCP would will AI to influence its agenda abroad, look no further than how it uses this capability. At home, Beijing is constructing a 21st century surveillance state designed to weld unprecedented control over its own people. With hundreds of millions of cameras strategically located across the country and billions of data points generated by the Chinese Internet of things, the CCP will soon be able to identify almost anyone entering a public space and censor dissent in real time. More troubling is the fact that these systems can be used to invade private lives, leaving no text message, Internet search, purchase or personal activity free from Beijing’s tightening grip As we speak, the PRC is deploying and honing its AI surveillance apparatus to support the targeted repression of its Muslim Uighur population. Likewise, pro democracy protesters in Hong Kong are being identified, seized, imprisoned or worse by the ccps digital police state, unencumbered by privacy laws or ethical governing principles. As China scales this technology, we fully expected to sell these capabilities abroad, enabling other autocratic governments to move toward a new era of digital authoritarianism. The contrast between American leadership on a I and that of Beijing and Moscow couldn’t be clearer. We are pioneering a vision for emerging technology that protects the U. S. Constitution and the secret rights of all Americans abroad. We seek to promote the adoption of a I in a manner consistent with the values we share with our allies and partners. Individual liberty, democracy, human rights and respect for the rule of law just to name a few. Indeed, we approach a I as we have other high tech breakthroughs throughout our department’s history, with rigor, standards for testing and fielding capabilities and the highest ethical expectations. Technology may constantly change, but our commitment to our core values does not. In February we became the first military in the world to adopt ethical principles for the use of AI based on core values such as transparency, reliability and governability. These principles make clear to the American people and the world that the United States will once again lead the way in the responsible development and application of emerging technologies, reinforcing our role as the global security partner of choice. To build on this foundation, we’re designing a comprehensive strategy to train and educate all deal D personnel from AI developers to end users so that everyone understands their role in accelerating the department’s digital modernization strategy. Over the last six months, for example, the department has stood up a responsible AI committee that brings together leaders from across the enterprise toe foster a culture of AI ethics within their organizations. In addition, the Jake has launched the Responsible AI Champions Program, a nine week training course for D. O D. Personnel directly involved in the AI delivery pipeline. We plan to scale this program to all D. O. D components over the coming year. Further, the Jake, in partnership with the Defense Acquisition University and the Naval Postgraduate School, will launch an intensive six week pilot course next month to train over 80 defense acquisition professionals of all ranks and grades. The’s trainees will learn how to apply a I and data science skills to our operations. With the support of Congress, the department plans to request additional funding for the services to grow this effort over time and deliver an AI ready workforce to the American people. As a Defense Department builds a digital security architecture to promote responsible and ethical A I development, we’re also shaping new international norms for the use of this technology. Next week, the JAKE will launch the first ever AI Partnership for Defense to engage military and defense organizations from more than 10 nations with a focus on incorporating ethical principles into the AI delivery pipeline. Over the coming year, we expect to expand this initiative to include even more countries as we create new frameworks and tools for data sharing, cooperative development and strengthened interoperability. We have come a long way since establishing the Jake two years ago. Today, more than 200 talented civil service and military professionals work diligently to accelerate AI solutions and deliver these capabilities to the warfighter from helping the joint force organized fight and win at machine speed to enhancing wildfire flood responses through computer vision technology. The Jake is utilizing every aspect of AI as a transformative his permit at home and abroad. The JAKE is also lowering technical barriers. Today I adoption by building a cloud based platform to allow D. O D components to test validate and filled capabilities with greater speed at greater scale. The goal is to make a I tools and data accessible across the force, which will help synchronize projects and reduce redundancy among many other benefits has an example of the tectonic impact of machine learning on the future of war fighting. The Defense Advanced Research Projects Agency. DARPA recently conducted its third and Final S 16 combat simulation between an AI controlled system and then experienced Air Force fighter pilot. The agents resounding victory demonstrated the ability of advanced algorithms to outperform humans in virtual dogfights. The’s simulations will culminate in a rural world competition involving full scale tactical aircraft in 2024 to be clear. Hey, eyes roll in. Our legality is to support human decision makers, not replace them. We see a I as a tool to free up resource is time and manpower so our people can focus on higher priority tasks and arrive at the decision point, whether in a lab or on the battlefield, faster and more precise than the competition. For instance, AI is advancing automated chemistry for a host of national security applications, including the continuous manufacturing of critical pharmaceuticals and the development of novel propellants for spacecraft engines. These advances free up time for our scientists and researchers to focus on next generation innovation rather than countless tests and experiments. Furthermore, through DARPA’s Spectrum collaboration Challenge, we are demonstrating the power of machine learning to improve spectrum usage through autonomous collaboration, which has enormous implications for future military and commercial networking, especially with five G. Lastly, to fully realize the potential of AI across the Defense Department, we’re committed to reforming outdated policies, divesting from legacy systems, modernizing our acquisition practices and strengthening our partnerships across industry and academia. Decades ago, the federal government supplied the bulk of research and development funding for many of the department’s most important hard work capabilities. Today, the private sector leads the way in software. We need to be fast followers of our industry partners and work hand in hand with them to develop a I at the speed and scale needed to stay ahead of the competition. During the department’s Corona virus response, for example, the Jake worked alongside industry to develop a Predictive Analytics tool that helped US Northern Command manage supply chains and identify hot spots with the greatest need. And in academia. The Army AI task force at Carnegie Mellon University and the Air Force AI accelerator sit at the crossroads of higher education, industry and defence rapidly prototyping, cutting edge innovation. Ultimately, these kinds of solutions are vital not only for our department defense missions but also for the good of the American people. With each generation come new technologies that fundamentally alter the way we think about plan and prepare for war. AI is one such breakthrough with balance potential for good or for ill. However, unlike our competitors, we commit to using these capabilities to preserve peace, promote prosperity, respect rights and maintain the free and open order that benefits all nations. As the epicenter of groundbreaking innovation, the United States will once again lead the way and developing and governing emerging technologies. We will stay true to our core values and principles. We were remained the global security partner of choice. And together with our allies and partners, we will defend the international rules and norms that have secure our rights and our homeland for generations. Thank you for your time and thank you for all that you do to ensure the department defense remains the standard of excellence at home and abroad.