 Hello ladies and gentlemen, take your seats. Candice Rondo, General, can everybody take their seats? Daniel Rothenberg. Well, thank you everybody for coming. Stu Brayden, we're starting, sir. Thank you everybody for coming. I'm Peter Bergen, I run the International Security Program here and I'm a professor of practice at Arizona State and I'm the chairman of the Global Special Operations Foundation. So I have three of my bosses here and one in the room, which is great. It's a rare opportunity for me. Well, so yesterday, of course, was the, we celebrated Veterans Day and of course on Sunday it was the 100th anniversary of the end of World War I and much of what we're going to talk about today in this sense is really about the collapse of the Ottoman Empire, the long kind of tale of that, the rise of the British and the French Empire in the Middle East and then of course the American Empire, which inserted itself in the Middle East after 9-11, creating vacuums in respect of Iraq and Libya, which produced many of the preconditions necessary for the proxy warfare that we'll be discussing. So this is sort of an appropriate date to have this discussion, this meeting. I'm going to introduce my boss, Emery Slaughter, who is the CEO and President of New America. She, five years ago, a little over five years ago, said, came to me and said, why don't we set up this thing called, why don't we look at the future of warfare as a program and from that conversation has come a future warfare program here at New America and also at Arizona State. We have about a dozen full-time fellows. We have multiple faculty at Arizona State. We teach an undergraduate course. We have a hundred people taking an online graduate course in global security as a result of that conversation, an idea that flowed from Emery. So thank you, Emery, for that. And secondarily, of course, everybody knows Emery as the CEO and President of this organization, but she was also the first woman to have George Kennan's job at the State Department as running policy planning. She also ran the Woodrow Wilson schools, taught at Harvard, Britain, or edited eight books. And so it's great pleasure to have Anand Overtraher, Emery. I really think in this crowd, my greatest credential is that I'm Peter Bergen's boss. And he's one of the few people when I'm very proud of the number of books I've written and edited about half and half, but with Peter, that pales in comparison to his regular bestsellers. So I'm delighted to have you all here and particularly for this day. The Special Operations this is a first. We hope it won't be a last. It's an absolutely ideal moment to be thinking about our special operations forces and specifically to think about proxy war and I'll come back to that in just a minute. I want to start by saying that we're delighted to partner with the SOF Foundation. Also the McCain Institute of ASU, as Peter said, we have a complex in the good sense, meaning many different parts, multifaceted partnership with ASU, which is the country's most innovative university for at least the last five to seven years and I say that sincerely in that I really think ASU is reinventing American higher education. I say that as no stranger to many institutions of American higher education, but when you look at what ASU is doing in terms of access, but insisting that you can have universal access and high quality research and innovation in terms of creating new subjects, new degrees combining what people want to learn. I don't know any university that is more innovative. We're also, for I think the first time, partnering with the Daniel Morgan School of National Security, which is lovely and so it's a powerful group. I also just want to say about the SOF Foundation. I'm a network theorist and the description of the foundation's work, its mission as Stu Braden describes, is to forge the good guy network, bringing together military government industry and intellectual leaders from around the world to advance the capability and the efficacy of our special operations forces and I love that mission. I think in many ways as we go forward solving public problems in every area and this is core national security government, but even here we need to bring together academia and policy, non-governmental actors as well, also industry. So it's a great partnership. Peter mentioned some of our faculty. I just want to list a couple of others, make sure everyone gets a shout out. Again, through the Center on the Future of War and the Project on the Future of War, a title we love until we're trying to advertise our conference in the spring and then lots of people go, why isn't it the future of peace? And we say it's actually one in the same in many ways. But we have Lieutenant General Ben Freakley, who's the professor of practice and of leadership. We have Nick Rasmussen, whom you're about to hear from, who's the senior director of the McCain Institute's counter-terrorism program and also the former director of the National Counterterrorism Center. We have Dan Rothenberg, who is the co-director of the Center on the Future of War and a professor of practice at the School of Politics and Global Studies. ASU titles are long because they have lots of different schools and also Candace Rondo, who's a professor of practice at the School of Politics and Global Studies. We're going to be hearing from lots of different people today in our three panels and particularly around people who are designing and implementing the technology that directly impacts our special operations forces. And I want to particularly call out Eric Grant from Raytheon. And military representatives, we have Air Flotilla General Vasile Tuider from Romania, who is again putting policy into practice. So welcome. And we wouldn't be having this forum without Raytheon's sponsorship, so we're particularly grateful. And I hope you will take the chance to thank the Raytheon representatives who are here today. So now to just come back to the subject, proxy warfare. We've got three different panels on looking at the overall question of proxy forces in modern warfare, AI in the gray space. And I will just say New America had its board meeting in California last week and we had a salon seminar on AI with the head of AI at the partnership for AI, from open AI, from Apple, and from the Y incubator. And it was quite the conversation. And then we'll have a panel on Iran's proxy warfare strategy. I just want to conclude by reflecting for a moment on proxy warfare now. Many of us, I can look around the room and say that and I have to say the more I look around the room the straighter I am standing. For all of you, it's great to have you here. Those of us who grew up in the Cold War are no stranger to proxy warfare. That was warfare in the nuclear age. We couldn't fight directly with the Soviet Union, so we fought proxy wars all around the world. And when I came of age as an international relations expert and national security expert, you were looking at proxy wars in Asia, in Africa, in parts of Latin America. That's still happening. I mean, we're involved, Russia's involved, China in places, but it is much more complicated precisely because of the rays of the rise of global actors. I don't actually like non-state. It is like calling an automobile a horseless carriage. You're looking backwards. Instead, I think we should be thinking about governments, state actors, and global actors. Those actors who can make a difference in the global space, but who are not government. So you talk about state actors and global actors. You have terrorist groups. You have criminal networks of various kinds. You have industry. If you look at global crime, it's often through industries that are legitimate on the top half and illicit on the bottom half. And of course, you have lots of non-governmental organizations. You can't quite tell what they are. All of that makes the proxy war space much more complicated. And all we have to do is look at Syria to see that. Indeed, it shifts the horizons of strategic surprise because there are so many different actors from which that surprise can come. It erases traditional front lines. It reshapes alliances and they don't last very long. And it transforms rivalry and rivalries. Much of what we'll be doing today is looking at how proxy war and proxy forces transform the rules of both engagement and strategic risk. So New America, in partnership with ASU with the Center on the Future of War, has started a multi-year research project on 21st Century proxy warfare. What we're discussing today and the paper presented is the first of a series of papers on conflicts in the greater Middle East and its periphery. And a large part of what we're doing, we're filling a research gap. We often do that and we think of ourselves at New America as being ahead of the game with big ideas, but very specialized expertise. And the summary page of our first paper in this series is available at registration for you to see and also on the website, and then you can sign up for getting the full paper. So with that, it promises to be a full and interesting and probably mind-bending day. My job is to introduce Colonel Dennis Willie, who is our first U.S. Army Fellow at New America, and he will then in turn introduce our first panel. Colonel Willie has been an active duty member of the Army for more than 23 years. He's participated in deployments in Bosnia, Kuwait, Iraq, and Afghanistan. And indeed, his year-long tour in Afghanistan took place at the ISAF Joint Command, and he was a military planner responsible for or focusing on integrating non-military applications into NATO's overall approach, which is an angle from which I think gives him a lot of expertise on what we're discussing. So I couldn't be happier to be here and to listen, not just as head of New America, but in terms of my own interests. And I welcome you and Colonel Willie. To ask the panel members to take their seats. Good morning, ladies and gentlemen. As Henry introduced me, I'm Dennis Willie, active duty Colonel here at New America, and it's my distinct pleasure to introduce the first panel, which will focus on the use of proxy forces in modern warfare. And as a military professional, I'm constantly on the lookout for insights into the evolution of the character of warfare and how our armed forces can remain prepared to succeed in the face of those challenges. Now, Peter, you reminded me in your comments about Veterans Day. I spent some of my Veterans Day weekend being a tourist here in D.C. and going to the American History Museum in Mount Vernon. And I was reminded that even at the beginning of the creation of our Republic, proxy forces were involved as Washington attacked Hessians at the Battle of Trent across the Delaware River. So things are not necessarily new, just different. A lot of questions come to my mind, but I'll save my own questions for later. But what makes the current version of proxy warfare new? What capabilities and techniques should the U.S. pursue to impact proxy warfare? And then what historical analogies can we use to inform our understanding of this modern change? To our panelists, Kelly Magsman, pronounce that correctly, is the Vice President for National Security and International Policy at Center for American Progress, where she focuses on Asia Pacific and Southwest Asia, foreign and defense policy. Prior to her current role, Kelly worked in government service at the Department of State, National Security Council, and the Pentagon for over 10 years, focusing on some of the world's thorniest challenges. Her last position in government service was as the Department of Defense's Principal Deputy Assistant Secretary of Defense for Asian and Pacific Security Affairs. I think you've got ASU. You can work at ASU now, all right? Kelly, thank you for joining us today. Next, I welcome General Vasili Tuada, who currently serves as the Deputy Chief of Staff for Operations and Training in the Romanian Ministry of National Defense. John Tuada joined the Romanian military more than 30 years ago and has distinguished himself through his multiple assignments to NATO missions and organizations across Europe and Afghanistan. I think we may have crossed paths briefly in North Kaia back in 2012 and 2013, as you serve there with the air detachment. Thank you for your service, and we look forward to your comments today. And our third panel member is Nicholas Rasmussen, currently the Senior Director of the McCain Institute's Counterterrorism Program. Nick is an accomplished government servant having begun his career at the Department of State in the 1990s. And in beginning in 2001, Nick distinguished himself within the intelligence community as a counterterrorism expert and served for President Bush, Obama, and Trump in the National Security Council and the National Counterterrorism Center. Most notably, he led our nation's counterterrorism efforts as the Director of the NCTC from December of 2014 to the end of 2017. Today, the very capable Peter Bergen will moderate our panel. I don't think I'm going to spend too much time introducing you, but to give as much time as possible to this panel, Peter, over to you. Nick, we'll start with you and then we'll do the general and then Kelly. Thank you. When you see a conference on proxy warfare, you tend to focus on, okay, you're probably going to focus on the examples, the exemplars where proxy warfare is somehow a threat to our security interests. And certainly, panels organized around Iran later in the day, and certainly some of the material Kelly will bring from her experience in government with the threat posed by Iranian proxy forces, you tend to focus on, okay, where is this something that is harming our national security interests? Where is this something where this evolution is actually putting us in a more dangerous place? I want to flip the coin for at least a moment or two, and maybe it would have been better if we laid out the scary side first. But there are ways in which proxy military forces actually are utilized today as a tool for us to secure our national security interests, if you're speaking from an American perspective. I can give two concrete examples from the waning days of the Obama administration. Two areas in which we found ourselves looking for ways to secure American interests but without necessarily having the tool ourselves or being willing to use the tool ourselves to actually employ our forces on the ground in a direct and engaged way. The first of these is, of course, obviously in the counter-ISIL campaign in Iraq and Syria. As much as you can look at differences between Trump administration and Obama administration, we are more or less employing the strategy today that we were employing in the waning years of the Obama administration. And that was to work with Iraqi security forces where possible on the Iraqi side of the border and to enable them to play the leading combat role in displacing ISIS from its territorial strongholds. And that strategy on the Syrian side of the border, and I say border in quotes because, of course, we know that wars are often fought in territory where borders aren't recognized. But the Syrian side of the Iraq-Syria war, that proxy force was a TBD when that strategy was first articulated by the Obama administration. We were going to work with some local force to displace ISIS from its territorial strongholds in Syria, and we were going to wait and see what that force looked like. We had an idea of what we wanted it to be. We were hopeful that it would have a largely Syrian Arab character. But the point was we had made a strategic decision as a country that we weren't going to deploy U.S. forces in large numbers to carry out ground combat operations against ISIS forces, but we were willing to enable a partner. We just didn't know who that partner was. Of course, over time that partner did emerge, that partner under the rubric SDF, Syrian Democratic Forces. Of course, we know that that force had a largely Syrian Kurd character to it, and so we had articulated a strategy that had a prerequisite that we had not accomplished before we articulated the strategy. And when we filled in the blank with that strategy and advanced our relationship with the Syrian Democratic Force, of course it came with considerable cost. It helped us achieve our objectives on the ground in Iraq and Syria, or certainly in Syria in terms of displacing ISIS personnel, leadership and forces from Raqqa. But it came at cost because of course our support for the SDF involved bringing our relationship with our Turkish NATO partner almost to the point of rupture. And of course there are many stresses and strains on our partnership with Turkey that run the gamut across political, military and other issues. But probably no issue put more stress and more strain on our bilateral relationship with Turkey than our support, American support for a group that the Turks viewed as more or less being a foreign terrorist organization. So again, my point is this is an instance where we created and developed a strategy around the idea of finding, identifying and supporting a proxy warfare, a proxy war partner. And I would argue that we've done it successfully. If you did, as the Obama administration did and as the Trump administration did, prioritize the effort against ISIL as your preeminent national security objective at the moment you are making that decision and articulating that strategy. But it was done with Eyes Wide Open. It was done with Eyes Wide Open that we were certainly going to be bearing some cost in terms of our future relationship with the Turks. And also we were going to bear at some point some cost in our relationship even with SDF partners because there might come a point in the future when that partnership might be shut off or curtailed or otherwise affected by our relationship with the Turks. We're still in the middle of that drama right now. That conversation is an ongoing conversation watching in policy circles. I'm happy to be outside of that conversation right now because it is filled with suboptimal choices, ruin a defense relationship with a NATO partner on whom we rely, or walk away from a battlefield partner who has done tremendous service, brave service, who have a Syrian Democratic force that has put blood and treasure on the line to defeat our enemies on the ground in Syria. So that's the policy dilemma that sits in front of the Trump administration today. And I think it's an important thing to kind of talk about today as we think about proxy warfare. The other somewhat less highlighted instance in recent years in which proxy warfare I think has been used to our advantage, certainly in the counterterrorism world that I came from is in Libya. Most of you will remember that the period 2014, 15, 16, 17 in Libya was a period of profound turmoil, profound civil war, lots of militias, lots of discussion about who we could partner with, who we could ally with politically in order to create a more stable Libyan political entity. And at the same time as this ongoing civil war was playing out, ISIS was beginning to prioritize North Africa and Libya in particular as a theater where it looked to put its most dedicated effort in areas beyond Syria and Iraq. When we were thinking about areas where new provinces of the so-called caliphate would emerge, Libya was very much near the top of the list. And given that North Africans had played such a prominent role in the conflict in Iraq and Syria in the first place, it was no surprise that North Africa would be a place where ISIS would seek to expand. And here we were facing this mounting challenge, again, in the waning days of the Obama administration, facing this mounting challenge from ISIS in North Africa, and not having a capable, competent state level partner in Libya, a military or even really an intelligence service with whom you could partner. And so after much consternation, and I would argue ringing of hands because there simply weren't good policy options, at the point when ISIS reached the peak of its power in Libya, having claimed, having taken territorial control of a large Libyan city on the Mediterranean coast, cert, it finally came to the point where the US made a decision to partner with local militias for the express purpose of ejecting ISIS forces from that metropolitan stronghold of cert. That was a somewhat more narrow objective than what we pursued in Libya, I'm sorry, in Iraq and Syria with a proxy force, but make no mistake. It was US air power in support of a local militia force on the ground looking to eject ISIS connected forces, ISIS link forces from a territorial stronghold. Again, there we were at risk of taking sides in a civil war. We were at risk of enabling actors with whom we did not necessarily have full confidence in terms of how they would carry out military operations, would they, so these are not risk-free policy choices, even if they end up being the only available policy choices. In some cases we were bearing some degree of risk. Would that partner live by the law of armed conflict? Would that partner avoid civilian casualties where possible? Would that partner treat captives or detainees in accordance with the law of armed conflict and with our own standards for doing that? All of these are open policy questions when you're making decisions about whether to employ a proxy force. So I'll stop there, but I think you can point to those two examples as instances in which proxy warfare was actually a tool that we managed to turn to our advantage in a circumstance where we probably lacked any other good option. Thank you, Nick. General. Thank you, Peter, for the opportunity you offer me to speak here. Thank you very much, Global Soft Foundation, for inviting me and speaking in this important forum. I would like, in my speech, to provide you an overview on the Black Sea region. This is my main objective today, and this is right the way I can see the Black Sea region and the way I would really pass this information to you to see and understand the Black Sea region. After the fall of the Iron Curtain, we had some young new democracies in the South Eastern part of Europe, and each of the countries was practically exposed to some of the proxy wars in that area. So it was not long to wait until Transnistria was showing up in Moldova, and Nagorno-Karabakh was coming in Armenian Azerbaijan, and also Georgia in 2008 with Abkhazia and South Ossetia, and right in 2014 with Crimea being taken over by the Russians, and also with the conflict in the Donbass region. There is a belt of frozen conflicts in the region that is creating a specific security environment, not a good feeling of security in the area, mainly because of the illegal annexation of Crimea. Donbass region, we could have added to the frozen belt of conflict, however it's not frozen yet, it's still active and there are also people fighting there and dying in the Donbass region. From the Romanian perspective, we could see that there are some efforts of Russian influence in the area. They are trying to regain the spur of influence they had before 90s during the Soviet Union era, so just trying now to be considered as superpower, the regional superpower, and sitting at the same table with the United States. Is there business what they are thinking about? The only thing is that there is a lot of influence in the region. And when we are speaking about the proxy wars, we are thinking about some other new mechanism of proxies. That means the amount of energy, effort and investment they are doing in cyber warfare, which is critically important at this moment. I'm not making any other theories or comments right now, you know very well cyber, there is no crossing points, no border, no window of opportunity is practically open for everybody, and it depends how clever they are to wipe out their traces. They can do practically everything they want, trying to perpetrate any kind of illegal activities in this area. There is also some economic investment, investment from the Russian side, specifically to my country. There were private companies who bought some shares from our private companies, young private companies, who finished practically in bankruptcy. So what we have seen from this perspective, there was not progress at all, no progress at all. We lost practically a big part of our heavy industry and we also have some other portion of the economy that is under Russian control, like a luke oil company, who's investing quite a lot in Romania, and Gazprom, who's providing a small chunk of natural gas to our country, but through a German company. So the big advantage for Romania is the fact that we are not heavily dependent on the Russian resources. Practically on natural gas, we are only 20% dependent on the Russian natural gas. So from these points of view, or point of view, are really okay at this moment, but thinking forward to the security and stability of the region, which is not really well at this moment because with the illegal annexation of Crimea, the Russian Federation deployed there a huge quantity of armament, ammunition, weapon systems. So we sometimes joke that Crimea will probably sink in the near future because of the quantity of so many weapons and weapon systems, but with the deployment of the system, they created, probably they inspired them from the Chinese history when they built the Chinese wall. Now the Russians, they built another wall. They built the A2AD wall. Everybody knows that the anti-access area denial wall just tried to preserve themselves and protect themselves from, I don't know exactly what for, because everything is invested from the West. It is negatively accepted or negatively commented by the Russian Federation, and every step forward which is done by the West in order to create good conditions of living high level of living standards into the southeastern states. The Russian Federation are trying to destabilize everything, to try to undermine every kind of trust we can build towards the West. The situation at this moment is still okay as soon as we are having a good investment into the force, good investment in the economy, good investment into the industry mainly, and speaking from the security point of view none of the country and Europe. I'm not speaking only about the southeastern part of Europe or the eastern part of Europe. None of the country in Europe can face Russian Federation. So with this huge difference in capabilities from Russian Federation to any of the states in Europe, the Delta which is created here, the difference in capabilities, could be filled only with intelligence, that means real intelligence, with huge investments on all the fields in order to secure the situation, and with partners, friendships, allies, and any kind of coalitions of willing like this in order to stop them before it is too late, because after that what we are strengthening at this moment is the need for deterrence of the Russian Federation. NATO has made a lot of efforts into the initiative of enhanced forward presence in the northeastern part of Europe, which is working pretty good, because Russians they have Kaliningrad Oblast over there and they very quickly isolate the three Baltic states, but in the southeastern part the region is really really interesting because it has cultural differences, it has religion differences, it has different language differences, it has political differences, economical differences, and so on. So in the southeastern part of Europe you will see a Latin-speaking, language-speaking country, but with an orthodox religion, and as you know most of the orthodox are oriented toward the east, toward the towards the Russians, however Romania is 100% oriented toward the west and it's standing tall toward its construction of the young democracy. There are many comments on our young democracy, I know there are, but we prefer the way of democracy in order to preserve our identity, our values, our national interest. In all of this picture what Romania has done is that invested, it reaches at the political level an agreement of allocating two percent of the national GDP into the armed forces and with this allocation we have plans to modernize the armed forces and also to have some major acquisition programs like the Patriot surface to missile systems, the Heimers, the high mobility artillery rocket systems and also to improve our naval forces with four new multi-role corvettes, modernizing two of our three frigates and also modernizing the other side of the fleet. We are also thinking to some very interesting projects for the future, you will probably hear about it in the near time. In all of this picture we have our contribution into the hot areas like the theater of operations and I could mention here that if 30 years ago we couldn't think about projecting force or deploying forces into areas like Afghanistan now it is like the second nature for our forces and starting with the 90s we deploy forces to Angola, to Albania, to the western Balkans and Bosnia Herzegovina and also in Kosovo. We deployed our MiG-21s to the Baltic states in 2007 which was a very challenging mission to deploy MiG-21s and conduct air policing into the Baltic states, air space and we deployed our forces into Iraq and Afghanistan. This was a huge amount of effort for us. We were supported by our allies and partners mainly the US. We contributed to the peace process, we contributed to the stability operations in Afghanistan but and of course everything is coming with the price in all this process. In all these operations we lost 29 persons there. We will never forget them as really hard to look into the eyes of their relative and tell them that for them the mission was accomplished. Speaking about the special forces, special forces is at a high attention at the leadership of our defense staff. We invest quite a lot in the special forces in the last year. The transformation process started practically in 2017 when we created Joint Special Ops Command. We practically transformed all the tactical units and also we were looking to some modernization progress to acquire some new weapons and some other system for them. We also deployed special forces into the areas like Afghanistan, Iraq, Kosovo, Horn of Africa and other hotspots in the world and our military personnel have given all their best to accomplish their mission. We also had five fatalities from the special forces and we have many other tents and hundreds of people who were wounded and who bear all the marks of these fights. The thing is that special forces at this moment is seen in Romania as an outstanding tool to be involved into the gray area, into the gray zone, until the conventional conflict is to be started. And as you very well know, it is now a very low possibility to start a conventional aggression. However, until the NATO will decide to apply the Article 5, the collective defense, there will be a lot of challenges into how to deal with the security aspects, mainly in our region in the eastern flank of the Alliance. Speaking about the future, we really have to look into the efforts which has done with the Romanian armed forces and we didn't do this alone. We had great support from the United States. We had great support from the United States Special Ops Command with expertise. We have our special officer, liaison officer here, who is an outstanding support for Romanian special forces. And also we have Congress that approved the appropriated funds which allowed us to make some investment into developing the proper special forces. It was highly, and it is highly appreciated in Romania and we are looking forward to continue the security system program which is very successful for Romania. I'm very close to the end of my speech and I'm just looking forward that you may have any kind of questions for me, speaking about this special region of Europe. And I will stay in front of you and provide you the appropriate answers. Thank you. Thank you very much, General. So Kelly, and one other thing that kind of Willie didn't mention, you're a director of the NSC for Iran which is relevant to this conversation as well as being senior director for for strategy. I had the great joy of dealing with Iranian and Pakistani proxies in the last 10 years. So first of all thank you, Peter, for inviting me here today. It's great to actually be amongst former colleagues, Nick, and I worked very closely together in both the Bush and the Obama administrations. And of course my former colleague, Norm Roll, just walked into the room who we worked very closely together for quite a long time on these issues. So I would just say a few things and I'll keep my remarks relatively brief so we can get to questions. Obviously proxy warfare is here to stay and I anticipate that in the case of Iran and Pakistan specifically, it's going to probably intensify, especially as the United States sort of makes strategic policy choices around where it's been to vote focus of effort. And I think the shift to dealing with China in particular is going to be and what that might mean for our engagement and presence in the Middle East I think is going to have a tremendous impact on how Iran starts to position itself in the long term vis-a-vis its proxies. Likewise on Afghanistan policy, I anticipate that we'll return to that issue probably in the next year or so under President Trump and suspect that we could see a potential move to withdraw US forces in which case I assume that most of the regional actors, including Pakistan, with its proxy forces will intensify their hedging strategy in this regard. So I think this is going to be a much bigger trend of proxy engagement in the long term. You know it's it's it's easy for countries because it's a relatively low cost not risk-free but low risk way of interacting with other states it's it is a sort of substitute for conventional deterrents in the case of Iran in particular. They sort of view their proxy forces as a deterrent Israel and to the United States in particular. It is easier because you can work with local populations who have more credibility in the space and so I think in that case you've seen that a few times with the Iranians. There's also ideological reasons. I mean the case that Iran in particular the export of the Iranian revolution is a key fundamental foreign policy goal of the IRGC and the Iranian regime and so the proxy forces are a mechanism to deliver on that goal. Right now the Iranians have close to a quarter of a million proxy forces in the region if you count them all up and that's you know pretty significant. Now I think the nature of Iran's proxy engagement differs depending on country to country and I think not all Iranian proxies are the same. I think it's important to understand the differences between them and sort of almost a hierarchical set of differences frankly in terms of what's important to the regime. I would say Lebanese Hezbollah being at the top of the list in terms of proxies and Houthis being at sort of the bottom of the list in terms of proxy engagement. So they're not all the same. You know I think one aspect that we should keep an eye on as in particular with Hezbollah in the long term as Hezbollah becomes more engaged in the Lebanese state politically in positions of political power I think you could see a shift significant shift in the way Hezbollah interacts with the IRGC and the Iranian regime. The set of stakes for them local stakes are going to be different in the long term in terms of their political stature in Lebanon so there is there are areas where if you're looking at a U.S. strategy finding ways to exploit those those differences I think it's going to be key. I think in the case of Hezbollah and also sort of the IRGC the IRGC became basically a kind of train the trainer of proxies in the region for quite some time is something we dealt with a lot in the Obama administration and Lebanese Hezbollah I think as it's growing capability also is going to present it's going to have more independence and I think this is something that the Iranian regime is going to have to to deal with in the long term so I think it's important to look at the proxy relationships with the home state as not a static dynamic very much driven by what's happening in the countries where the proxies are engaged but also their own politics. In the case of Pakistan the biggest challenges are I mean I think proxies are part of the martial culture of Pakistan and you know getting the Pakistanis who obviously give up this tactic as a way of advancing their foreign and security policies extremely difficult in the context of South Asia and Central Asia you're also dealing with a government in Pakistan that is not unified in terms of its a knowledge of these issues internally in terms of you know what Hukhani what the Hukhani Network is doing in Afghanistan what it's not doing you have pockets of of information and others were completely in the dark and so when you're engaging with these governments it is very challenging to present to them a set of choices and have a coherent reaction from the Pakistani government that was that's obviously playing out in real time now Trump administration is pursuing a much more aggressive line with the Pakistani government removing security assistance etc. I think the challenge is that I'm not certain it will deliver a change in Pakistani behavior the Pakistanis are going to hedge in general I think they see American engagement and favor is relatively fickle and I think that's going to continue to drive their employment of proxies in Afghanistan so you know I think my sort of overarching comment as I think there are ways to exploit divisions tactically at least I also think we need to account for what is going to be strategically in their those countries interest to keep on doing what they're doing I think we can't disregard that that's they're in their national interest I think we fool ourselves if we think they're they're going to suddenly give up a means of warfare that they have been engaging in for for quite some time so I'll leave it at that and then we can get to the how to solve the problems well thank you Kelly I'm gonna ask each one of you a question and then open it to the floor so starting with you Kelly on Pakistan you know on August 21st 2017 Trump made the announcement about the new Afghan policy we're gonna it's going to be conditions-based not time-based the second he said is we're gonna get tough on Pakistan everything you just said suggests that that's kind of not gonna happen because we've tried carrots we tried sticks there's a great scene where Ali Gears James Baker what if they don't like carrots you know in the context of discussion of Iran so we we we tried everything and it seems that we the Trump administration sort of is it's wishful thinking just to say hey we're going to get tough with Pakistan because as long as we have troops in Afghanistan we need their ground route and their air route to supply those troops if we don't have troops in Afghanistan we don't care about this so I mean what how would I mean you've been following this for ten years or more right how do you in your discussions with the Pakistanis on the let's say the Akhani issue or and you know obviously the government as you said is not united and different levels of knowledge so I mean what what if anything and then also just adding to this if we if we're in a serious talk for the Taliban which we are now we obviously we can't be penalizing the Pakistanis and kind of so how do you solve this conundrum well clearly I didn't I think you know to your point about carrots and sticks I think we have to decide fundamentally what we care about and I think the Pakistanis have to do the same thing and I think it's more of a this is going to sound it's going to have to be some sort of negotiation of interests that goes on you know I do think you know for looking at the G locks and the A locks and tell everybody what those are ground lines of communication all these guys know what these are airlines of communication I think frankly I actually think that's less of an issue over time for us anyways I think the Pakistanis know that but we have a lot of other issues in the relationship nuclear issue India I think you know obviously the Trump administration is doubling down on what began the Obama administration which is a much more closer military and defense relationship with India and so you know these are the these are the things that drive neuralgia in Pakistani mindset so I think it has to be a strategic conversation about the boundaries of the relationship and how we're going to work together within those boundaries because I think the sort of like we're just going to punish you or we're going to give you favors isn't going to isn't going to shift the conversation any meaningful way now that's just you know yeah just off the cuff thought but that's what I think is missing right now general does so what is Putin's game plan in your world part of the world yeah very interesting question I could not answer instead of him but I am just doing my personal analysis and say that he's he really enjoyed in destabilizing the area he's really enjoying and doing whatever he can do in order to stop the advancement of NATO and the U.S. investment into the southeastern part of the country of the euro and now it does matter what what is thinking about and what kind of games he has because I'm sure that he has quite a lot of other games much more important that southeastern euro but in in our terms we are saying that instead of the investment investing too late into our capabilities and into our partnership and into our cohesion as NATO countries or EU countries we we need to look right now into the deterrence capabilities we have because I personally consider that this is the most important at this moment because if in the north-eastern side with the cleaning grid already existed there they they create he created a special effect over there because also cleaning grid has a lot of weapons system deployed in in that area and has also bubbles for air defense and for missile launchers however in the north-eastern part of Europe that it sounds like the level of cohesion reach a different level much upper than in the southeastern region of Europe so in the southeastern region we have the Black Sea which is governed by the Montreux convention so in terms of Navy involvement or NATO Navy involvement into the Black Sea that might be possible to a certain point we have Turkey a country which establishes own course in along the next month or years nobody knows exactly we have also states in the southeastern part of Europe like ours like Hungary like Bulgaria like Greece like the former Yugoslavian state so each and any of the country is responsible for their foreign policy and for their domestic policy however speaking from the regional point of view we have to work more into the cohesion side of the house thank you and Nick you've lived through the debates that you described obviously about whether to arm the Syrian the SDAF or and to what extent that would anger the Turks and maybe it would be helpful to tell the audience kind of what the discussion was under Obama and how it changed under Trump well I guess one distinction to draw and I think you can draw it pretty cleanly is that that some of the states were talking about turn to the proxy tool as a preferred tool because it allows them to achieve effect they would not be able to achieve but you know through their normal application of their military power or whatever their capabilities were the United States it's different it these are the turning to the proxy tool in most cases it's recognized as the discussion is proceeding but you're talking about a suboptimal less than desirable course of action but one that is preferable nonetheless to a large-scale deployment of u.s. ground forces or some of the other policy choices before us so that discussion in the waning days of the obama administration was going on in full recognition of all of those downside costs associated with support to the SDAF and those discussions included what kind of weaponry would be provided what what conditions would be placed on that provision of weaponry what kind of intelligence support how close would our forces our our our soft forces deploy in order to provide that advise and assist mission and yet we also were aware that making decisions on this would hand the new incoming trump administration with a significant set of facts on the ground and so I think there was a realization in the waning days of the obama administration that these were decisions that the trump administration or any new administration deserved to have a voice in and yet before there's a transition you can't have shared power between administrations so in some ways these decisions I think lagged for months and months and months beyond their natural life cycle because we were at a moment of political transition and it wasn't fair I put fair in quotes there it wasn't fair somehow to poison the relationship with turkey as the obama administration was walking out the door on the other hand if you're the trump administration you walk in the door and you say wait a minute you've left me with almost no option other than to proceed on this course which you were already pursuing you've got a strategy that says we're going to defeat isis using ground forces tbd as I said earlier and the only tbd available uh is a syrian democratic force now our turkish partners again and again and again promised that there were other available forces that would more meet the criteria uh that we would all prefer that would have not you know created the same set of problems uh that having a Kurdish force did but that force never materialized and so I think over time the trump administration quickly realized that if the president did prioritize defeating isis and accelerating the progress of the campaign against mozal and then on into to the eastern part of syria there was only one means to do that and that meant biting the bullet accepting the cost associated with uh with angering turkey at a time when we were angering turkey on a number of other fronts at the same time and it was understood that that would that would likely have long-term repercussions for nato and long-term repercussions for our bilateral relationship and that's I think the world we're in today and the george toppy bush administration it made a similar kind of decision when they sort of essentially deferred the afghan decision to the obama administration who was incoming right you could argue that too I mean and again this is you know political transitions don't often line up with strategic imperatives on the ground and yet yeah so and I and I actually give both administrations credit because they were trying not to tie the hands of their successors in ways that would have been that would have seemed unfair if you have a question raise your hand and wait for the mic and identify yourself please okay gentlemen over here is it brian williams from you uh darkness a couple of you touched on obviously both the pakistani and the iranian use of proxies uh and uh both of their proxies of the taliban and associated networks in afghanistan and uh and the shia militias in uh in iraq uh collectively inflicted thousands of casualties on us forces uh in the post 9 11 era it struck me as a reporter um how hard the u.s. government tried to keep the roles of both pakistan and iran from the american people during that period um and i wonder uh if you guys could shed some light on why that happened and whether or not you think uh you think that was the right approach to take well thank you kelly for both of these i suppose that a requires you agree with the premise i'm not sure i agree with the premise um in fact if you look at public testimony of defense officials state officials and on the rare occasions when we in the intelligence community would give testimony we were pretty up front about um what we saw proxy forces from a backed by iran backed by pakistan doing including their impact on on us forces deployed in those areas areas of operation what we lack was good policy choices what we lack was the leverage i think kelly talked a little bit about this in the case of pakistan uh in pakistan it was a case of leverage and as as peter noted characteristics of all you know if i if i could count the number of times the mission to go deliver the no kidding this time we really need a message to a pakistani security official dating back to richard armorych's armorych's first trips to pakistan in the aftermath of 9 11 and going intermittently every year or two since it wasn't for lack of awareness or self-awareness that we were in an unacceptable situation on the iranian side i'll defer to kelly on this but i think there was just a question of trying to play out in our minds how escalation would look and what options we had that would allow us to inflict retribution if you were if you if you would on iranian proxy forces but wouldn't leave us in a worst place yeah i think that's right i mean i also disagree with the premise of the question i mean i had someone who had to testify of myself one pakistan we've you know got tons of questions and we talked about these issues pretty openly um i you know i think on iran uh obviously they inflicted a lot of casualties on us uh during the iraq war uh and in the in the period right before we drew down and i think that was pretty open i think most people talked about these issues pretty uh publicly but to to nick's point i think uh the policy options were you know suboptimal in many in many spaces um so i guess i just would echo what you have laid out gentleman here hello uh my goodness uh my name is ali joffrey i work with nsi uh i was wondering if you could comment a bit on china's use of uh proxy forces if it all it seems to me that uh their participation in the gray zone of conflict happens more within um economic um and pseudo diplomatic um avenues and not so much kinetic i was wondering if any of you could comment on their use or not use of of that as a policy option thank you i'll take that um china is very effective at using gray zone tactics uh you know in terms of looking at just the south china sea for example is the sort of the example that everyone talks about in terms of gray zone uh they take action just below the threshold of what they think will produce a direct conflict with the united states or its allies and they do it slowly and over time um which is you know generally how they approach those issues and they use you know similar to russia using little green men uh they use what they have as a maritime militia uh which is essentially just pla guys in a you know fishing boat uh you know to conduct some of these uh these operations in the south china sea so they are actively engaged uh in this space um and you know i think for in terms of policy options as it relates to to chinese gray zone tactics i think uh you know one thing people need to understand is that the united states also feels constricted in its ability to execute options which is something uh that people tend to leave out of gray zone discussions it's usually what the the tactic of the adversary but what gray zone does to the to the state that's being affected is it the state that's being affected feels like it has constraints uh that say for example china doesn't have so we are a law abiding international law abiding nation in terms of how we execute uh naval operations uh we follow the rules and the chinese don't follow the rules and so we feel like we're in a bit of a constricted space in the on these issues so that's going to be a challenge going forward um for the u.s and china context on the south china sea kelly you mentioned the hoodies what happens if an iranian supplied missile blows up in downtown riyadh and killed several dozen people they fired uh the last time um i mean you know something like 200 missiles already uh most of them are you know have fallen short but play out for us what um what what would happen next well i think uh you know if you're assuming that with us operations have sort of ceased in terms of support to the Saudis of the stage right yeah so i think that uh obviously the Saudis would hit back but where would they do that uh i i i tend to think they will do that uh in yemen um i you know there was an operation some sort happened in uh in iran which killed a lot of security officials not too long ago you could see something like that as well and then get an open question about whose uh whose responsibility that was um so i think but you know the Saudis i think will tend to cabinet to uh retaliation inside yemen um but that's my speculation and for for any of you you know we we've always tried to encourage the gulf states to kind of take more responsibility for their own security which is now what they're doing and suddenly the Emiratis are showing up in yemen and libya and jibouti and Somalia and the Saudis have got this very aggressive foreign policy um i mean is this a good thing that they are um sort of have a more aggressive foreign policy which is you know enabled by proxy warfare or is it a bad thing and yet you just signed a letter from with 30 senior obama officials it was it yesterday or yesterday yesterday sort of calling for an end to any american support but obviously this began under obama so and it will thought so on on on any of those questions i mean throughout the post 9 11 period we've looked to build partnerships with countries with the idea that over time their capabilities would improve and that we would in a sense bear less of a burden for carrying out counterterrorism but also other military operations in some of these theaters because our partners would get better and achieve higher levels of capability that is happening that is playing out just as you described peter but you also lose control you lose some ability to dictate outcomes when you ask your partners to step forward and play that role now the governor you have over that of course is sharing of equipment sharing of intelligence and sharing provision of training and assistance and that ideally gives us some ability to calibrate that support based on the way these forces conduct themselves and whether the forces are acting in a way that is consistent with u.s. interest but it's imperfect um i would argue though that i don't i'm not sure i would prefer an alternate scenario in yemen where we were bearing right now you know we're carrying out a very high operational tempo of actions against a qap you know enclaves on the ground in different parts different remote parts of yemen that is not something that i would argue would be in our interest right now and so we're left accepting or at least to date having accepted imperfect conduct of a campaign with devastating i would argue humanitarian consequences i mean i i would just add i mean people can read the letter if they would like i think we acknowledge the fact that we made a set of decisions we placed a bet frankly on on this assistance to Saudi Arabia and i i think at this point we're evaluating in the context of a lot of issues you know whether that it was the right decision and what it's going to produce and to the next point you know i think our control over what the Saudis are doing is is waning and you know whether or not we actually have leveraged through some of those means i think is a really open question when it's something that Saudi Arabia is doing in the context of its own interest and not necessarily ours so i think that leaves open questions i mean this is the challenge of all partner operations our partners have their own interests and so you know i tend to fall on the side of it's better for us not to be involved in this particular set of engagements but i also think there's been an imbalance in our approach to Yemen in general we've sort of outsourced it to the Saudis we've de-emphasized diplomacy which i think is what is needed now to bring this terrible conflict to an end and so i think what we need actually is a much more vigorous us diplomatic engagement strategy to get the parties to come to the table and make peace under the auspices of the UN Security Council resolution so over the weekend we stopped air or fueling of Saudi planes and do you have a sense of we stopped intelligence sharing to what extent that would really impede the Saudis or i mean as you said they're doing something that's so clearly in their own national interest that it you know probably wouldn't make much of a difference it may it may or may not but at least the United States wouldn't be complicit in some of the the targeting that's occurred uh... any other questions Mr. Allen back thank you Jeff Selden from VOA the new national security and defense strategies have placed a much greater emphasis on great power competitions US conflicts with Russia with China how do you see the use of proxy warfare and perhaps use of terrorist groups by Russia China Iran and others playing into that is that being accounted for enough and when looking at proxies we've been talking a lot about actual military forces on the ground what about the use of groups like some of the nationalist groups across Europe and the US that Russia has been establishing ties with it's a great question because i think in some ways we honed in very quickly and very narrowly on proxy warfare in terms of combat forces when in fact i think the china question elicited from your colleague out there elicited in my mind a whole set of questions about gray zone conflict that in the cyber domain in the political domain and i'll tell you those are even harder to counter because if you're talking about asymmetric warfare the asymmetry is even broader wider deeper whatever whatever adjective you want to use in that set and also attribution is that much more complicated you know attribution isn't difficult with a weapon system because ultimately there's a serial number or a tag or some indicator that tells you upstream who was the provisioner of that military technology or that that resulted in whatever military action but who who's the hidden hand behind a nationalist group interfering in a political election much harder to say and also much harder to develop a set of of counter options that are meaningful and i think the national security strategy that you're talking about are the national counter terrorism strategy that was meant that came out a few weeks ago i think did a good job of articulating that state power conflict is not now much more central to our set of national security challenges but it may not have you know adequately articulated that great power conflict can also be carried out in the gray zone i think that's right another question over here this gentleman here uh george perrier the national center for urban operations um speaking to the success of proxy warfare that you have described by great powers from the eastern states to the middle east and obviously as an application by the united states and then the united states pretty successful deterrence of decisive action through its training and maintenance of um conventional combat means can you guys speak to potentially the tension that exists between this newfound um reliance upon special operations forces to carry out the type of proxy warfare from proxy warfare to counter terrorism that's existing across the spectrum and as the do d attempts to keep up with that workload through various i think structural changes such as the security force assistance brigades can you guys speak to how policy needs to accommodate that kind of disconnect in the sense of the workload kind of outpacing the forces that are um formatted to do it i guess i would say that we thought for for much of the last 15 17 years we've thought of our our unique high-end soft capability as being at the service of a counterterrorism mission that was what the preponderance of the of the of the focus was and yet as some of this great power conflict has reemerged as being a key you know part of our set of national security challenges the soft contribution to those to responding to those threats is every bit as as profound and every bit as resource intensive and yet i don't know that we've thus far engaged in the investment that would allow us to handle both and i i started to see this a little bit in the in the my waning days of government service where i was starting to see some of my military colleagues in the counter in the counterterrorism community starting to think well i need to worry about north asia as well i i may have missions that impact i may have to develop capability to deal with north northeast asia scenarios or i may have to cope with a china set of scenarios and all of a sudden they're not simply thinking about how to go after loose nukes in a country in the middle east or or or carrying out direct action missions in support of our counterterrorism objectives so that the the array of missions that we're throwing at our soft community has gotten wider and wider and i don't think we've sized up or scaled up in a way that would be fair to our soft and is that even possible is that a contest well yeah that's a multi-decade investment those are you know force management decisions that are going to be made by the the secretary and the chairman i would agree with nick in terms of the trade offs and the sort of stretching thin of our special operators i will tell you a story when i went to visit so come for the first time when i was at the duty uh this is like 24 early 2014 i went down there and i was the acting assistant secretary for asia i get there and i on my mind are a couple things afghanistan pakistan which were in my bucket of responsibility but china and of course the rest of asia was as well and so one of the first questions i asked of my so come in a lot of others when i got there was so how are you guys thinking about china and i got blank stairs from pretty much everybody i spoke to and that was 2014 so that wasn't that long ago so i think you know to nick's point i think we're in a bit of a lag stage in terms of both implementing what the prioritization of mission is across the soft enterprise how the allocation of forces should go across the combatant commands with respect to the mission set so i i think there is a lag to your point and i just think it's going to take a little bit of time for the apparatus to catch up i'm rizlota sorry just a quick and more conceptual question as i've been listening to you i've i've been thinking of course i know what proxy war is and then as i'm listening i'm wondering what's the line between proxy war where they're fighting for us and allies where we're a kind of i'm a network theorist so i always think about people connected in sort of more horizontal clusters of adversaries nick you lived this well but you know well you know you already had me wondering if i'm allowed to use sub-state actor anymore based on what you said at the beginning of the conversation because it in many ways it's not really much of a proxy at all it is really where attribution is so obvious and so direct it isn't really a true proxy it is simply another form of state action and to the point of the question asked about casualties inflicted on us forces and we're kidding ourselves if it was if we didn't know and believe that it was iranian technology behind much of what was what was causing deaths of american service personnel in iraq so in that sense proxy can be a misleading term and might take us away from really understanding what we're talking about in terms of a challenge to national security i guess where i started though was trying to think of with my comments at the beginning proxy tends to be a dirty word we tend to think of it as negative having a negative impact on our national security but there are opportunities does provide certain opportunities to find the right way it provides opportunities where at a lesser cost than might otherwise be the case the u.s. can engage in action to protect some of our core national security interests that is not an ideal way to proceed in every in every case but it is it is a part of the toolkit that i think we could our adversaries are clearly making use of when we'd be crazy not to as well one more question can you wait for the mic hi my name is uh jeff gobel i uh work in the j3 international at us so com so for the general sir you know that large segments of the special operations capability are perfectly suited for assisting nation states with undertaking proxy warfare so in the transformation of romanian soft are you are you thinking about that in terms of a capability in your soft community or is it traditional you know direct action counterterrorism force that you're thinking about great question thank you very much yes of course we thought about this and with the transformation process we are now with undergoing with the special forces we're thinking for Romania into the southeastern part of Europe to become a regional hub in terms of trying to look at the other special forces into the region and try to establish any kind of cooperation with them and also for the states like Georgia and Moldova to establish a kind of exchange and training programs in order to improve the capabilities we want to bring them for special ops exercises we want to bring them into the joint exercises we want to perform last year sorry next year and with this occasion we we would like them to to be together with us and also practice all the procedures as we are going into the training program right now for the Romanian special ops forces we have established the school the tactical school for special operation forces and it is a huge investment in this application school and we are providing our operators with pre-training the pre-qualification course the qualification course and also the advanced courses we want to implement this will give us the opportunity to increase the level of training for our troops of course with their intention to use them in the hot areas mainly focused into the region here but also continue our commitment with Afghanistan, Iraq and Kosovo of course okay well we're gonna leave it there there is such a thing as a free launch in Washington that's going to happen next door and we want to thank Nick and the general and Kelly very much for their brilliant observations good afternoon welcome back I'm Megan Keeler Petigrew I'm the chief operating officer of the Global Soft Foundation and I wanted to thank New America again for hosting us here today and the McCain Institute at Arizona State and the Daniel Morgan Graduate School of National Security and I also wanted to extend a special thank you to the many military officers we have here from our partner nations that made a trip to come up for this event so thank you and we're glad you're here I only agreed to introduce this panel because there was no expertise required on artificial intelligence on my part so I'm looking forward to the discussion here about how AI might both clarify and further complicate this murky space called the gray zone so to my immediate right is Cara Frederick she is a research associate for the technology and national security program at the Center for New American Security prior to joining CNASH she worked at Facebook and was also an intelligence analyst for the Navy and the Department of Defense next is Kristen Sharp she directs the initiative here at New America on work workers and technology and prior to leading this team she has held many staff positions and on the Hill and the U.S. Senate though specifically and Eric Grant is our engineer on the panel he is the technical director of the mission support and modernization in Raytheon's intelligence information and services division and lastly our moderator for this afternoon is Peter Singer he's a strategist and senior fellow here at New America and Peter is the author of several books the most recent of which like war was just published just a few weeks ago so thank you for joining us and I look forward to the discussion so thank you very much for the kind introduction and also a big thanks again to our partners who helped make this event and this panel happen and then finally a special thanks goes to my colleague Kristen who joined us at the last minute we had Wendy Anderson was to join us but she is fighting there's a really bad flu if any of you have kids you know it's going around so she was unable to join actually emailed saying she was going to the doctor this morning it was that bad so we particularly it's great to be in a place that has in-house experts who both know the topic and are willing to help out so again thank you for joining us I'm particularly excited about this panel because it brings into our discussion what is I think arguably the not only the most exciting but most important area of technology change that's going on both in business and in geopolitics this is a space with artificial intelligence where essentially there is a arms race that's going on between companies if you look at the Fortune 500 of them 244 have identified AI as the core to their business strategy moving forward and that's true whether we are talking about traditional technology companies like the Googles the Facebooks but it's also true about companies like John Deere as an aside how do we know it's 244 because an algorithm went through and read all the business strategies for us in turn this is clearly a defense and geopolitical arms race for example China is very open about saying that it wants to be the world leader in AI by the year 2030 and has a national strategy of investment to make that come true and we have a wide array of activities across both US technology and defense industry as well as the Pentagon so it's a crucial space I think an interesting question which we'll get into is in terms of the historic parallels is this going to be a technology like the airplane like the tank or is it going to be a technology like the steam engine or like electricity in terms of the effect on the future of work and the future of war politics so I thought that might be an interesting way to kick us off to ask Kristin essentially what do you see as the key applications of this technology to work and in particular what areas are more versus less apt to be automated perfect thanks for letting me kick it off and many thanks to all of you for letting a non-military expert participate in the discussion and lend some perspective particularly this week and this weekend also want to start out by saying thank you to all of you who have provided military service to the country and for Veterans Day we really appreciate your service so thanks very much for that so in the world of work automation and artificial intelligence the things that usually get talked about as the most easily automatable are things that are repetitive and hierarchical so things that require either manual service that is done the same way on a repeated basis or things that are aggregating information in a way that is a repetitive process so in some sort of practical examples that might have crossover stuff driving any kind of driving is easily automatable because it's a sort of constant looking at the particular sort of street signs or things that are happening and the machine processing that information so that's one that gets talked about as an occupation that is at high risk but more than that things that use big, big swaths of data to make decisions about things so in the particularly in the insurance or healthcare or financial industries that use data to process paperwork so loan back office loan processing things that use big pieces of information to decide whether something is a good application of something or not that stuff is highly likely to be automated in short order so those kinds of things where somebody gives direction and then somebody else carries it out on a repeated basis great so let's take this into the military realm and in specific the special operations community so I'm going to first ask Eric and then I'm going to turn to you Kara on terms of what do you see as key needs that the special operations community has that AI might be applied to so sure again thank you for having me here today and also I want to echo the thanks to all the veterans out there for service happy to be here and waiting for some questions from my perspective I'd say you know the special operators that are out there today are facing a much more difficult role than they have in the past because not only are they trying to gain technology and have that edge but so are the adversaries and so you're now going into an environment where the internet of things is there you have devices that are monitoring you from all different types of phenomenologies whether it's video or cameras or RF energy or other types of things tracking your digital fingerprints as you get through so special operators are going to have some difficulty in terms of some of the traditional ways they operate that they're going to have to adapt to and they can apply AI to some of those so I think the idea of the soldier whether it's a special operator or even a future infantry man or any soldier it's really going to be that that AI of the edge right they're going to be carrying AI capability with them they're going to be trying to connect back through communications and to broader AI capability speeding them the right information at the right time so whether it's drone technology whether it's using things like communications for resiliency so that I can always get my communications back to headquarters especially in areas where that might be a little bit contested it could be things like understanding the environment I'm in from navigability and that could be urban that could be something like a military version of Google Maps where I have other things identified that I need to watch out for maybe there's certain cameras I'm awarding or other biometric identifiers I have to watch out for but it's going to be kind of this AI versus AI and some of these these gray zones where technology is going to be there in urban environments and you're going to have to apply technology to avoid you know if you're trying to avoid detection or avoid understanding what you're trying to do there in a proxy battle it'll be similar you're going to have AI against AI in terms of people trying to employ different tactics and that will obviously extend into cyber so you know I think it's there's a wealth of opportunity out there for special operations to employ the technology but they also have to understand how it's being employed against them I think that's going to be one of the to your point that the ongoing battle of AI against AI not just company-wide but also across countries and across organizations that are trying to each use it to gain advantage over someone else can I build on that is this the technology the way you're framing it that will be widely proliferated if we think about compared to previous capabilities yes yeah I think when you're seeing that already even within the commercial sector if you look at how much money's going for Silicon Valley for AI technology it's just continuing to take off and I think when you look at the defense department and some of the recent announcements from DARPA some of the investments they want to make in AI research the stand-up of the joint AI center you look internationally at what other countries are doing you know there is a little bit of an AI arms race going on and but everybody's trying to figure out how do you really apply it right they understand they want it the trick is how do you actually make use of it in the real world and cause effect to your advantage and I think that's where back to the special operation side is where the focus has to be on I can do a lot of things with AI but what are the things that are going to allow me to accomplish my mission more effectively than I can today without it great so again I'll echo all the sentiments that I've been voiced here today and I spent the entirety of my intelligence community career supporting special operations forces both in the field and at home so it's a it's a huge pleasure to be with you guys here today and really appreciate the opportunity to sort of share what I've been working on and what the Center for New American Security has been working on especially with regard to artificial intelligence and international security and its applications to the battle space so I would again echo what Kristen said it makes perfect sense I was an intelligence analyst for for SOF so I think that Intel Exploitation Surveillance and Analysis is one of the things that AI can do very effectively so generally AI can help make it easier for human analysts to sift through large volumes large swaths of data so that's sort of the groundwork that we're working off of in terms of force pro threat triage some of the things that we'll kind of elaborate on a little later in this program is what AI can do with regard to help racking and sacking priorities right how do you ID predictive trends and have it lead to conclusions how do you let the machine do the work first and then reserve the analytical capacity of the human being to to help make the decisions later so Project Maven undoubtedly is gonna come up so I might as well just say it first so it's initial mission right the DOD's algorithmic warfare cross functional team established able 2017 its initial mission was to field technology to augment or automate pet for UAS and mid altitude full motion video in support of the defeat ISIS campaign you've seen it in the news because of Google basically stopping work on or not renewing the contract to support it with the subject recognition algorithms but that type of machine learning computer vision that is what we can build on to have applications to the battlefield so very almost simple ways to apply AI to some of the missions that we're working on with regard to special operations I think a lot about SSE obviously as a former intel analyst the person who was sort of going through those excel spreadsheets of a lot of metadata so if AI can help us parse through that to maybe do some follow on missions because things move so quickly especially when you're in theater if you can get if you can get meaning out of the information quickly then that's going to help you be more effective in the battle phase so those are I think two things really quickly that can be applied to special operations forces and AI how might it change you come into this with a very interesting perspective how might it change the workforce when it comes to example number of analysts needed and the type of training that's a good and you're not in your head which means you've got a way and all of you're going to have to weigh in next on that so in terms of I think it can help make the analysts more efficient right so the Air Force is making a lot inroads in this way they've actually advertised the fact that they want to reduce the number of human beings working specific problems like when you're sitting in a talk you have a line of FMV analysts sort of staring okay that's a tree that's a truck trying to figure out what these things are if you can get a machine to do that then those FMV analysts whether for they're from NGA etc they can be again their capacity to make harder decisions can be reserved for other other missions so they don't necessarily have to be eliminated you know you're not gonna some of their tasks can be eliminated but they basically just make the person more efficient and more effective so Kristen you want to yeah so I think that's a hugely important point that most AI technology can be used to augment human capabilities not replace human capabilities when we talk about people having their jobs automated or changed as a result of artificial intelligence we often talk about it as something that is significantly changed rather than something that's eliminated and that's because the way AI works usually is that it takes out of all this huge pool of data it takes a million tiny signals that an individual human wouldn't pick up on even if he or she knew what the landscape of the information was and takes all those signals and pulls out the ones that are important and that are indicative of something else and so sure this is a hugely unrealistic example but roll with me for a sec if there was a situation in which you're watching a terrorist enclave or something and the AI spits out that there's always a huge uptick in activity on a Tuesday and the it's the intelligence officer that would say oh well that's actually not that significant because there's a market on Tuesday in that village but you need somebody to you need the AI to pick out here are the markers of things that are happening that are different than a pattern and then the analyst takes that information and says here's what that means and so in terms of workforce and in thinking about workforce it's how do you develop the the context and landscape analysis so that the people know when to interpret the data in different ways sure yeah and I think to that end I think the how the human interacts and the human machine teaming I mean a lot of the research is showing that the machine by itself or the human by itself are not as effective as both together and so I think you know certainly from the defense department's perspective and everything they've been advertising they're not necessarily looking at taking the human out of the loop they're looking at making the human that's on that loop more effective whether it's a soldier out in the field or whether it's an analyst back in a in a talk whether it's somebody repairing an aircraft I mean it's it's how to make those different roles more effective by using the technology and the information that AI can deliver and I would like to also point out that from the perspective of how how it's going to influence the operations a lot of times they're not thinking you know the rush is to get the algorithm out there and get that FMV so that the throw the algorithm out there have to identify the objects and and remove the analyst from the loop however what they're finding is it takes a long time to train all the different objects that analyst is watching for and so even Project Maven has not been able to remove those analysts because those analysts still see other things that have the machine hasn't been trained on so it takes time and as you're doing that you have to work through the con up so what's the analyst's new role that analyst now has effectively a a machine beside helping him or her but they have to understand how to interact with it how to use it to affect how does it change their workflow and their their daily working environment how does it change their output and that whole process has to be analyzed and and adapted it affects the TTPs on the military side and that has to be part of the whole process it can't just be the shiny new algorithm that I throw into operations it's really got to be thought through from a con ups perspective how does this really change the way that person does their job and how does it change the outcome that we're looking for I find it's often difficult for people to understand how real this technology is so I'd like to ask each of you to name and sort of describe for us one either research project or real world actual application of the technology that you've seen in the last year that was particularly exciting and exciting might be defined as wow that was great to wow that was really scary exciting is open to interpretation so it's we'll begin on the far end and work our way back yep so I'm really interested in data visualization platforms so there's some software that basically allows you to control multiple reapers at a time in one space so it's the idea of these predictive indicators all of this analysis is sort of running around in the background but then when there's some sort of anomaly you're pinged right so when we think of the future of intelligence analysis we're sort of looking for these pings and then figuring out how to analyze them so if you can get everything in front of your face in one go then to me the more the better when you can see in one place the most the most sort of novel and interesting real world application I've seen recently was in a Chinese company and one that was that's you already sent a shutter well in commercial applications not in military applications but it was in analyzing who is eligible for credit and in what circumstances and the company had a platform for analyzing all of the different kinds of signals that it sends for who is credit worthy and in what circumstances and because of that was able to have a much much higher threshold of of evaluating whether somebody was going to pay money back that they had borrowed to make a purchase and in what circumstances their their their fraud rate was 99.9 percent effective at detection as opposed to commercial banking industry right now which is about 97 or 98 percent and so because of that they were able to take people who either hadn't had access to credit or hadn't had a credit score you know the Chinese equivalent of that before and allow them to make purchases in different ways and so I think that there's some real potential applications for economic stability in using big sets of data differently than we do now I mean there's a lot a lot of things out there to pick on so I'm actually going to take it a little bit of a different direction and talk about counter AI you know there there's a lot of rushing out to go employ AI without necessarily understanding the inherent biases that are in the algorithms or necessarily the you know the way it's going to operate and a lot of AI is not very explainable so for example MIT's been doing some work on looking at can I actually trick some of these computer vision algorithms to think what it's seen as something else and they have some interesting examples where they 3D printed a turtle and were able to convince the algorithm it was something else and so that could be done from any angle so it was much more complicated than well I'm just looking at it from this direction and I can trick it it was tricking it from every possible viewing angle to think it was something else and basically it exploited a flaw in the algorithm and so I think there's a lot of research going into from the military application is if I'm going to put these capabilities out there and I'm going to ask the soldiers and sailors and airmen and Marines to trust this I've got to make sure that it can't be tampered and so the research is going into the security of the data how do I protect the data to make sure I don't train on the wrong data how do I protect the integrity of the algorithm and how do I really understand the vulnerabilities or get to that explainability of the algorithms so that the human understands why it gave me this course of action I think that research is going to be critical to really getting the adoption into the the military community and having people trust it and being able to take it to war with them I'll abuse my moderator's role by bringing in the example of just a couple weeks ago saw there's been a lot of discussion around what are known as deep fakes which is applying this technology to create hyper realistic online imagery and the like and it's been described rightly as a national security threat that you will be able to for example create imagery of U.S. soldiers or allies committing a war crime and it will go viral and it will be impossible to detect whether it's real or not and the application of it that I saw was someone took the clip out of the movie solo which is the was the recent Star Wars prequel about a young Han solo many people in this room probably did not go see it because it wasn't successful they basically layered into it imagery of a young Harrison Ford so it looked like a it was a hyper realistic as if Harrison Ford but the young version of them was starring as a young Han solo and for me one it was way better than the original but two it captured the duality of this technology and terms of both deep fakes we're seeing this used in a good way for entertainment if we also can quickly come up with examples of it being weaponized and also showed how it's a technology that will inherently proliferate this was someone did it for fun and yet achieved something that a Hollywood studio couldn't have even done a couple years earlier let alone a national government and so really sort of for me captured a lot of the interesting things here let's get to let's get back to being particularly wonky what are the key barriers to utilizing this technology to its utmost barriers in terms of research in terms of acquisitions at DoD what are some of the key challenges to us actually gaining the positive sides and when we'll begin we'll kind of last time we went from far to close we'll go from close to far I mean I think from the research side the data right getting good data that you can take in train on is probably one of the biggest challenges you know that's why folks like Google and a lot of the big platform companies I mean they they harvest all that data because then they can train their algorithms on it when we look at what the defense industry is doing you know and what the government wants to do that's a different set of data and in some cases you may not have a lot of data on the things you're interested in those those targets may not have as much information you know so getting access to that data and then being able to label it appropriately and have it conditioned so that an algorithm can exploit it is really important and that takes a lot of effort and I think that's one of the lessons they're learning on things like Project Maven where that takes time I mean that doesn't just happen overnight you have it's a very manual process in some cases that go in and hand label all that imagery to make sure it knows that that's really a tank and when I see that tank in five other images that's still a tank that that all had to be hand labeled so there's a lot of energy that goes into doing that and I think that that slows down the the adoption right that's that's another piece so the research that speeds that process up will be important to see how we can get to label training data faster I think on the acquisition side it's just part of the whole discussion on OTAs and how do we contract faster and you know getting around some of the far limitations and the and the rush to bring some of this technology into the into the DoD space a lot of different things are being tried but I think in terms of AI it's really going to come down to even when I do the research and I have the algorithm can I get people to adopt it and use it can I get it into the people who cares hands and will they actually trust it and use it as part of the weapon system I mean I think that's that's going to be the major barriers overcoming that that leap of faith that says I'm going to put my trust in this this algorithm to go out and execute a mission for me because I'm ultimately accountable for the outcome how do we get past that leap of faith a lot of testing and that's part of the problem is when you go to test that algorithm you might get a slightly different answer today than you did yesterday and the explainability is very important so I think the research into how do you really get to explain ability and be able to peel back some of the layers of those algorithms so it's not quite a black box you can kind of start to understand you know good examples like navigation if you look at your Google Maps and you plot in your course you can generally figure out why I picked that course because you can see the traffic and it makes intuitive sense to you but a lot of these algorithms it's not necessarily intuitive to a human why it made that decision and so I think the more we can have work going on around that that helps peel that out and make that available to the human to understand the trust and build that trust and that this thing is actually giving me the answer I need that'll over time build up but I think it's going to take a lot of different testing than the DOD does today and how they approach test a lot of modeling and simulation and a lot of computing to really look at how that those outcomes can be predicted in the future I mean along those same lines I think that one of the things that's uniquely troubling in the military context is transparent I mean the when you talk to people about what they fear or appreciate about automated systems and artificial intelligence it's the more they know about how narrow or broad the scope of what the the artificial intelligence analysis is doing the less fearful they are of it and so like I mean people the sort of constant narrative in the media is that we're going to have robots that can do translation and that can you make decisions and that can analyze data and that can do all these things and in reality there are not systems that can do all of those things at once right now there are systems that can do some of those things and do some of them better than others and like people people want to know what kinds of information goes into creating the algorithms and what you're getting out of it and how you use it and that understanding the both the sort of parameters of what it's doing and how it got to that decision-making process is really helpful in in having people accept the systems themselves and how do you do that in the military context when you're trying to keep some of the information you know from adversaries or or it's classified or whatever it's a it's a bigger problem I think in this context interesting one at least and then I think a huge foundational element too is sort of creating an environment that encourages the behaviors that you're seeking right so this could be IT wise this could be this is cultural as well so instead of soft very platform-centric right we're going to get this Reaper it's going to be magic it's going to do everything but instead of thinking as a platform in a platform-centric way we can think in a data-centric way so basically everything that you were saying Eric so when you start with the data and you work backwards working towards the same commercially relevant standards if we can get to that point we need to be able to identify where commercial methods can generate efficiencies by integrating system development and operations kind of what industry refers to as DevOps so you'll probably hear that a lot I think it works I think it's been proven to work especially in some small isolated DoD context Castle Run Air Force Lab in Boston they're doing that there and then sort of an interesting idea that was posed by an intelligence community representative is we tend to have a opt-in culture right so when you get the pop-ups that DoD in the IC you sort of opt-in to provide data well what if instead of opting in we automatically we had to opt out I think it would really change the quality of the data that we were getting and generate better efficiencies as a result great so I want to end with one last question what is the single biggest or maybe better phrase most annoying to you misconception about this space that you'd like to dispel i.e. that the audience should walk away from here never thinking again so we'll begin with you Eric and then go down the road well I think we kind of already talked about it it's that these are not silver bullets right these algorithms are not going to just change fundamentally the way softforces or any forces fight there's going to be a lot of work that goes into how to employ them and I think people have to understand there's a lot of neat things they can do I think from the intelligence analysis perspective tremendous opportunities to process that data and really give those analysts much more information and power at their fingertips and get them out out of having to watch predator video all the time but I think at the end of the day it really comes down to understanding that it's more than just a software application on your iPhone or on your desktop or or that the soldiers carrying around it it's really about the what's the intended outcome you desire with that algorithm and how are you going to get the human to use it in a way that gets to that outcome because the human machine teaming is going to be critical to success in the DOD with this I mean he kind of took mine but my biggest misconception is that people don't have anything to do with this and that in truth like it is people who write the algorithms it is people who put the input into what the scope is going to be it's people who interpret the information at the end it's people who sort of talk about how publicly how you're using the information all of that is is designed and run by people and we need to have people who understand it effectively and I think on that note too and I could be kind of a poor offender here but I think that when people think that the private sector and the public sector sector divide is insurmountable I don't actually think that's the case I think I mean take myself for an example right I went to Facebook and now I came back working tech policy issues with the intent of forwarding American interests across the globe so I think that as long as there are very interesting problems to work on you're going to attract the next generation of talented engineers the next generation of software engineers who are going to want to figure out these problems and do it in the service of democratic nations of their own nation so I don't think that there's an insurmountable schism between the private sector and the government I think as long as we figure out ways to bridge that divide then people are going to keep coming back to the government and be very interested in helping to solve these problems in a safe and robust and reliable way especially as regards AI great so let's open it up to questions from all of you if you could raise your hand and wait for the mic and then introduce yourself that would be great so first hand it's over here on your left Hi there my name is Brigham Bechtel and I'm with Mark Logic so it was really great to hear you talk about data and its central point here what are we going to do about ontology use the word tank which we all know means different things in every service just as an example so how are we going to overcome some of these ontological problems of what something means when we're teaching a computer I think NLU so far the use of natural language understanding and natural language processing we're kind of not there yet but there have been initiatives within the government to sort of work that out on the sign side so those are sort of in incipient stages commercially they're making inroads semantic analysis is a big thing in commercial industry intelligence analysis I think it's moving a little faster but I think that certain agencies are really thinking seriously about this and using NLP to sort of solve those issues but it's not there yet it's when you kind of weaponize that is what I'm a little more concerned of you talk about spear fishing automated spear fishing especially at scale the information security pieces is huge with regard to AI and what we're going to have to defend against in the future deep fakes notwithstanding and I think natural language processing can sort of be employed to sort of figure out the semantic interpretations of things and then turn it against you so I think both offensively and defensively these are it's a race as usual and but the right people are thinking about these things in my experience yeah I mean I think Cara said it well in terms of where the ontologies need to go it's there's not going to be a one-size-fits-all it's critical in terms of the ontologies how they're applied into the labeling for the training data but in terms of the broader ontologies I think the IC has realized that they're not going to have the ontology that everybody fits we're going to try to figure out how to federate those ontologies together so that when one says tank the other says well that that's just a moving vehicle or or hey that's a that's a target right so I think I think there's a lot of work and the natural language processing is going to be a key enabler for helping link those things together automatically I think where it really matters is making sure that when you're training the algorithm you're very clear on the ontologies that are going into that labeled training data set I just added one other thing one of the interesting things that we're learning is that even in areas where we thought humans agreed more and more research and in not just terms but for example we all assume we all know what the color red looks like and yet we now are learning that we each experience and understand red in fundamentally different ways not just in how we visually process it but even how it evokes different emotional markers within us and so you know just telling the machine we'll label it red has very different consequences if we don't understand those those differences let's give someone else a chance any other questions yeah right there I'm Colonel Shorichika Romanian exchange officer to US Stockholm I'm happy that the AI is helping me transforming the information into intelligence but from the commander perspective think that intelligence is part of my IPB the IPB is part of my MDMP at the end of MDMP the military commander is choosing the best course of action from that course of action depend not the success of my mission also the life of my soldiers but let's put not in so dramatic way think about the MDMP as a part science and part art the art of war think all the military men in this room they know what is it the question is how do you teach a machine to learn the art of war if it's possible thank you so I think you're on the mark there in terms of at the end of the day in the military it's a human commander making that decision to commit his forces and he's accountable for that and that's going to be a very difficult thing to replace with a machine and maybe we shouldn't right I think if you break it down in the way you were describing it you know the step one is how can I get the best situational awareness possible right and AI can enable me to do that and as long as I trust what it's showing me where the targets are where the enemy is what's going on what the different activities are if it's giving me better situational awareness than I have today that's a good thing right it's never going to be perfect there's always going to be inherent risk that I'm either being deceived or maybe my algorithm tripped up here there's always going to be some uncertainty and that's why that human commander is so important to make that that decision based on the information he has this is the course of action but I think there's going to be a lot of work on how to give him alternative courses of action and that's where the modeling and simulation and wargaming come in where very rapidly the computer can play out multiple scenarios of force against force and then give a recommendation but I still think it's up to the training that goes into that commander and the experience they've gained over their career to then make the call and commit the troops can I follow up on this we'll bring all of you in how will we weigh the recommended courses of action whether it is you should go this route versus that route because you were less likely to face casualties or detection to courses of action in terms of this is the route of traffic you should go how will we legally weight that in terms of it's fine if everything works out well and we've said okay well we'll let the human decide the human will always be in charge how will we deal with situations where it gives a recommended courses of action to him he follows it but he was in charge he made the decision but the computer told him to do so and bad consequences happen what will how will our legal system be at military law or civilian law deal with that I don't think we know yet and and that question comes up a lot in the private sector too in terms of how do you program automated vehicles to turn one way to avoid an obstacle versus another way to avoid an obstacle it's but that that whole question goes back to this question of what are the circumstances in which you have to have a person making the decisions and what what are circumstances in which you can seed that or delegate that to to an automated system in some way and that's a I think we're developing the sort of both the norms and morality around that and and what the actual regulatory and legal structure will be around that but I mean there's a reason why ethics counselor is predicted to be one of the types of roles that big corporations will have in the next 20 years or so it's for exactly how to deal with these questions as we deal them out and I think it's an interesting element is that the the tolerance for machines making mistakes in our society is pretty low so car crashes they happen all the time but you hear about it when a self-driving car strikes and kills a pedestrian so I think that's going to be huge element at play but I think people also have a pretty good understanding of sort of letting humans do what humans do best and letting machines do what machines do best so if we have delegated authority for certain actions that can be made at machine speed especially on the battlefield then within those frameworks we can sort of operate pretty well and you know frame legal regulations et cetera around it but in terms of back to your original question and AI helping commanders process information faster right so instead of there's a command and control element too so again let humans do what humans do best and let machines do what machines do best if you can make it faster then that's great but in terms of you know giving full authority to a machine I don't I don't think that's going to happen in America just yet but we we do need a posture for some of our adversaries potentially doing such things question over here Dennis William military fellow here at New America and about 10 or 15 years ago the military went through a doctrine of effects based operations and largely that was that boiled down to measures of performance and measures of effectiveness and today even though we still use those two types of measurements in many of our operational level plans we struggle because we may not have the measurements right we may not understand the environment very well and as we're talking about this and the decision making component of it will artificial intelligence potentially bring back a resurgence of this effects based decision making because in theory the data sets are defining the environment in a way that we wanted it to to define it 10 to 15 years ago we just didn't have the technology to pull it together yeah I think so and I think fundamentally you can take that data now and do a lot of things with it in terms of measuring a lot of it's how you instrument right so you know you're collecting the right data and then it's going to be what's what do you really want to do with that data what measure of effectiveness or what outcome you're looking for I think the big thing will be that because you can compute and run simulations you know far faster and war game that out you can you can run enough opportunities through to see what the potential outcomes are and start trying to you know arrive at some conclusion of what might be a better outcome I don't know that you're ever going to get to the single answer that there's always a best answer for an effect I think especially on this topic with you know in a gray zone world it's not always going to be clear you know you're not always going to know you know you might have a desired effect but there could be second third order effects that you didn't anticipate especially when you get into cyber so I think that there's going to be a a lot of work that has to go into really thinking through all those different orders of effects and being able to to run those through and understand what's what's the real likelihood of a particular outcome and at the end of the day is that is that the course of action I want to take versus understanding you know that there's a perfect answer every time I was going to disagree with them initially but then you you rectified my my qualm with your answer and I think distinct to the special operations world too I mean you guys are always going to be a little different right you're going to you're going to operate where humans are really important in the in the human domain essentially to some degree and I think that as things get a lot murkier and we can't necessarily make sure you know this is the effect in a cut and dried way I I think that you guys are going to have a particular role to play that is not going to be based on clean effects along anymore let's get one last question hi Ben frankly from the McCain Institute Arizona State University advancements in change detection where is AI today in looking across a broad spectrum of operations economic operations movement of funds looking at activities in one area operation moving to another area of operation where the advances in change detection seems to me to be a problem but it's one area where AI could really help I'm not as knowledgeable on the financial side I mean when you get to machine vision there's a lot of work going on in automated change detection so not just identifying an object but identifying the patterns of objects identifying things that change over time where I may not have a video I may have an image every day or I may have you know an image every month and I'm identifying changes over time and measuring you know I might be measuring agricultural changes I might be measuring construction changes you know and a lot of that is now being automated I think one area where you see that that ties into financial and the construction business they fly a lot of drones now around construction sites and they can automatically fly those drones image it and basically feed that back into their financial models and decide are they on you know from afar is that construction project on schedule are there issues you know what's the risk to my investment and so I think that's where from a machine vision perspective you're seeing a lot of money being poured into how to extract information out of that imagery and feed that into different financial outcomes that the commercial sector is watching and I don't think I think one of the biggest problems is the architecture to support the fusing of all of this right so a lot of this is done in a siloed manner you know a few years ago NGA was working on ABI right that was their their biggest thing but we haven't found a way as DOD as intelligence community the intelligence community is a little better but we haven't really found a way to fuse all of that information together and and do so quickly and do so officially so I think that's a big ISR architecture problem that's a big IT environment problem and you have entities within the Pentagon who are attempting you know sort of to be the champions of of this change but again there's a lot of cultural barriers it's a big lumbering organization that's not conducive to the rapid innovation that made Facebook famous for moving fast and breaking things so I think that getting all of those elements together in one and then in order to extract valuable information from that we I don't think we're there yet but I think we are recognizing the problem and seeking to rectify it so I want to end with a question that tries to bring all of this together and applies also to the purpose of this event you've spoken about how this is an incredibly important technology it has a wide array of particularly useful applications but each of you also reinforced that no the human is still crucial in everything from the development of the use so given that how do we need to change education be it overall education to professional military and intelligence community education to have a cadre of people that can use this technology to its fullest whether it's out in the field whether it's as an analyst whether it's as a business or political leader as an acquisitions officer what do we need to change in human education to keep up with the AI revolution so I'll begin you are nodding the most which means you're rocking it the most so we're going to start from that end towards me okay so MIT is actually sort of leading the way I don't know if you've guys heard but they basically just announced they're having a college of artificial intelligence so while the first classes get to matriculate or anything like that so we don't know what their syllabi are going to look like we don't know exactly what they're teaching but I do not suspect that it will be an isolated program I would imagine that the graduates when they do eventually go through the courses will be exposed to a variety of the humanities so people can sort of retain the critical judgment and the critical thinking skills that are necessary to make the data good and the outputs actually work and have a lot more integrity so I think that they're not just going to have siloed education systems but there is an idea that we need to think very seriously about these problems going forward and MIT is sort of creating a blueprint for that so I would watch that closely and see if it works well I'm going to press you though because you have a very interesting perspective to bring here what is the type of training that you received as an intelligence analyst that you believe that that will be different from what was you know you were brought up under versus what might be needed say someone joining the field right now with all the potential of AI head yep so we had some I'm not going to let you get away but just say MIT is doing a great job it's awesome but that doesn't help the ICU yep I think I think technical acumen is sorely wanting I think that if not for some of the rotations that you were allowed to do at specific agencies with the agency that I was at then you know you could sit in rot in front of your computer only doing one going over one target set for years and years and years and some people did but I think that people need to be aware that they need to actively seek out a fluidity or a facility rather in other ints and be able to work with the tools in a way that is a lot quicker and just a lot more immediate than most people do you can inertia is a big thing I think that happens in the intelligence community so being constantly sort of made to seek out other opportunities to get those technical skills is absolutely crucial because you know in my last tour in Afghanistan I could pretty much count on one hand the number of analysts in the IC who had NSA training who had NGA training and all of the above because it's very easy to sort of get stuck looking at your guys and only focusing on your guys but you need to also have that ability to manipulate the tools and not just buttonology too like a basic understanding of sequel of Python that kind of thing that can help you help America in the future and not just within the intelligence community so I think many many more people and many many more kinds of occupations will need to be able to use very specific problem solving and entrepreneurial skills to develop specialized courses of their career within their occupation or within their role one of the trends that you see most in the future of work is that we need many many fewer big classes of people that are all trained the same way you don't need 300,000 accountants trained anymore you need 100 you need 100,000 different speciality specialty focuses of three types of accountants and for people to be able to identify the things that aren't being done and problem solve their role to be able to provide value added there isn't really a skill that our education system focuses on right now but I think it's going to be the defining characteristic of people who are successful within their their jobs even in traditional hierarchical context going forward so I guess I'll break it into two pieces one is the kind of the MIT piece in terms of even going back to the age of your kids in elementary school what do we need to do to start in STEM programs prepping them for understanding what it takes in terms of the math and sciences to to go into the AI field and how do you get more kids into that I mean if you go back to you know the shutter on China right part of the issue is they just have far more people going into those technical fields than we do and I think we've got to spur that on and with all the the money and activity going into AI there's just like cyber there's far more opportunities out there than there are people to fill those jobs today so I think it's it's how do you start in the schools get people motivated interested and educated enough that they're comfortable with the technology to go into college go into MIT and other schools and come out with those degrees to go fill that workforce so I think there's that side of it but the other side of it is kind of I think where Kristen was going in terms of how do you get people more comfortable using it right I mean we all walk around with our cell phones and realize there's a lot of AI sitting on your iPhone today that you don't even necessarily think about but it gets you around day to day it's how you communicate social media everything else so I think creating more and more opportunities for people to be comfortable with AI understanding its limitations as well as its opportunity and value and being able to smartly understand what's coming out of it for example if you look at Facebook and some of the things that happened with the the Russian involvement I mean how do you get people smart enough to understand that hey that's probably not real that that tweet or that comment probably was something that you know just doesn't logically make sense or or how do I ask the question of we'll verify that's true before I just believe it so I think you have to get people understanding that AI is going to be influencing us everywhere on a daily basis whether it's on our social media the way we drive to work and just being comfortable with that technology and understanding as users you know how do we make the most of that as a as a human in that loop versus letting the machine do all the work for us great answers all around please join me in a round of applause for us thank you thanks all time for a break we'll be back at three thirty for our final panel thanks good afternoon everybody no no person in the university environment would sit down with that little remark so thanks everybody for joining us we're here in the final session of today's event my name is Daniel Rothenberg I co-direct the center on the future of war with Peter Berger at the our center links our university Arizona State University with New America and we're thrilled to be cosponsors of this special operations policy forum event this is our second year doing this we have sort of a theme as you can tell this here on proxy war and really briefly I think one of the real issues is why is this term becoming gaining new salience and one issue is just sort of to what degree do we need a new language or do we need new definitions to make sense of our changing world so very briefly this is our panel the moderator is Lieutenant General Benjamin Franklin who is a professor of practice for leadership at ASU an advisor to President of ASU he's the former commanding general of Assistant Division Commander of 101st Airborne Division in Operation Racking Freedom and the commanding general of the 10th Mountain Division in Afghanistan where he was also the commanding general of Combined Task Force 76 in Operation Enduring Freedom Afghanistan he is a huge advocate for so many things that go on in ASU and particularly reaching helping to redefine what our university is in this changing educational environment Norman Rould served for 34 years in the Central Intelligence Agency managing various programs related to Iran and the Middle East Iran being the theme of this panel including as National Intelligence Manager for Iran at the office of the Director of National Intelligence from November 2008 until September 2017 Karim Shajapur is a senior fellow in the Carnegie Endowment for International Peace where he focuses on Iran and US foreign policy toward the Middle East he is a regular contributor to the Atlantic and writes for foreign affairs the New York Times the Economist and the Washington Post he's also an adjunct professor at Georgetown University's School for Foreign Service Candace Rondo is a colleague professor of practice at the School of Politics and Global Studies at Arizona State University she is previously the strategic advisor to the US Special Inspector General for Afghanistan Reconstruction a senior program officer at the US Institute of Peace an analyst for the International Crisis Group and she was South Asia Bureau Chief for the Washington Post where she was also part of a Pulitzer Prize winning team so we welcome all of you and Ben Floreshawes thank you Daniel thank you ladies and gentlemen for being with us this afternoon thanks to the panel for your time and expertise and I want to thank Arizona State University and New America Peter and Daniel in particular for hosting this and co-directing the program speaking on behalf of all of us we're honored to be with you so the purpose for our panel is to discuss Iranian strategic proxy or Iranian proxy strategy we're going to dive right in Karim what is the Iranian proxy strategy thanks for having me I'd say that since from 1979 the 1979 Iranian Revolution to the present there's basically been three pillars of Iranian foreign policy in the Middle East number one is opposition to the United States what they would call opposition to US hegemony wanting to rid the region of the United States number two has been the active opposition to Israel's existence number three this has increasingly become a pillar is the rivalry with Saudi Arabia and I would argue what's unique about Iran in the Middle East is it has kind of three tools which it can draw on to pursue these objectives number one is sectarianism which I suspect we're going to speak more about today but Shia sectarianism has been a powerful tool for Iran to employ in areas where it's demographically expedient right in places like Lebanon Iraq Bahrain to a lesser extent countries where there is an active or Shia minority like Yemen so sectarianism is a powerful tool which Iran employs number two is anti-imperialism and Iran doesn't only align itself with sectarian militias but it's also been a patron to groups like Hamas and Islamic jihad and also other anti-imperialist fellow travelers throughout the world they don't use them as proxies but Iran has friends in Pyongyang and Havana and Caracas and these types of places and the third tool which Iran can employ is Persian nationalism it's among the most nationalistic countries in the world certainly in the Middle East in which many countries in the region are less than 100 years old Iran can go back 2,500 years as a civilization so there's this very proud Persian nationalism they have and so I would argue you have some countries in the region that can employ maybe one or two tools but Iran is unique in that it has these three very powerful tools upon which you can draw to pursue these three objectives opposition to America Israel and Saudi Arabia the final thing I'll say before handing it over to my colleagues about what makes Iran so potent as a country which uses proxy warfare it's the fact that I think there's a widespread misperception even among people who do this for a living who believe that Iran is Shiite has its Shiite proxies like Hezbollah and Shia militias and Iraq and Yemen and elsewhere and a country like Saudi Arabia Sunni and it has its Sunni extremists it can support and in fact the huge huge asymmetrical advantage which Iran has over all of its rivals in particular its Sunni counterparts in the region is that almost all of the world's Shiite radicals are willing to work for the Islamic Republic of Iran if you're a Shiite radical from you know India westward you can find a place of employment if you want to go fight for the Islamic Republic of Iran and particularly in Syria these days and the other side of the spectrum Sunni radicals groups like ISIS and al-Qaeda for the most part want to overthrow governments like Saudi Arabia and so this is a huge asymmetrical advantage Iran has and that none of its rivals can really fight fire with fire because they don't Iran really kind of can own these Shia radicals Sunni countries like the Saudi Arabia but out to some extent they periodically can rent Sunni radicals they don't own them and at the end of the day the spread of that type of Sunni radical ideology poses much more of an existential threat to country like Saudi Arabia than it does to Iran thank you Norman what do you think their strategy is do you agree or other elements of the strategy well first I'd like to thank New America and Arizona State University for this opportunity to speak before such an August crowd with such stellar colleagues it is always full hearty to disagree with cream Sergeant Porte so I will begin by by burnishing my credentials by saying I fully agree with my brother Karim I will answer this a little differently though I think when we talk about strategy with Iran in the region it implies that there is some large plan that they began in 2003 and then that plan has rolled forward I don't think that's the case I think one of the great myths is that the U.S. invasion of Iraq allowed Iran to do what it's done in the region I wouldn't put it that way I think it's during the U.S. invasion of Iraq we did not stop Iran from doing what it subsequently did found it could get away with learn how to do better build on past experiences which Qasem Soleimani had employed since 1979 when he worked Kurds in the north since that time Iran has pursued an incremental approach wherever it has gone in the region for Iran to succeed in the region it requires four capacities and this is both a limiting factor on Iran's activities but it's also an enabling factor it requires chaos state that has failed or failing it requires beleaguered Shia who have no one else to whom to turn it requires a logistics pipeline and the extent of that pipeline defines the nature of the program on the ground and finally requires the absence of a third party adult on the field so to say what their strategy was in the region you could say to sustain allies to bleed adversaries to develop new opportunities that's been achieved but to say that they knew that going into Syria that they would be able to do that in turn I'm not sure is the case I think in 2014 2014 or 15 before the Russians entered the conflict I think it's questionable as to whether Iran's strategy was working but once they developed an air proxy through the Russians things shifted in their direction they've been successful but it has this program has limits I think right now their view is that as long as they have achieved these goals and there is no opposition in any area they have crossed many red lines they will continue to push more red lines in the future okay thanks Candace well I mean I don't think I'll say anything terribly different from Karim or Norm or what was said earlier on our panel in the morning I mean deterrence is destabilization and destabilization is deterrence in this instance and so there's a obviously a heavy investment in locally grounded forces sometimes migrant forces in the case of Syria in particular we've seen a very heavy investment of course in Shia Hazaras from primarily from Afghanistan or at the border regions and there are some historical links there that are really important that I think Norm has just sort of alluded to and Karim as well which is that you know Iran has this distinct advantage in having a very thin purely Persian Shia majority but actually a very large and very diverse minority of Kurds Hazaras etc etc that they can tap into that aren't integrated fully into Iranian society for lots of reasons and that are useful conduits for advancing armed conflict in places like Syria Yemen and elsewhere that's the armed group side and we can talk a little bit more about kind of I think the various brigades that are operating in Syria in particular I think they're a really interesting study actually in the kind of continuation of this investment in revolutionary ideology sectarian ideology and identity as a means of expanding the deterrent force in the region it's very critical but the other side I think that we don't talk about very often is the sort of soft power side which is the number of television stations that are owned and run by Iranians in places like Afghanistan Bahrain and elsewhere this is really actually I think often underestimated is the sociological and social impact of local media that is in either the Persian language or speaks to the Persian identity or speaks to the Shia identity this is true also in social media and there I think we've seen some very interesting developments vis-à-vis Saudi Arabia in the last couple of years this summer we had a bunch of protests as we all know in Iran particularly in the east again a trouble spot because of kind of the ethnic makeup there and we saw online you know battles between bots and the deployment of Saudi bots and Iranian bots facing off against each other messaging around identity around you know events on the ground I think we'll see a lot more of that and I think the big piece of the strategy that we have to talk about in the future okay so stay in the kind of the strategy level for a minute they've got proxy forces deployed four plus decades in Lebanon Syria Iraq and Yemen most recently so is this a coordinated effort by someone at the governmental level of Iran overall or is these just independent activities in different areas to disrupt and to advance nationalism and counter imperialism and counter the West I mean coordinated or not coordinated this strategy go ahead this is a whole of government effort it's obviously authorized by the supreme leader of Iran it's led by the Quds force which has been transformed since 2011 it's something more approximate to a combatant command but the elements of Iran's military which have been involved in this effort include their militia their navy their aerospace forces their ground forces just in small numbers Iran has demonstrated a capacity which is seldom seen in Middle Eastern conflicts and that is self-control it is willing to put serious skin in the game it has lost approximately according to public reports more than 2000 personnel more than a dozen generals it keeps fighting but it keeps its forces small its focus remains on developing proxy forces that are willing to do what Iran needs but the level of direction is not always precise one of the difficult words that is often thrown around is proxy and if you want to have analysts or think tanks work themselves into a frazzle ask them the definition between the difference between proxy surrogate and partner and come back four days later and see who's still moving on the floor but Iran has created a transnational Shia militancy which is capable of fighting on disconnected battle spaces against different foes simultaneously with different degrees of discipline effectiveness and capacity and it has done this again using this whole government effort by providing a cookie cutter approach of capacities which begin with trainers ground training and finally ending up with advanced missile technology or UAV technology I'd like to do a footnote on Iran's foreign ministry which is generally irrelevant on regional issues Javed Zarif has only one goal and that is to blunt and he does this effectively blunt Western or international pressure against Iran's activities while they build put facts on the ground and he does this effectively so again it's led from the top but it does involve multiple elements okay You know you made a comment on that they have the asymmetric advantage of all people from the region west of India can be in support is there a unity of effort is this controlled I mean from the top and dispersed in these different geographical areas where the fighting has taken place Well I think Norm put it well that at the top of the pyramid you have the supreme leader I argued that he wields a sword and he wields a shield his shield is Javed Zarif blunts as Norm said against international pressure sanctions his sword is Qasim Suleimani of the Quds Force and so you know Suleimani is one below the leader and then they use Hezbollah very effectively as their contractor and subcontractor it's easier for Arab Lebanese Shiites to train Arab Iraqis or Houthis in Yemen and the ways of warfare then it is to send Iranians there and increasingly you know as a society few Iranians want to go and spend time in places like Yemen or Syria they want to go to Dubai and so they subcontract that sure Hezbollah but it's become an effective efficient model and a lot of this goes back to something Norm said earlier which is who else wants to be on the ground in these places when you talk to members of Congress in the U.S. they'll say you know we want to do nation building at home we don't want to send our sons and daughters to these places likewise Iran's adversaries regional adversaries young Saudis and Qataris and Emiratis they don't want to go and inhabit these kinds of places so Iran is very good at exploiting recognizing power vacuums exploiting those power vacuums and they in contrast to their Gulf Arab rivals you know the quality of life for the average Iranian has never been the primary or even secondary consideration of the Iranian regime they're willing to spend and they've spent billions booing the Assad regime at a time of enormous you know economic difficulty within Iran so you know that that is I think what distinguishes Iran from its regional rivals Candace you were going to say well I was thinking you know one thing we didn't really talk about is the unique moment that we have in which you have Iran working cheek by jaw with Russian forces in Syria and the degree of coordination that was required particularly in Palmyra in 2015 Derazor even Aleppo between the proxy forces on the ground be it Hezbollah or some of the Hazara Niva Fatima you and Brigades and so forth with the Russians really illustrates certainly a heavy investment in reviving that relationship strengthening that relationship taking advantage in some ways of the log jam that we have now in the Security Council in the UN very critical and of course that plays well for Russia on a lot of different levels we see with the Astana process I mean the U.S. is really nowhere to be seen the EU has only just begun to kind of get into the conversation as we saw in Turkey late last month in Istanbul so I think that's a key part of the proxy kind of strategy it is not to say and I think it's important to make a distinction between allies, partners and proxies a difficult one to do but I will note in our glossy briefing paper that you can get out there we do try and make some sort of distinction and I think the distinction is with allies there are a set of rules of the road there are agreed terms there are rules of engagement there are often technical military technical agreements that kind of spell out what's supposed to happen with proxies the greatest advantage you have is not to spell those terms out that is the strategic advantage of using a proxy force is that they cannot be restrained or it appears that they cannot be restrained or that they're on their own that there is some sort of autonomousness in their actions and their motivations so that's another thing that I think Iran has played very well is this sort of again going back to the identity politics of these many many many disposable heroes I think that's right that most people would like to go to the Dubai Mall rather than go fight in Syria or elsewhere and that exploitative I think prowess is very illustrative of the strategy at the same time I think we shouldn't forget that it also demonstrates the limitations right of its conventional forces that in fact it could be a terrain issue but it is also an issue of just the ability to raise a conventional force in the same way that some of its local competitors might that is a constraint and it is something that is certainly opens a window to some sort of counter on our part I think on the U.S. part and that we have to be creative about that and really look at it in its eye okay so and taking up this point I want to ask the panel the question is an earlier panel today said that when the West looks at employing proxy forces there's this danger of loss of control that you can't control the direction you're going in so you laid out a clear strategy of disruption countering imperialism countering the West advancing nationalism so are these forces somehow you're arguing they've got autonomy and but is there some level of control and direction of these proxy forces in these four geographical areas to move towards an ultimate outcome that that the Iranians want to have in the region or is it just they've employed these forces and they're letting them go what's your what's your assessment of of a level of control is this earlier assertion that you can't control them a good assertion or not they're they're certainly under control but the word control is the next phase if any of the analysts or think tankers have survived after discussing these leaves today saying you know I'm with you thanks to come so they'll spend three or four days talking about what the word control means and and and frankly about the ability to have an outcome and a move in the direction now now that is actually very precise you've just saved a lot of think tanker lives because you will have based on very little limited information people talk about Iran's level of control over Lebanese Hezbollah which is far from absolute versus their control over the Houthis which is far more than people suspect but it's it's really more of a what is our goal at the end of the day and our goal is to fight the war in Syria well Lebanese Hezbollah is there you don't need to have a lot of control as you're integrating the Fatima Yuhn and the Iraqi elements together in a battle space Iran does not and that but with control they do have they have generals on the ground Afghanistan but also using their expeditionary forces which many of these circuits now I'll fill in to train Bahraini militants you have now Iraqis training Bahraini militants and Lebanese Hezbollah training Bahraini militants your goal is to create problems I always go back to the Al-Qaeda issue when Iran captured hundreds of Al-Qaeda operatives leaving Afghanistan in 2001 it permanently shifted the course of the war in terror and it was unpunished for it it captured hundreds of personnel coming from the Kandahar and a few from the Dorempic camps interviewed them fingerprinted them took their pictures and said go home and cause trouble now had they said we will put you all in a plane and send you back to the security forces of your various countries imagine what the war in terror would have been like that's a very different thing so doing the same thing with Houthis the only difference is you don't give someone a ballistic missile and tell them do whatever you want with it you may do that with claim wars a few types of UAVs a few types of small arms but once you get into serious UAVs anything requires training anything requires precision and cost and a footprint that requires control and that's when each of these elements you need to look for the technology and the nature of the fight I'm not sure this is an answer to your question but as Norm was speaking it occurred to me that another big advantage Iran has and the region vis-a-vis both its Sunni rivals and vis-a-vis the United States is that in most places where it wields power projects power it's simply trying to either disrupt the United States you know oppose Israel but it's not really in the business of trying to govern you know it's not its interest is not good governance and so if your goal is to simply you know sabotage U.S. efforts you know sabotage Saudi Arabia counter your region arrivals that's a lot easier than actually trying to build something positive and constructive and so you know I Norm has seen intel that I haven't seen I just say I having spent a year in Lebanon as a Fulbright scholar years ago when I was working on on based in Tehran and in Lebanon at that time 10 years ago there was a lot more question marks about Hezbollah's independence from Iran I would argue now Hezbollah is almost a branch of the Iranian Revolutionary Guards I don't not saying that Iran micromanages micromanages Hezbollah but I think they do macro manage them and I haven't seen any example of Hezbollah taking a decision which would in any way be contrary to Hezbollah I think there's a lot of things which are counter I would argue to Lebanese national interests I haven't seen them do anything that is counter to the interests of the Islamic Republic of Iran so we've established they've got a strategy they have a whole government approach they've got some top driven direction they've got forces on the ground achieving effect so this is a policy forum what's the West do about it what are the West's policies that can counter this work in the region the first thing it does is give up the fantasy regime change as the silver bullet I really I have to emphasize that there's no plan afterwards it never is there wasn't in Afghanistan there wasn't in Iraq there isn't one now for Iran so you have to try and you know picture what regime change in Iran would actually look like there was a peace in foreign affairs this month by Secretary Pompeo how to confront Iran and I was really struck by how many echoes of the sort of in fact many references to the Soviet Union right as the as the parallel though I think people forget exactly what happened with the Soviet Union I mean not only did the government collapse but the economy collapsed and actually sort of the psyche the national psyche of you know several hundreds several hundred million people so we have to really think about what it means to set as our strategic objective regime change in Iran behavior change is something quite different how we get there I'm going to leave to Karim and Norm I'll just skip the beat how do we counter it what should we do what's the right policy well if you again if if you kind of reverse engineer this and say that the places where Iran wields power and influence Syria Iraq Lebanon Yemen all countries embattled by other civil war very weak governments fractured societies places that do have strong governments you know less sectarian friction you see far less Iranian influence you know that is much easier said than done to rebuild fractured societies and it's a challenge when you're the United States as opposed to say Russia and China and that when you care about or you purport to care about democracy and human rights it's much more difficult to just go partner with an absolute dictator and say you know establish a hundred percent control and rid your country of any Iranian elements but but that's the bottom line is that if to echo something Norm said earlier it's not that Iran is so strong it's that a lot of the countries in the region are so weak and Iran is taking advantage of that so you'd have to take it country by country start with Lebanon I would argue you need to have a Lebanese military you know strong cohesive Lebanese military and in Iraq likewise I think you would need a strong Iraqi military to kind of counter the Shia militias and this is a strategy which is beyond something that the U.S. can achieve it would require also the help of our partners but it's to engage a lot of the Shia Arab countries and leadership and societies and think about ways in which they can put their Arab identity before you know for a long time now they've put the Shia identity before the Arab identity the Saudis have made an effort to reach out to some Iraqi Shia leaders like Muqtilesad to try to see if you know their Arab identities can come first you know I think that's a welcome strategy but it's not you know it's going it's actually going to take quite a lot of time and focus to change the way societies have been looking at themselves the last decades well we've been rebuilding countries uh no we're trying to for decades now we've been in Lebanon for a strong period of time 16 years or 17 years in Iraq and Afghanistan what what's the policy counters to you know here's one rebuild failed states but what do you think norms so what's policy rebuilding failed states requires a lot more than any one external country can give and I think it's almost fruitless to discuss that but with Iran you need to do first you need allies they have always responded to multilateral economic and diplomatic pressure but that doesn't mean you go through the UN there's no question that the Russians and to a lesser extent the Chinese have blocked every significant diplomatic pressure against the Iranians for for more than a decade and the idea of going to the UN Security Council and asking for an investigation I think I can name four or five that the Russians have consistently uh reluctantly tolerated allowed to undergo the undertaken uh declare invalid and and hope for another crisis so you need allies but I would not look at you in security council you must impose the cost on Iran and ultimately for policy makers you really only have three options you have diplomacy does anyone think talking to Zarif about ending missile strikes on Saudi Arabia will work this has got to be a yes or no question and not a 40 minute conversation then you have war is anyone in favor of sending 700,000 U.S. ground troops into with our colleagues from other partners into Iran okay you have a problem with that so now you've got sanctions and that hurts the Iranian people and in theory presses the government to say economic destability instability provokes a security political instability which is that is our price worth this price but you also have the option of precision military strikes and there is in the west you know different flavors of this you'll have people say we must do more diplomacy usually that is spoken by people who are sitting in again you know we're in a think tank next to a Starbucks where the greatest threat is how hot is the coffee but if you're in a country where you have 211 missiles falling on you you have a different perspective as to the pace of time and I should note that in those countries there are thousands of Americans and Europeans and other nationals and missiles don't choose targets so the consequence of if Iran is involved with the missile there's no one who doubts that Iranian supports proxies they are a tool of Iran nobody thinks Qatab Hezbollah is not an Iranian element if they attack an American to hit Qatab Hezbollah means Iran has a lot of Qatab Hezbollah needs to go through you need to punish Iran directly you really only have a couple of options heavy sanctions with international support for going UN Security Council because the Russians or Tomahawk missile strikes now there will be a day after but keep in mind that if you don't do that because your consequences are always what happens if we do this what happens if we don't what happens if we don't I would suggest people ask what was the last red line for which Iran was punished since 2011 think about that if you're in the Quds force right now sitting in front of the supreme leader and you say we would like to attack the next Saudi ambassador in Washington we would like to extend troops to Nigeria to help the Shia there what is the consequence the international community will impose upon the Quds force I'll close on one more point when you look at Lebanese Hezbollah in 1983 how do you all think we're doing so we're often asked is Lebanese Hezbollah question I put to some people is an Iranian proxy is it a member of Lebanon's polity is it a terrorist organization and the answer people give is all three which is the wrong answer because you've now just produced months of policy wrangling and no one is going to say we will stop the policy work the worst work etc etc we've done that for how many years how do you think we're going to do against Qatab Hezbollah or the Houthis when the Houthis are extending themselves into Djibouti at some point in the future did we have a chance to stop them now by just saying the problem is cutting off Iranian weapons and money what happens if you cut off Iranian weapons and money to Lebanese Hezbollah that's not always it sounds easier if I'm saying this easier than we all know it is I don't mean to do that I've been in these conversations for 35 years but simply to say diplomacy will get us there and no one wants a war well you need something and you must have your lines pay a price for their actions so that you provoke a debate among their senior most leadership to say do we want to do that again okay I want to follow up on a point to me you said they need stronger militaries so our special forces work on a document to buy with and for does that work that's what we've been using we've now expanded into advise and assist so we're twisting the names around we've had mission creep across conventional forces special forces and even special operations forces so does buy is buy with and for a viable doctrine and a viable policy to try to build the military power while you do maybe precision strikes and put other elements of national power pressure on Iran I would defer to Norm on that because he's had experience on the ground working with these militaries I think your problem is Iran has the largest ballistic missile force in the region Saudis spend more money than anyone else to reach it they're number three or four in the world for military expenditures as the Saudi military expenditures deterred Iran no so if you were to make the Saudis Saudi Air Force stronger is that going to deter Iran what about their cyber force one of the problems we face is in some ways we're we're not there aren't military rivals in essence looking for superiority their word rivals also a tough word intelligence community argues about a lot of words as you can tell we really do the the wars are more of paper rock and scissors we train militaries in the region to look like us we being the British or European partners etc etc the Iranians train militias to look like Iran so if I were to ask you to name with the definition of be carefully limited a Saudi proxy group in the region you cannot if I ask you to name all the Iranian proxy groups Shia and Sunni in one breath you cannot that's it I'm not sure how you get the Saudis to the point and I'm not even sure we should be in the business of creating more proxy groups in the region we're we want to be the good guys and we don't want to create groups that create another version of Hezbollah in these fragile countries instead we should strengthen state institutions and teach Iran that there is a price to be paid for its external adventurism and that lesson needs to be taught in Tehran and not going through the endless number of your Yemenis Iraqis Syrian Syrians would and this last question I'm going to ask I'm going to ask you all of your ready to ask questions but would would an information operation of any kind of sort of in the Cold War radio for your voice of America that that talked about American values Western values during the height of the Soviet time would any kind of information operation help in the region with getting people to move away from some of these extreme acts of violence and moving toward a more peaceful region would that work at all yes but pardon me yes but okay I mean I think you know when we think about information operation we look at for instance the state department's global engagement center which has struggled to find its footing for a number of years now and probably will continue to do so for a couple of reasons one governments have a difficult time being credible interlocutors in the public space today period anywhere in the world especially our government but two we're thinking of the problem often I think in the wrong way the problem is not messaging that will be received by the right group of Shias who will then suddenly rise up against the powers that be in Tehran if that's not going to happen a commitment to destabilizing messages that has some moral implications for the United States and also of course you know moves into a free speech area that I think we don't easily trample on even in today's climate one antidote really in this gray zone information situation that we have today truth and the big truth will be when these sanctions start to really pinch you will see the migration of Iranian money into offshore devices you will see the formation of new organized crime blocks you will see more coordination and cooperation between actors particularly state-run corporations that have a vested interest in a particular industry particularly oil that is where everybody needs to be watching and all the messaging needs to take place around the money that's really being stolen out of the pockets of average Iranians that's the real messaging campaign the investment has to be in not so much disinformation or better messaging but actually clarifying who's doing what to whom and what it's costing the Iranian people let me just accentuate that exact point which is one of the things I think we've done poorly as a government is simply make public all that Iran is spent in the region whether that's on Assad on Hezbollah on Houthis this is a source of real anger for many Iranians in fact over the years among the slogans people have been chanting when they take to the streets is forget about Syria think about us forget about Palestine think about us and I've always thought that the US government to the extent you know we have we have information facts about the billions spent subsidizing Assad the Houthis should make that information public second when it comes to you know an information campaign the most potent tool we have in our toolkit vis-a-vis Iran is Voice of America's Persian language service which is an absolute disaster and it has been a disaster for about 20 some years now and everyone knows it's we spent about 20 million dollars a year on this and the joke is that the Islamic Republic will be reformed quicker than Voice of America's Persian language service no one really has ownership over it and it's not journalists who are working there it is government bureaucrats in a building which resembles you know 1950s Soviet Union so I would say do with the Voice of America what BBC does you know there's a kind of a they have they actually staffed BBC Persian with journalists and this is a country Iran which has so many talented journalists who have had to flee the country who are desperately looking for work who would love to come and work in Washington and it's just been woefully managed this is a tool we have to potentially reach 40 million Iranians and we're just you know throwing it away great how about some questions please right here down front hi I'm Ken Mayercourt the missiles that the Houthis have been firing into Saudi Arabia supposedly were supplied by Iran but at least one military expert claims that their Russian scud missiles supplied to the Yemeni military ages ago that have been modified by the Yemenis if they are in fact supplied by Iran how does Iran get them to the Houthis there is absolutely no question whatsoever Iran has provided advanced missile technology to the Houthis period nothing else to be said how they get them there you just have to look at a map they've either dug a tunnel from Tehran into Sada they're coming through Oman or they're taking large missiles and technology and putting on boats through the maritime blockade and going into Houdaida and Salif if you think about the latter then it becomes all the more important for the Arab coalition to ensure that the Houdaida port and Salif ports are at least controlled sufficiently to ensure that only humanitarian supplies come through the country because Iran can easily set up transshipment points of some sort in East Africa there have been public reports that some activity has been going through Oman and some reports that perhaps that has been recently constrained but as you look at missiles it's important to think that people are more important than missiles when you're sending in a toy like that you need a technician to show you how to use it you need training you need care and maintenance and those Lebanese Hezbollah and Iranian personnel are again only entering three ways they're coming through that Jules Verne tunnel they're coming through Oman where they're coming through the maritime blockade Hi, Selena Hayes appreciate the comment on the messaging but I find with messaging you must have an understanding on how the audience that you're messaging to how they receive information how we receive information here in the US is obviously different how they receive information in Iran and I also would say that different groups within Iran probably receive messaging differently so while I think Voice of America revamping that is a great idea do we even understand how messaging is received there and I also refer to gender and generational so you know there's differences on how gender different men and women receive messaging and then different generations within a country receive messaging are we looking at different groups on how they would receive our messaging because I think we'd have to have a different narrative for different groups I'm not talking about messaging I'm talking about just news and information messaging sounds like we're going to send propaganda to them what's unique about Iran is that if you look at the Arab world media landscape there are hundreds of satellite television channels from Jazeera, Arabia, etc Iran has state-owned television they have these Los Angeles Exile channels they have BBC Persian which on a much lower budget than Voice of America has a much wider audience and then there's a one or two broadcasts from London which are very effective but VOA probably has the largest budget of all the overseas outlets and it has perhaps the least impact so I'm just talking about information now in terms of how people receive information when the protests in 2009 in Iran happened only one million Iranians had smartphones with mobile cameras the more recent protests January of this year now 48 million Iranians have these and the app which has become the most popular one is called Telegram so it's much more difficult for the regime to control information control communication but the reality is still a majority of people get their information from television and it's not these days if you produce good television then the clips are shared on Twitter and Facebook and Telegram and social media but in terms of what we can as the United States do I'm not even talking about messaging propaganda etc just create a first rate news program a news outlook I would monitor after the BBC or NPR and I think that can be the most effective thing we do can I add to that actually I just I want to sort of point out there is something to be said for your point is well taken which is that we don't know a lot about the audience I say you know the weeb here being the United States primarily I think there is still a great deal of sort of a huge deficit in our knowledge of what resonates what's meaningful what narratives are important to pay attention to but again you only know that I really you know I'm a former reporter what can I say I really believe that investing in you know just public institutions like the media is a really critical part of understanding what's going on in a place so to Karim's point there's no message really gendered or otherwise that's going to get through in any sort of meaningful way the message is there's something else other than the reality that you see on your on your state-run TV and that's that's a very powerful message we've seen that work in the past and we know it's power and we can't control the outcomes and I think we have to be prepared for that but we have seen it work and it would be foolish to think that investing in bots is going to do everything for us it's just not please hi Jeff Morley um the Trump administration was clearly depending on Saudi Arabia as a pillar in its Iran policy how will the Khashoggi affair affect the ability of the administration to advance its Iran policy I'm interested in hearing from all three speakers impact of Khashoggi and his murder in Turkey on our relationship with Saudi Arabia I think you have to first begin by saying what does Saudi Arabia do in the Iran coalition if you want to use that phrase you have a lot of outlandish statements in the west Saudis are pushing the administration to war with Iran or Israel it's usually Israel or Saudi Arabia is urging a war with the administration with Iran and that's just goofy that's just not true the Saudi effort is even when you talk of the word rivalry which is a difficult phrase a difficult word because the Saudis have not attempted to control Damascus or Baghdad in the exception of maybe the Hashemites in hundreds of years and no one has controlled Yemen for a very long time so maybe there's a competition and maybe there's a sense of pushing back on Iran's influence in the region Saudi main involvement is to ensure that there is sufficient oil in world energy supplies to prevent any sanctions on Iran from hitting back on us and the Saudis have done an admirable job for all of those who predicted earlier this year that there would be $200 barrel oil I predicted $83 just in case you wrote it $83 right now is what I said so I'm off to the Saudis are talking about cutting oil production now oil predictions are pretty squarely but when people say we want to use the Saudis in our campaign against Iran what else do we want them to do? We want them to develop some inroads into Iraq to support the Sunnis maybe provide some funding for our operations in Syria but what do we really expect to see? That Saudi land forces are somehow in Basra or Saudi air forces are hitting the Beqaa? I don't really understand when people say what the Saudis are going to do in the coalition they have a limited important powerful role and they're playing it the Khashoggi issue makes it more difficult regarding Yemen and I think the danger there is if the Saudis if they're in a war they can't win, lose or endure and they leave peace may be a bad thing for everybody in one extent except for the 12 million Yemenis who desperately need food and medicine I get that if Iran is able to fly C-130s in or planes in to Sana'a airport the way they fly into Damascus or Beirut what does that mean to the Babel Mandab which has more than 15 percent the world's trade 4.6 million barrels of oil a day about 12 percent of the world's rice I think that's the number submarine cable with all the internet traffic from western Europe east to England to Asia what does that mean to have that on that on that body of water and I think I think you have to wonder where is the Saudi what are Saudi going to go in Yemen and how is our interpretation of the Khashoggi issue and the terrible humanitarian tragedy which our entire generation along with Syria and Libya deserves a stain how that will play out I apologize for taking so much time well a couple of things one at the beginning of the Trump administration it was clear that the this team believed that the Obama administration obviously tilted too closely to Iran and from a global perspective whether you go to China or Russia or even our European allies when they look at the Middle East it's much different than how we view the Middle East they view especially you know even European Union they would argue that Iran is actually one of the few stable actors in the region they would argue that it's a source of stability in Saudi Arabia has been fueling radicalism Iran has been opposing I'm not making that argument I'm arguing that's the perception and I think the view that the US was trying to push is the Saudi Arabia being you know a positive actor in the region fighting against radicalism and trying to reform itself into something positive was totally destroyed by Jamal's assassination and I knew Jamal and so I think it just makes it much more difficult for the United States to convince our allies and countries around the world that we should be pressuring and isolating Iran right now because I think far more countries around the world including many of our allies believe that Iran is actually much more upstanding regional actor than Saudi Arabia I mean there's no question that this opens the way for a Congress that has been teetering on the edge vis-a-vis the policy in Yemen to begin making a push to certainly limit the mission I think we can anticipate that that will happen to what degree we don't know but I think that I certainly expect some sort of curtailment I think it actually highlights once again this fundamental challenge that we've been wrestling with since 2001 which is just the moral credibility of the United States when it comes to human rights in general and the protection of journalists and the right to free speech has been completely eroded and you know we've allowed that to happen collectively we can't really point to the government it's all of us in this room outside there's just no excuse for the kind of abdication of our own moral conscience that leads to backing somebody like MBS who then goes and murders not just Jamal but you know several dozen others we now know that this is a concerted campaign on behalf of the regime what do we say about that do we keep investing are the arms deals that we're going to make are they really worth it at the end of the day our moral credibility is the shield it is the shield and yet we continue to throw it on the ground step on it allow others to step on it we have a real challenge here I think with the Khashoggi case in particular really highlights that for us but it's not the last time that we'll see this unfortunately so I wish I could say otherwise but I suppose the the silver lining is that we will see a change and a shift that I hope will result in some sort of cessation of violence in Yemen well thank you all very much for your attendance this afternoon thanks to the panel for an interesting discussion we'll watch for those key watch words that cause the movement of furniture and things thanks everybody have a great afternoon thank everybody here for for many of you came for the whole day so thank you we have a small reception outside we also want to thank Raytheon and Chris Boyle for you know basically supporting this event and thanks to all of you for coming and thanks to this last panel