Technologically Speaking with S&T’s New Under Secretary!
In this very special bonus episode, hosts John Verrico and Deepak Saini are joined by S&T’s new Under Secretary, Dr. Dimitri Kusnezov. He has previously served at the U.S. Department of Energy and National Nuclear Security Administration and is eager to now put his leadership and scientific background to use guiding S&T into its 20th year and beyond. Dimitri, as he prefers his colleagues call him, discusses everything from artificial intelligence and machine learning to quantum computing and 5G. He also touches upon why it’s so important that science has a seat at the table when it comes to making policy decisions. Listen until the end to catch him letting his nerd flag fly during a series of rapid fire questions.
Record date: October 19, 2022
Technologically Speaking Transcript: Let’s Get the Science Guys in There
Guest: Dimitri Kusnezov, Ph.D., Under Secretary for Science and Technology
Hosts: John Verrico, Chief of Media & Community Relations, and Dee Saini, Media Strategist
[00:00:00] John Verrico: Hello, I'm John Verrico with the Department of Homeland Security’s Science and Technology Directorate, or S&T, as we call it. I'm here with my co-host Dee Saini for a very special episode of Technologically Speaking, where we meet with the science and technology experts on the front lines of keeping America safe. We're thrilled to welcome as today's guest, Dr. Dimitri Kusnezov, who is the newly appointed Under Secretary for Science and Technology.
[00:00:28] Dimitri Kusnezov: It's great to be here. First, it's great to be on the podcast, but second, it's great to be finally confirmed and here at S&T and at DHS.
[00:00:36] Deepak Saini: So, it's my first time meeting you. And, Mr. Under Secretary, I just want to ask, how would you like to be referred to, especially as you're going around the Directorate and you're meeting new people.
[00:00:46] Dimitri: I always have had kind of a more, casual way of thinking about that. I've never been big on org charts or titles or these things, as an academic. First, I grew up in California.
[00:00:58] Deepak: Go bears!
[00:00:59] Dimitri: And yeah, go bears. And so, there was a certain informality in, in how I grew up. First name is fine. As, an academic it wasn't about title or it wasn't about anything other than ideas. Where are the ideas? Who has good ideas? And you find them with students, you find them with faculty, you find them just about anywhere. And so, you couldn't let titles or positions get in the way of trying to figure out what the most interesting things are and what the opportunities are. And so, going by Dimitri was always just fine. Then I didn't have to explain how to pronounce my last name, and it made life simple. I'm more informal when it comes to that.
[00:01:40] Deepak: You and I share that, you know, with unique names here or there.
[00:01:43] John: Oh, absolutely. with Verrico and people would destroy that too. And it, you would think it would be pretty, fairly easy to pronounce. You know, you mentioned something, Dimitri, that really struck home for me when you talked about good ideas being all over the place and coming from students, from academia, from professors. You don't have to have a title to have a good idea. The atmosphere here at S&T is we're all open to new ideas and fresh ideas, and we don't care about rank and structure so much as to just who's got the good idea. With that in mind, and knowing that we are pretty much an idea factory here, what does it look like for you as to how you were envisioning working with a group of researchers?
[00:02:25] Dimitri: I think science and technology and mission agencies are different in character than how you find them in agencies that don't have to deliver, that aren't operationally focused. In a mission agency, you're faced with far more problems than you know how to solve. I think broadly, societally we have more problems than we have solutions. And so I'm happy for progress against any problem we have from wherever it comes from. And the only way to try and get that kind of, flow of ideas is to try and foster an environment where people aren't constrained by imaginary lines, that I can't talk to you unless I talk to them first. Or I have to ask. I, I want ideas to flow. I want conversations to happen that kind of bring out the best of what people have to bring to the table. Because I don't know how we get ahead of the problems we're going to face in this changing world today without everyone bringing their A-game to the table. And it, it's gotta start with creating the right kind of environment so you're not limited in how you push your ideas and socialize them and test them. And not every idea is a great idea. Not every great idea is worth pursuing because it might not be timed right. There are many factors that impact, again, in an operational organization, what is doable and what's not. But, unless you go out there and try and open up the aperture, you're, you're gonna miss these things. This is a great place. Like you said, I've only been here a couple of weeks now, and so I'm really in a very strong absorb mode, but it's clear there are a lot of passionate people about the mission, about S&T, about trying to do the right thing in, in many different ways.
[00:04:15] John: What do you want people to know about S&T that they might not already know?
[00:04:23] Dimitri: It boils down to, really the mission, the broad mission statement of DHS, and protecting our values, protecting homeland, protecting our society. And when you think about everything out there that you worry about today, where technologies are taking us, where the politics around the world are taking us, the instabilities, the list of things that are probably on the short list of what you worry about, it's probably just growing. And it's probably different now than it was even a couple of years ago, and probably that was quite different than even a couple years before that. And so, there's a very dynamic space in which the things that could cause us harm, that could impact us as a nation keep changing. And the question is, how do you get ahead of that? If you're only responding to today's problem, you won't be ready for tomorrow's. And we can spend all our time looking at today's problem, but who's taking the long view? This is what S&T has to do. How do we get ahead of what's changing and how do we ensure that the best that this country has, the best entrepreneurs, the best science, best tools are readily available for the people that are taking all the risk and all the operational parts of DHS? So, it's a daunting challenge, but it's, I think it's vital.
[00:05:44] Deepak: What is it like for you to come here at a time just before we embark on the 20th anniversary of S&T, considering what caused us to be formed in the first place?
[00:05:57] Dimitri: A lot of great institutions were built out of crisis. Where I was before our National Laboratory system and the Energy Department and its predecessors were all from the Manhattan Project and the Second War. And so already, after almost 80 years these things are going strong and they've proven to be important engines for the nation, filling in gaps that aren't being done by industry because the problems that the country faces are often beyond market and don't have a natural home. They're, they need more than just academic work and knowledge to, to solve them. And so, their important roles that governments have to play. I hope it affords the opportunity for people to step back and take another long view of where we're headed. It is easy to get caught up in a lot of the noise in the system. There's a crisis every day at some level in every agency and it can consume you. And sometimes, marking anniversaries like a 20-year anniversary of DHS affords leadership time to reflect, that otherwise might be harder to get with the day-to-day challenges of, of the jobs. And so, I think these anniversaries are important in many ways.
[00:07:16] John: Dimitri, as we're talking about, kind of the importance of S&T and the important role that science plays, you know, your new role here is as the science advisor to the Secretary and to the Department. And formerly with DOE, there was also the same kind of a great responsibility to help kind of direct the science and then to help those people who are making policy decisions and real-time decisions understand how the science can be brought to bear and how they can inform actions and policies. And you wanna, can you elaborate on that a little bit on, on how science can inform policy?
[00:07:54] Dimitri: I think that's a great question. Whenever you have a, you know, I, I've seen it through the lens of crises, and there's always one. Any given year, there will be something that will draw you in. From Ebola and Fukushima and the Gulf Oil spill and, pick your challenge, the, the Columbia Shuttle disaster. There are so many of these are hard challenges where people are going to have to make decisions about something. And the question is, do you do that in the vacuum, or do you try and at least frame what the decision space could be? Because whenever there's a problem or a crisis, the immediate questions for those who are faced with making decisions are of the kind, you know, what happened?
[00:08:43] John: Right.
[00:08:44] Dimitri: And, and so you have to bridge that gap. You have to try and understand, okay, we can try and attack this question this way. I've been involved in a lot of crises over the years. With Fukushima, for example, and the question was should we evacuate U.S. citizens from Japan? The disaster just happened, we had models of wind patterns, we had models of the geography, we had a sense of what could happen as a source term of a nuclear event, which is not a full nuclear detonation. And you could then make estimates of, okay, this is the impact to Tokyo. And so, it was the middle of the night. We had about an hour to do some estimates and feed them into the meeting. And our sense was Tokyo was not at risk. But, there'd been things like this where you can draw science in, but you have to be in the room, because if you're not in the conversation, then you're not part of this. And it's not always a natural request from those making decisions. You get all your operational people there because they own a lot of risk, but you don't say, hey, let's get the science guys in there. That's often not the mentality I found. So, finding the right way to inject yourself into the right conversations.
[00:09:58] John: So, you made an excellent point in talking about how science has to be at the table. So let me ask you this, how do you plan to ensure that science is at the table with the decision makers? And then we'll expand on that a little bit too. Not just in times of crisis, but also, during the everyday times between crises, so to speak, so that they know to plan for that the next time the fecal matter hits the air circulation device.
[00:10:27] Dimitri: Yeah. Well, well said, I guess. It's about being aware of what kinds of conversations are going on. There is no perfect answer to that because if you sit around and wait to be invited into the right conversations to offer input, you'll be sitting around waiting by the phone all day. You can't wait for the calls to come in. You have to really be constantly thinking about what's happening in the world around you. What are the tools and resources you have? What is the place where you can have the most impact? You have to be paying attention, and then you have to inject yourself. And it's entirely possible that the problems that you see are not identified yet, and you have to inject them and carve out time to discuss.
[00:11:14] John: How would you like to see how R&D impacts, not just during times of crisis, but the everyday people's lives?
[00:11:25] Dimitri: You know, if people had less things to worry about, then we've had an impact. It's about trying to figure out how science and technology can take the worry out of the everyday life of people here in this country. It's not a, a single path. I worry a lot about emerging technologies. I worry about things coming down the line. I've spent a lot of time thinking about the convergence of technologies and what that would mean in terms of bad things happening. There is a lot of interesting technology happening and with all the technology development, the thresholds for use of those are coming down. Anyone can learn how to do almost anything these days.
[00:12:08] John: Excellent point.
[00:12:08] Dimitri: And the question is, what does that mean and what's it going to mean as, we go from 5G to 6G? When I think about communications these days, for example, 5G looked like, well I've got a five on my phone now, instead of a four or three, so I can look at the web and download content faster. The thing about 5G wasn't just that, what I saw. I wasn't the only one that saw it. What was important there, I should say, was the opening up of the spectrum, far more parts of the electromagnetic spectrum were opened up for communication. And so, as you think about smarter devices. These days, small startup companies with a hundred people design and build smart chips for all kinds of functionality. Your iPhone or whatever phone you have to be all kinds of interesting detectors from nuclear detectors to whatever you want. There's an app for that or something to add. And so, there is a ubiquity of sensing technologies that are being built in. There is broader communication. There are smarter systems that can learn on their own based on data that's fed in.
[00:13:18] Dimitri: How do we do triage on the kinds of technologies we were faced with? We're in the position where we can shape the development of what's coming. If we have a little bit of foresight of what is the new spectrum of things we should be worried about. I think there are things in these systems of systems that are coming that we have yet to really bracket and identify. But it doesn't matter because it's coming anyway. And so, there's an urgency in trying to figure out, how do we get our head around so the world we're heading into is safer and more convenient and allows us more time to do things we want than to be worried about every next step we're going to take.
[00:14:00] Deepak: Dimitri, you laid out the tech landscape so beautifully. At S&T, tech is a part of our vernacular so I feel like AI, ML are buzzwords that people have become well accustomed to, even though that's also a huge part of our portfolio. But now, folks are starting to hear about quantum computing. In terms of the general public, outside of DHS, having studied at Berkeley and then Princeton and then having touchpoints at Yale and your former work at DOE. from where you started to where you're at now, are you fascinated by how far we've come up to technologies like this?
[00:14:36] Dimitri: Yes. and no, it is moving at a remarkable pace. I would say, especially with respect to machine learning these days, largely because of the commercial pull and the private capital going into commercializing everyone's little bit of data. Whatever it is, everyone is recognized this has value. But it's very narrow in terms of its impact and what it means. It's focused on specific kinds of things, not always what the government needs or what the government worries about. But I think there are still very hard problems in all those spaces in quantum, in, in machine learning, in all of these spaces that have yet to be surmounted. I distinguish those in that AI is a broader topic that includes machine learning as kind of an approach to trying to understand what is in data. What do get data from? What are you sensing? What are you doing with those streams? Whether it's sensors in your smart vehicle or on your phone or in a city. So, one end of the spectrum, you have the sources of the data. Once you get data, use machine learning methods, to, to figure out, well, what's in it.
[00:15:48] Deepak: That's really interesting.
[00:15:49] Dimitri: And, you might find things that are real. You might find things that are imaginary. It's still a bit of a black art in terms of how that works. It's an engineering tool as opposed to a science. Once you learn something, it's an entirely separate issue. What do you do with it? Is it actionable? Can you make decisions based on it? That is not lived within machine learning. And how you do that is, is really a wide-open question. If you're trying to make decisions from complex sets of data, there isn't any machine learning algorithm that will tell you how well you should understand the confidence of any prediction that comes out of it. Then there's the autonomy part of it.
[00:16:30] Dimitri: There's the human, AI interaction, and so AI is this broad umbrella of smarter systems. Whereas, machine learning is a piece of it that is just focused on how you're getting an understanding of the content of rich data sources. So I, I have watched that evolve and there's a lot of noise there, but most of the noise in that has been people applying it to new data sets. Your question was, am I surprised about the pace? Not really, because most of people applying the same kinds of methods to different or novel data sets, and so it's kind of more of the same. But it touches the public more broadly than just science would. And I think the public's willingness to adopt it, I think is impressive.
[00:17:17] Dimitri: Machine learning has been developing, at a rapid pace, but there, it's really plagued by some very hard fundamental questions of how useful ultimately it can be. And so, I don't see it as an end point. Quantum is still emergent. Quantum computing will have limited application for a lot of things we're worried about in the near term. But quantum sensors, quantum communication, quantum information. There are fields tied to quantum that open the door for doing things more securely that make sense worth looking at. But they're really engineering applications of a theory that was developed more than a hundred years ago.
[00:17:55] Deepak: Dimitri, as you're working your way around the Directorate, and I'm sure you're experiencing, we have a vast portfolio with a large mission. We have a lot of projects, a lot of research, a lot of programs, a lot of innovative tools and technology we're building to support our Components and our frontline responders. Are there any projects or any programs right now that as you're working your way around S&T, that's piquing your interest or you're really excited about?
[00:18:21] Dimitri: I've only seen a glimpse of what some of the work is so far. I've had a lot of high-level briefs of this is what we do and how we do it. And I did have the opportunity to go out to the Transportation Security Lab in Atlantic City to look at what are they thinking about, portal monitors and detectors of the future. And there are some interesting things there where you don't have to stop in the same way, where you simply walk through and things are detected in passing. And so, it opens the door for the experience that we face now in entering any building, any secure facility, almost anywhere from, baseball games and football games. I mean, there are great ideas out there people are building and it's, it is exciting to see what could be coming.
[00:19:06] Deepak: When it comes to protecting the homeland, do you think there are things that we work on that might be surprising to the general public?
[00:19:14] Dimitri: Certainly surprising for me, whether it's surprising for others, I don't know. But I think last week as I was talking to portfolio managers for first responders, the fact that, we work with up to about three or three and a half million first responders across the country was a bit of a striking figure. I know DHS has about 240,000 people, but this is more than 10 times more that are being impacted in many different ways by the kinds of things being done here at S&T. And that touches everyone's community from rural to city.
[00:19:50] Deepak: That's a really good point.
[00:19:52] John: And everybody always wonders of course, when a new leader comes into an organization, what's this dude gonna be like to work for? So how would you describe your leadership style? What kind of a leader are you, sir?
[00:20:06] Dimitri: It's sometimes hard to evaluate yourself so sometimes those are, questions better asked of, not only your immediate family also the people you've interacted with in the past. What I try to do is at least create an environment that is favorable to, to exchange of ideas. I need the environment to be open. It's sometimes hard to see if you're sitting alone in your office tending to email and Zoom calls, what all of these are. And so, people have to be comfortable in coming up without, again, the sense of, well, I've gotta go through this procedure to do that. You know that you can just stop by and say, Hey, and I mean it when the door is open. If I want it to be conducive to expression of ideas from anybody. Good ideas are anywhere. People have different lenses that they view the organization, that they view the responsibilities, that they understand the challenges. And everyone brings something different to the table. But talking to people I've worked with in the past is probably the best way to answer that question more honestly. I think we all have a sense of shared fate in whatever we do here. I sense it in the people here, in terms of how committed they are to the mission and to each other in terms of supporting the overall goals here.
[00:21:22] John: Dimitri, you are an academic and a scientist, so what advice would you give to students or pass on to others today who might want to pursue science careers and how they can be applied to national security?
[00:21:38] Dimitri: So first, that's a great question. I think science for, again, a mission agency, science for national security has a different reward structure than academic science. And so first, it's not for everybody. Academic science is based on citations, it's based on promotions, on how you push your own ideas and your work and demonstrate to others that the intellectual content that you have contributed exceeds the importance of other people's work. And so it is a space for egos, which is an essential driver of academic progress, but also pushes a lot of individualism. And that's important. That's how science has worked for centuries and will continue to work. When you think about the value structure on the national security side, from my point of view, whether you publish or not, I don't care. It doesn't matter. Question is, do you have ideas that can help impact the problems that we face as a nation? What do you bring to the table? And sometimes you can publish those and sometimes you can't. And so, the reward structure for people in this space is different. And that's not for everyone. You have to be satisfied that you made a difference rather than being satisfied that you have more citations in your colleague in the next office. And so, the metrics are different and for some, this is important to them. And they work hard in the national security world, and you don't know who they are, but you've seen the impacts of their work because the world around you is safer, and they know that. And so, if you're satisfied with that, there is a welcome place for great ideas and smart people. But you have to be willing to commit to that kind of outcome.
[00:23:24] John: And knowing you made a difference.
[00:23:26] Dimitri: Yeah, no, exactly. And it does make a difference.
[00:23:29] John: Thank you, Dimitri. Dee.
[00:23:31] Deepak: Dimitri, it's been great getting to know your leadership philosophy. So, as we steer into the future what do you envision your first 100 days, or of the next few months looking like?
[00:23:42] Dimitri: I'm trying to understand how everything connects. I'm trying to understand now how the science is done and how that connects. I've run science organizations in operational environments. With a changing world, what are the risks that are coming down the road and how do we place our bets if we don't have enough money to put against everything? What are the things that we should worry about the most? And I look at trying to understand the science through that lens. And so first a hundred days, well, I hope to get a sense of how this place works, how it connects and understand how to think about borders of the future, how to think about transportation of the future, how to think about homeland security of the future. What is it that we should be preparing for in tomorrow's world? For today's world we have a sense of what the triage is. We know the problem. It's still a hard problem. What's gonna happen in the years to come? And so, what I want to do in this first period, and hopefully less than a hundred days, is have a better sense of what it is that we're doing and who the performers are, how we engage the community, how we engage all the instruments we have in this country. And then, test people to think ahead. What is the future that we should be prepared for, that you worry about? And I think that's gotta be part of the vision that's crafted on where we're going. And so, it has to be a balance between hedging against future threats, but also working the tactical issues of delivering what people need right now.
[00:25:20] Deepak: I like how you put that. So now we're gonna have a little bit of fun.
We're gonna do some quick rapid-fire questions.
[00:25:27] Dimitri: Doesn't always work well with me, but I'll try.
[00:25:30] Deepak: Dimitri, in your opinion, what is the most important tech advancement of the 20th century?
[00:25:37] Dimitri: You know, maybe the microprocessor.
[00:25:39] Deepak: That's a good one. What would you name your own supercomputer?
[00:25:43] Dimitri: I wouldn't have a supercomputer. We need something beyond that.
[00:25:47] Deepak: Oh, okay. I like that answer. What piece of tech can you not live without?
[00:25:52] Dimitri: Microwave, probably. I have grandkids with me, couldn't live without it.
[00:25:57] Deepak: What's your favorite element on the periodic table?
[00:25:59] Dimitri: Oh, I like the ones at the end, Lawrencium, Nobelium. The ones that were named from some of the really imaginative people in the last century.
[00:26:09] Deepak: What's your favorite book?
[00:26:11] Dimitri: There are two of them, but no one would know the titles. One is, Lie Algebra and Some Special Functions. Another one is, it's more of a symbol. It's just called SL2R.
[00:26:22] Deepak: Hey, that's great. As long as you like them. That's less competition to see if they're available online.
[00:26:30] Dimitri: I tend to like math books.
[00:26:33] Deepak: Last one. Okay. Speaking of. Your kids, one day tell you they don't like math and science, how are you gonna respond?
[00:26:41] Dimitri: I take them to the whiteboard and explain to them why they should.
[00:26:46] Deepak: That's a great response.
[00:26:48] John: That is the best answer ever.
[00:26:50] Deepak: All right, Dr. Kusnezov, you are a busy man. We greatly appreciate your time. Welcome to S&T. It's been such a joy for me and John getting to know you, and we look forward to getting to work with you more.
[00:27:03] Dimitri: My pleasure. This is a great place to be. I'm looking forward to my time here and thank you again for welcoming me so nicely.
[00:27:10] Deepak: This has been Technologically Speaking, the official podcast of the DHS Science and Technology Directorate. To learn more about S&T and to find additional information about what you heard in this episode, visit us online at scitech.DHS.gov and follow us on social media at DHS SciTech. Thank you for listening.