School of War - The AI Arms Race: Can Our Intel Community Keep Up? With Anthony Vinci

Episode Date: June 5, 2026

Anthony Vinci—co-founder and CEO of Vico, an AI company that empowers judgment and analysis in finance and national security, and author of The Fourth Intelligence Revolution—joins School of War to discuss the technological evolution of spycraft. How does an intelligence officer operate in a world of rapidly advancing technology? What happens when machines begin to assist, or even replace, human judgment? And are we being spied on constantly? 02:33 - The Job of an Intelligence Officer 03:43 - Technology and Intelligence Work 05:58 - Living Under Surveillance 07:43 - AI as an Intelligence Analyst 09:37 - The Origins of American Intelligence 10:15 - Who Was Wild Bill Donovan? 12:37 - The Modern Intelligence Community 14:25 - The Evolution of Spy Technology 16:39 - The Intelligence Gap Before 9/11 18:58 - A Mossad Chief's Critique 23:10 - The Fourth Intelligence Revolution 25:02 - AI and Autonomy 27:49 - Are We All Being Spied On? 30:05 - China's Political Warfare 34:00 - The COVID Intelligence Failure 39:59 - The Dangers of AI Language Models 48:37 - TikTok and Surveillance Follow along on Instagram, X @schoolofwarpod, and YouTube @SchoolofWarPodcast Find more at The Free Press. Learn more about your ad choices. Visit megaphone.fm/adchoices

Visit BetMGM casino and check out the newest exclusive. The Price is Right Fortune Pick. BetMGM and GameSense remind you to play responsibly. 19 plus to wager. Ontario only. Please play responsibly. If you have questions or concerns about your gambling or someone close to you, please contact connects Ontario at 1-866-531-2,600 to speak to an advisor.

Free of charge. BetMGM operates pursuant to an operating agreement with Eye Gaming Ontario. We talk a lot here on School of War about how the digital revolution is transforming war fighting, but we haven't really zoomed in yet on how it's transforming the intelligence community. Today's conversation, a really interesting one, is going to do just that. We did record it before hostilities with Iran began, just for your situational awareness. Don't think that we're somehow ignoring all of that. Let's get into it.

If you read the headlines of Vietnam is to end in a state. We continue to ban. If you read the headlines about Israel, you're only getting a tiny slice of a long, complicated story without depth, context, or sometimes even the basic facts. I'm Norm Weissman, the host of unpacking Israeli history, the podcast that dives into the fascinating and sometimes controversial events and figures that have shaped Israel's past and present each week on unpacking Israeli history. I explore the layers of Israeli history, debates around the Palestinian and Israeli conflict, the cultural forces at play, drawing from a variety of sources and perspectives. So if you're looking for a nuanced, thought-provoking take on Israel, one that avoids the oversimplifications and political spin,

I think you'll really appreciate the show. Find unpacking Israel's history wherever you listen to your podcast or on YouTube. Hi, I'm Aaron McLean. Thanks for joining School of War. I am delighted to welcome to this show today. Anthony VINC, he is the co-founder and CEO of VECO company that does AI-backed forecasting. He's got a long background in the intelligence community served in the Central Intelligence Agency, the Defense Intelligence Agency. He was the chief technology officer, the first chief technology officer at the National Geospatial Intelligence Agency. He served all over the world to include in Iraq.

Anthony Vinci, thank you for joining the show. Thanks so much for having me. Appreciate it. So I want to talk to you about your very interesting book, which is called The Fourth Intelligence Revolution, the future of espionage and the battle to save America. And you go a little bit into biography in the book. So let's start there. Why does a nice young man join the DIA, which I think came first, right? What possessed you and moved you to that? Yeah, like a lot of people who signed up to become intelligence officers, I had somebody in my family who served.

And my father was a Marine and he served in Vietnam. And I always had that inculcated into me that when the country comes calling during war, you need to volunteer and sign up and go. And this was when 9-11 happened, like a lot of guys, I signed up and said, it's my turn and signed up. And that's how I ended up in it. What do you understand as an intelligence officer in an organization like DIA or when time came and you switched over to CIA? What is the job of an intelligence officer? Yeah, look, at the end of the day, there are decision makers in our country, the president being the number one, but there are generals. There are diplomats.

They have to make decisions on behalf of the whole country and decide the best course of action. and those people need the best possible information to make those decisions. And sometimes they need to get at information that's hard to come by. That's either classified in secret or that our adversaries, the people on the other side of the equation are purposely trying to make deceptive. Right. And what intelligence officers do is get that information and analyze that information so that they can get it to those decision makers so they can

make their best possible decision. A big theme of your book, obviously, is the role of technology in intelligence gathering. And we'll talk about that in a historical sense soon. When you were a young officer operating in places like Iraq, talk about the role of technology in your day-to-day work, just serving in the early 21st century in the post-9-11 war on terror period. You know, I was a case officer, so that's somebody who goes out and recruit sources. So I find people who have placed them in and access to information. I recruit them and I get them to give me that information to go find it, right? Maybe that's a person who has access to a terrorist organization or an insurgent cell

and I want to know what they're planning to do. And I have this person, I manage and handle that source and I have them go get that information. people have been doing that basically since civilization began, right? That's been going on for thousands of years. You could go back to ancient Rome and there were probably people running and managing sources. They were doing it in, we were doing it when we had our Revolutionary War. George Washington had sources, right? And most of the time, they were basically doing it in a pretty non-technological manner, right?

Like it's all about wits and, you know, finding people, being able to talk to people, potentially using some basic technology, like maybe you're signaling your source by, you know, hitting a wall with a piece of chalk to let them know you want to meet. And by the time I got into, you know, became an intelligence officer and went to Iraq, 90% of what I was doing was pretty non-technological. It was still the old way of doing it. I think if you took Nathan Hale and he traveled forward in time to join me at the farm, he basically would have understood what we're doing.

But we did use a little technology to keep ourselves more secure. Maybe we're using that to ensure using SIGN, for example, signals intelligence, to see if maybe there were some threats using remote sensing, G-WIN, you know, to use a satellite imagery or drone image to see if there are any threats that might affect me or my source. But it was still basically the same. What's happened since then is all of that is changing, because now we're living in a world where essentially everything is surveilled. And the business, we call this ubiquitous technical surveillance. So you can step outside and you'll see it.

There are cameras everywhere. We're walking around with cell phones everywhere, which are basically boxes full of sensors. that follow us around. And so now even case officers have to use technology to deal with that technology on a day-to-day basis. Yeah, you know, it's funny. It's just to draw the analogy over to the life of the infantryman. I was just having a conversation yesterday with a young guy whose service is more recent than mine. And, you know, on a battlefield that is much more technological, has access to more tools.

than I had access to. And of course, we have America's ground combat organizations today paying careful attention to what's going on in places like Ukraine where, you know, the softer architecture of everything, the, you know, ever-present surveillance, as you point out, is so dominant. And, you know, I don't think I'm that old. Like, it wasn't that long ago that I was working. Just like I'm going to do you the favor, Anthony, of saying, you're not that old. It wasn't long ago that you were working.

And I didn't have any of that stuff. Like 60-millimeter motors were like. In terms of organic assets, 60 millimeter mortars were about as sophisticated as it got. And that's only 15 years ago. And it's kind of amazing to reflect on the pace at which the digital revolution is transforming war and intelligence gathering. Absolutely. And it's a similar driver, right? It's artificial intelligence and autonomy have driven both sides.

And I write about this a bit in the book, you know, that there are a lot of parallels. For intelligence world, AI can now do a lot of what was done by intelligence officers, you know, like analysis. And so intelligence officers need to learn how to use that. And similarly, in the military side, you know, we always had aircraft, but now an aircraft can be essentially piloted by an autonomous AI system, right? Or it could be a ground combat vehicle of some sort. That's really been the big change. And you're right. It's happened within our lifetime.

Yeah. Right. The last war was one where more or less, it was similar to the Persian Gulf War or even Vietnam in many ways. The next war is going to be completely different. The battalion, if I recall correctly, as we were preparing to deploy my battalion, had one drone organic to it. And a company lost it. They just made a mistake and it basically ran out of batteries. It was flying around 29 poms.

And you would have think a Marine had been, you know, abandoned. in the desert, the investigations, the recriminations. I mean, it was a whole thing, capital W, capital T. And, you know, it was a $30,000 piece of equipment. Now, that's not nothing, especially for the Marine Corps. But, you know, today this stuff is proliferating down to the, you know, the squad level, if not below. And so let's tease out what this means in the intelligence world, because that's our subject

today and it's going to be really interesting. You, before we come, we'll go back to the past down, then we'll come back to the present after that. You talk about these four revolutions, these four phases in the history of the American intelligence community. And let's work through them. The first one is the foundations of the modern I see, which is the founding of the Office of Strategic Services, Wild Bill Donovan. Tell us a little bit about who that was, about why the OSS was started,

and what the nature of that first phase of intelligence work was. like I said we had intelligence even in the revolutionary war but it had always been mainly a military affair it was we might think about it more like reconnaissance in many ways and it was nascent and world war two the stand up of the OSS was where we really learned from the British in fact how to make an intelligence agency and to do this at a professional level at a world class level And that's what OSS was. You know, they learned from SIS, from MI6, their version, the British version of the CIA, had to stand that up. And while Bill Donovan was exactly the right guy to do this, he was, you know, he was a Medal of Honor winner from World War I, but then became a lawyer. And he kind of was running a source network, in essence, privately throughout the 1930s and collecting information and providing it back for his own. law practice, but it was ending up in government hands as well, including all the way to President Roosevelt at the time. Even though Donovan was a Republican, he was still on kind of the same side.

And so when Roosevelt said, hey, we need somebody to stand this up, Donovan was exactly the right guy at the right time. And he was a cowboy, right? Like, he was a guy who he snuck on to the D-Day landing. He was not supposed to go. He was that kind of guy. He was the kind of guy who was like, no, I'm going. My guys are there. I'm out. And he was that kind of leader. And he was a creative leader. And he was willing to take risks. He was willing to use technology.

The Navy, for example, had turned down the use of scuba gear. And that was invented by this guy, Christopher Lambert. Donovan saw it. He had him test this scuba gear in the pool of the Omnishore Hotel in Washington, D.C. He watched this guy swim underwater for an hour. and he's like, I want that. And he created the Maritime Unit, hired a bunch of surfers. Literally, he went to, he went to Santa Monica in L.A., found some surfers, said, look, guys,

I'm going to give you some scuba gear. And I want you to infiltrate beaches in Burma. And they're like, that's the kind of person he was. That's the, that is where our intelligence agencies came from. And that DNA, it's still there. And he built the structures, you know, that we have today. Essentially, the CIA is still. organized more or less in the same way. There's an operational unit, there's an analysis unit,

there's communications unit, and so forth, that all goes back to OSS. And so what then marks, OSS gets shut down, but then as the sort of the foundation of the modern national security state, as the academics would put it, you of course have the CIA and other agencies brought into being, what's the difference between that original phase of the OSS and the Cold War phase that begins shortly thereafter. Yeah, there's a saying that the OSS were called glorious amateurs, right? And they sort of were that. They were there for the war. They stood up. But they, you know, they were not long-term professional intelligence officers. And the majority of them went back to regular life after the war. And that's how America liked it, by the way. We did not like the idea of a

peacetime intelligence agency. President Truman Shurray. down the OSS because he said he didn't want there to be an American Gestapo. We always have been suspicious of intelligence. But what happened is all of a sudden we're going up into a Cold War and America is up against one of the greatest intelligence agencies in the world, which was the KGB. And sort of its precursors at the time and then the KGB eventually. And we needed our own. We needed professionals. We needed a well-funded professional organization with very, very strong counterintelligence. And we needed more than one agency. It couldn't just be done by one. We needed the CIA, which is set up in 1947, to focus on human intelligence, all source analysis,

and covert action. But then we realized there was new technologies like satellites, so we needed, we need agencies to focus on that. There was signals intelligence. So we stood up the NSA. We stood up DIA, and from there, intelligence proliferated and became very professionalized, very well-funded, and started to look like what it looks like today. And you do see the sort of early examples of sensors, right, of heavily technologically driven surveillance. You talk about the U-2 and things like that in the book. Talk about how that sort of that element comes under the scene.

There had always been this technological tint to intelligence. You know, like I said, Donovan tried, it was the first to use scuba. He was the OSS was the first to use a suitcase radio, which sounds big, but back then was that was a big deal, and it allowed you to report intelligence back from behind enemy lines, which is completely new in warfare. Well, that carried over into what I call the golden age of intelligence, right? The Cold War with the CIA and so forth. And America did what we do, and we went big on the use of.

technology for intelligence. And it was in a few areas. We were launching satellites by satellites. That's when they happened. We moved from aerial reconnaissance to space-based reconnaissance. That was very costly, very difficult to do, but it gave us a significant advantage. And we moved into using computers for analysis. That was also very new. And whole forms of intelligence that couldn't have existed without technology, like signals intelligence, spying on radio communications of our enemy or electronic intelligence, spying on the telemetry from, say, missiles from one of our adversaries. So these are new forms of intelligence that essentially had not existed before World War II. Or if they had, like it was in a very small scale.

And then the end of the Cold War, you know, we talk a lot about obviously the Department of Defense, Department of War now and the defense industrial base on the show. the end of the Cold War famously is a period of consolidation and sort of trouble on some level for the military and elements of industry behind it. I'm curious how that plays out in the intelligence community. And then, of course, obviously, we have to talk about the post-9-11 era of which you yourself are a veteran. So what's the next phase? Well, yeah, the end of the Cold War for the intelligence community was marked actually by a movement away from human intelligence. And it was the biggest drop off in case officers like myself that had ever happened, and a shrinking a budget, just like in the military. But really what I believe was the biggest deal is that this

massive kind of intelligence complex that we had built with the CIA and the NSA became so good at what it did, at compartmenting information and having very, very strong counterintelligence that all of a sudden this new enemy, this decentralized terrorist enemy in the form of al-Qaeda, was able to slip through the cracks created by this compartmentalization. So the CIA and the FBI weren't talking to each other about threats because they overlap domestic. Al-Qaeda was operating internationally and domestically. FBI was primarily in charge of domestic. CIA was primarily in charge of international.

They weren't talking to each other. so these al-Qaeda operatives were able to slip through the cracks. And ironically, it was because we were so good at how we used to do intelligence. So 9-11 happened. That was partially because of that. And that's sort of what the 9-11 commission report concluded. And so post-9-11, we realized that, again, we had to kind of revolutionize our intelligence community.

And we created what we call the intelligence community. And one of the things we realized is it can't just be a few. super secret intelligence agencies operating alone, removed from the rest of government. We needed this whole government approach. We needed to bring in other agencies like law enforcement, like the FBI and the Department of Justice, like Treasury, to look at, you know, counterterrorism threat financing, for example, right? And so they brought them all together in this intelligence community, created this Office of the

Director of National Intelligence to kind of oversee that. And we started adopting new technologies that primarily this time, instead of being developed by Lockheed Martin Skunkworks, for example, we're starting to be developed by commercial industry. So we're using cellular phone data, advertising data, commercial software. The CIA even created this just before 9-11 created this venture capital fund called Incutel to help to bring in that technology. So this was all new technology, new structure, that was the revolution. I recently had Yossi Cohen on the show, former director of the Mossad, talking about his career, which was heavily tied up with human intelligence, the running of sources. And then obviously driving pretty spectacular operational benefit from that from time to time. I was struck by how Frank he was, he was full of praise for the CIA.

for directors he had worked with, especially Pompeo, you know, very, very, very loved Americans in particular, loved the American intelligence community. That said, he was pretty critical of the agency's human intelligence capabilities compared to the Mossad. I'm sure there's a scale question there. I mean, Mossad's a much smaller organization, I think, than our world. But, and one, you know, it's funny. One thing he said about it was, I was pushing him on it. And he said, you know, one of the things that I don't quite get is, um, you know, you know, when we're recruiting agents, we the Mossad are recruiting agents, we're not, we're not Israelis in the sense that no one goes to some, you know, a person who's associated with the Iranian nuclear program and says, hello, I'm from the Mossad. Would you like to spy on your country for Israel on the reasonable assumption that that would be a pretty low probability play? So, you know, these officers are leveraging their language skills and elaborate legends and, you know, they they don't appear as.

Masah, they appear as, you know, French businessmen or whatever who have an interest in industrial information or whatever it's going to be. And he was basically said that as far as I'm aware, we're the only Western intelligence service that really operates like this. Now, I can't verify if that's true. I was struck actually by how detailed he got in his critique and the nature of his claim. And I just want to put it to you, you know, have we really recovered on the human Human Intelligence Collection Front since 9-11, you know, are we doing it at the level we need to be doing it today? Well, post-9-11, America got back into the business of human intelligence, right? We realized pretty quickly that, you know, Al-Qaeda was smart, and they were not particularly technological in nature,

and they were keeping a low profile, and the only way we were going to, you know, be able to kind of tackle that terrorist group and the insurgents, in Iraq and Afghanistan was to have human sources, human intelligence sources, right? And so we went deep and I was part of that movement. And I would say we got pretty good at it. It's true. We do things in a different way in America than Israel or then Russia, for example, right? And I think part of that is that we want to create, in essence, a reputation for America to know, look, if you're going to get recruited by somebody,

we're your guy. We're going to treat you well. We're going to take care of you. You can trust us, right? And in fact, the CIA just released, or there was a story in the news that the CIA just released a video targeting Chinese intelligence officers and put this up on social media. It's possible to access this video, essentially saying, you know, hey, you should be a little worried, a little disgruntled. Xi is beginning to, you know, remove your best officers. And that's bad for your country. And we can help with that, essentially, right?

We're trying to build trust. That's different. And so what can look to Cohen like a weakness to us in some ways is a strength, right? And I would say that goes for America in general. That is our approach to the world, which is that, look, we. can be your ally, you can trust us. Yeah, do we do some, some hard things sometimes? Are we willing to occasionally take out the, you know, the sword, as it were, when we have to defend ourselves? Yes, we do that. We're strong, but we're trustworthy. And that, that's a strength in my view,

and I think always will be. This episode is brought to you by Activia. You might already be eating yogurt, but not all yogurts are created equal. Activia contains over one billion probiotics per serving to survive and reach the gut alive. When it comes to gut health, Activia is the number one family doctor-recommended probiotic yogurt brand. Choose Activia, feel good from the inside out. Visitactivia.ca for more details. So that's really interesting, thanks.

Let's get to sort of the core of your argument, which is this question of what you call the fourth intelligence revolution. So if we go from the Donovan era to the golden age to the post-9-11 reconstitution and re-sertification, the creation of the modern intelligence community, your case is that actually we are at another inflection point as we speak. And it has something to do with, I don't exactly know how to phrase it, you'll phrase it better than me, but in my own words, it would be something

like the breadth of our ambition when it comes to conceiving of intelligence work and intelligence collection, that rather than something highly specialized and narrow, in fact, the nature of technology and its proliferation, the proliferation of sensors, of the ease of communication, kind of all these factors which you can tell us about, mean that we need to completely reconceptualize how to collect and organize information for strategic decision making. How does that actually work? Why is that the case? Why is that the case? is it different now than it was 10 years ago? Just help us understand the nature of this new revolution.

Yeah, absolutely. Look, these revolutions are driven by two things. One is a new enemy, new adversary. That was Nazi Germany. That was the Soviet Union. That was al-Qaeda and other terrorist groups. And this one is being driven by China fundamentally. And that China is a new, powerful, technologically advanced, economically.

advanced, militarily advanced, peer competitor to America. And that's created pressures that are transforming the agencies, transforming intelligence. The other transformer is technology. Just like satellites transformed intelligence during the Cold War, artificial intelligence and autonomy are transforming intelligence now. Together, those forces have led to three broad changes in intelligence that I think are revolutionary. And by revolution, and by revolution, I truly mean revolutionary. Like, that the intelligence world will look so different, 20 years from now. It will be unrecognizable to us, right?

Just like if you went back to 1982 and you asked Director Casey, who was director of the CIA at that time, and you said, hey, in 20-something years, the CIA is going to fund and operate a public venture capital company. that funds Silicon Valley companies, he would have looked at you like you were insane. Well, that's how different that last revolution was. This one's going to be that different. And it's going to be marked by these three things. One, the scope of intelligence is expanding radically. And instead of a focus on essentially political and military issues,

intelligence is going to focus on economic issues, science issues, technology issues, and so forth. Because that's now the nature of the competition with our adversary. with China. It's not just a political and military competition. It's an economic competition. It's a technological competition. It's an AI competition. So we need our intelligence community to essentially supply information about those issues back to our decision makers. The second change is due to this new technology of AI. AI is going to change everything about how intelligence is done. And just like we're all sort of watching in all of society and economic life right now. AI is sort of doing different jobs. And there's talk about how, you know, law work might be done by AI or, you know,

how radiology might be done by AI. Same thing is happening in intelligence. AI is going to perform intelligence analysis. It's going to operate sensors. We're going to have, you know, autonomous drones that are sensors. But more than that, that, we're going all the whole process, what's called the cycle of intelligence, the whole cycle will be done by AI systems. So I have an AI system that collects intelligence, feeds it to an AI system that analyzes it, which feeds it to one that disseminates it, that may send that intelligence to another AI system, say out to a drone. That's truly new and revolutionary, and begs a lot of questions about the role of people in intelligence. The third, and I would

would say most radical and revolutionary change is that now intelligence affects everybody. It used, most intelligence agencies used to essentially spy on and quote unquote important people. They would try to spy on a minister of defense or a general or someone maybe a member of parliament. But now we are all being spied on, right? and when you turn on the news and you see that yet another, you know, insurance company has been hacked or bank has been hacked, all that information is being collected oftentimes by people like the MSS, which is China's CIA or the GRU, which is Russia's DIA, for example. And so all of our information is being collected and then it's being used for information operations. And this is what was happening with TikTok, for example, where China was wanted to use that for information operations, ultimately to influence the public, not just a general or a politician, but the American public as a whole to bring about its geopolitical goals, right? And that's very new, and it drives the need for a whole new approach to intelligence that involves the entire kind of U.S. society.

Well, let's take each of those in turn because I think they're all important threads to pull on. On the first question, this issue of the need, as you put it, for the intelligence community to work in just new terrain, industrial terrain, you know, health pandemic, global health, et cetera. I guess part of my worry here, and maybe you're going to tell me that I'm too worried at what we'll see, is I think about a country like ours. fundamentally, you know, a liberal democracy. Political warfare, strategic thinking, you know, these are not, these are not our strong suits as a political culture. Maybe we have people thinking seriously like this who are full-time employees of the IC, but ultimately these folks answer to politicians and their appointees who come from, you know, American society. And my instinct is that senior leaders in China wake up every day thinking about ruthless political warfare with the United States being played for all the marbles. And probably see, by the way, military action is totally subordinate to that political warfare, which is occurring every day and is not even really specialized, intelligent, you know, specialized work for intelligence professionals.

It's just how we do statecraft. It's just how we, it's just how we advance the interests of China and the Chinese Communist Party. We just, it seems to me, we fundamentally don't think about things that way as Americans. And so I worry, I worry about our capacity as a country. Okay, so we go out and collect a bunch of information about Chinese industry. Some element of our intelligence community has some new bureau that it stands up that does good work on that. What do we do with that? Do we do we have a plan?

What is our plan exactly to exploit this information? How do you, does it seem like a reasonable concern to you? How do you think about that? No, it is a very reasonable concern because it's so new to us, right? In a communist system, in particular in a Maoist system, there's no distance between political warfare and economic issues. All aspects of the state are enlisted to support the direction of the party. The Chinese Communist Party is in control for their entire society, and they will use whatever means necessary. So they don't have this separation that we have in America, right?

And so it can be very hard for us to conceptualize that, where we have always had this strong separation. And it's obvious we do when you see how when Americans notice the breaking down of that separation, it makes us very uncomfortable. So if you look back at Cold War history and you see how Alan Dulles was involved when he was CIA director in what we're called kind of, you know, banana republic wars, right? And in Guatemala and so forth, right? And changing governments there that may have been involved, the, you know, dull fruit company and American company. We get very uncomfortable with that, right? And we kind of halted doing that because of that discomfort. So that's, we have a very, very strong separation. But for better or worse, we're in a world where we're up against an adversary that is using, that is combining

these economic tools and technological tools for political warfare. So to your point, we have to ask this question, how do we do it? And we have to do it in a very specific way. We have to do it in a way that does not remove what's special about America. We separate out our economy on purpose because it makes, we have a capitalist system that works best when it's efficient and is not government controlled. So how do we kind of walk that line? And that's the subtlety of the question, right? And that's, you know, I address some of that in the book, but we're seeing it actually happen in real life right now as, for example, the Trump administration begins to make investments directly in companies, for example, right? We're trying to figure out what is the right line to walk here.

on a support or supply chain, but at the same time, we don't want to ruin our capitalist system and just start picking winners willy-nilly who might fail, right? And we're trying to walk that line. Same with intelligence. We want to be able to support decision makers who are making economic decisions, like, hey, should I put this company on the entity list or create an export ban or fund an American company? So I'm worried about these threats. How do I make those decisions?

but at the same time, we don't want to, like, completely break the system or give unfair advantage to one American company over another. So we're trying to kind of walk that line. One of the more interesting bits of your argument was your discussion of the coronavirus pandemic and how that constituted intelligence failure. This is actually a subject close to my heart, which I have a little bit of experience with at the time. But you talked about November 2019 detection of, in effect, what becomes the pandemic. What happened in November of 2019? What did we see? And why was it an intelligence failure?

So, you know, a lot. This actually came out in the news in around January 2020. And I distinctly remember it that the president was briefed on this. the possibility of a pandemic. And he was warned, and we later learned, didn't, didn't take it. It's not that he didn't take it seriously. It's just that he didn't take the appropriate actions in retrospect, right? It's always easy to look back at a decision in retrospect.

And, you know, a lot of us forget there had been SARS and MERS and previous, you know, kind of large-scale epidemics, but not really a pandemic for 100 years. And the failure to me was one, it was a failure of imagination, which happened with Pearl Harbor. We failed to understand that the Japanese could attack us. It was a failure of imagination in 9-11. We failed to realize that there could be an enemy that could do that kind of attack and have that kind of damage in al-Qaeda. And so there is a failure of imagination, but a failure to truly comprehend the issue and to communicate that information to a decision maker in such a way that they can make the best decision, right? It's not just about saying, hey, we detected some virus and we're seeing some things happening in China.

It's really illustrating how that matters, why it matters. This is the job of an intelligence officer, right? And that was a failure. It's not a policy failure. It can be a policy failure. I should take that back. It can be a policy failure. They may just make the wrong decision.

But the decision maker has to comprehend it. And I believe that it was a failure because we did not treat things like a pandemic as a national security issue. And it was a national security issue. Look at all the national security implications of it. China ended up effectively taking over Hong Kong during the pandemic. They used it as a cover, in my view, to get their political issue. We had seven million Americans die, right? I think that is the latest number.

I might be off on that. But we had millions of Americans die. Like, these are, I believe, how we should mark what is a national security issue, are those repercussions. And we didn't treat it that way. And I believe we should treat these things that way going forward. I remember seeing at the time, and I can't remember if this was January or February, maybe even later, like March of 2020, there was this sort of mysterious story

leaked to NBC, I think it was NBC News, of a crash in cell phone traffic at the Wuhan Institute of Virology. Back in that sort of November, maybe even October timeframe, sorry, I don't have it in front of me, so I'm maybe getting some of the precise timing details wrong, as though the place had been evacuated

or, you know, mostly evacuated. So the normal cell phone activity you would see there had plummeted out. and I remember I remember seeing that and I had done in my role at the time as a Senate staffer I had done some back of the envelope math based on what was publicly available in terms of what was happening in Wuhan in December and sort of track back to like when would patient zero have been and I picked I had picked just through pure inference from publicly available information sort of mid late October and I was pretty proud of myself I was pretty proud of myself when that when that leak came out but this is all to your point it's like what a way to sort of zoom in on the Wuhan Institute of Virology is the fact that we apparently could monitor the cell phones that were wandering around that building, whether or not we were listening to them. We knew that they were there. And here, look, I think you bring up another good point here, which is why intelligence agencies?

Why did they have to help? Well, look, the CDC and other civilian agencies can do some of this work, right? They can track. But when you're dealing with authoritarian states that are actively trying to deceive, and in my view, it's very clear that China actively trying to. to deceive the world about this issue of COVID. They tried to cover it up. And the only way that you can collect information from authoritarian societies like that sometimes is to use intelligence tools, right? You have to have human sources or you have to have SIGAN sources and so forth.

And that's why you need intelligence agencies. But it doesn't mean you only need intelligence agencies, nor does it mean necessarily that they should even be in the lead. And so I think this is what is a really important issue to kind of parse out here is that there may be these national security issues like health care issues, medical issues like COVID and pandemics, or like technological issues or economic issues, where we need intelligence agencies to be part of the competition and part of the support. But they don't necessarily need to be in the lead. And in fact, they probably shouldn't in many cases. and where you may want to have other civilian agencies be in the lead with support from intelligence agencies or use commercial industry or nonprofits or this or you know what you could call public private partnership to be in the lead but that involves intelligence agencies at some level. So let's talk about AI for a minute. I've confessed to you in the past that I occasionally have a pessimistic streak when it comes to ways in which we might use.

AI, large language models in particular going forward. And I want you to respond, probably push back on what I'll say now, which is when you talk about autonomous drones, you know, acting as ISR platforms, I have no negative response to that whatsoever. Seems like a pretty good use of technology. Why do you need, you know, some very expensive, highly trained person to sit there with a joystick and a trailer somewhere when a computer can fly the relatively inexpensive platform for you? It just seems efficient.

It seems like it's going to improve productivity. It just seems like all around, like, what is really here to complain about? But when you talk about analysts using AI, I start to get more nervous. And I won't make the full case for you now. But I have this long spiel about how, you know, the integration of large language models into secondary and higher education is going to be a disaster. for our country because when you use the AI to do your writing, you're using the AI to do your thinking, that learning to write as a student is basically the same thing as learning to think. And so I think about professionals.

I think about young military officers, but in the context we're discussing today, intelligence analysts, using AI to sift through information, formulate their thoughts. I mean, there's a way in which when the data is so vast and it exceeds what any human mind could actually comprehend anyway, then yes, then we need new tools to sort through that data. And you see plenty of interesting applications of that today to include just necessary battlefoot applications. But when you get to the point of the analysts sort of formulating their own thoughts about things, I start to get nervous about them relying on AI. And I want you to tell me, I want you to make me feel better, Anthony. I want you to make me, I want you to, I want you to tell me

that I shouldn't worry because it worries me a little bit because I wonder at what point when the analysts, when they're coming to their final conclusions about this or that important subject, is sort of outsourcing that thinking to a highly automated process at what point we don't really have analysts anymore. Look, it doesn't just make you uncomfortable. It makes us all uncomfortable, right? As a society, as a global population, to watch this new technology, all of a sudden be able to do these things that we thought we were the only people, you know, the only entity

on the planet able to do. But we're all watching as these technologies get better and better at doing it. Now, I'm not arguing in any way that the technology is already there. The AI still has many, many problems from hallucinations in LLMs to non-determinism in LMs. But when you're talking about the future and I'm looking at this sort of 10-year, 20-year time frame of what intelligence agencies are going to be like, I do believe that within that time frame, these tools will become so powerful that they will be able to do what most analysts, for example, do today. And that includes, we could call it base level analysis, for example, change detection, look at an image and tell me what has changed. Have they moved the aircraft that's on the runway

at this military base in Siberia, right? That is a form of intelligence analysis. It's very, very important. It's the literal difference sometimes between nuclear war and not, right? Very, very important. But I think most of us now would admit that computer vision systems can do that.

We see that happen all the time and they can do it at a level that is approaching, and in my view, in some cases, better than what people can do. And actually, we may rely on it because it will be better, right? It doesn't have to sleep. It doesn't take a coffee break.

It's going to see it as soon as it happens, right? We want that. But then there's, so I think we're already approaching that world. Over the course of the next 10 or 20 years, maybe even five years, we're going to see higher and higher levels of intelligence analysis happen. And that is going to include what we in the business call all source analysis, where you're taking in lots of forms of intelligence, putting it together and coming. up with a view for a decision maker.

That might be a colonel trying to figure out whether to move his troops somewhere. That might be a president trying to figure out how to have a negotiation with another country, right? And these systems will initially be able to support the human analyst. And we, most of us are now starting to see how that happens. You know, like a lot of people, I'll sometimes run my emails through chat GPT to get it to sort of draft something for me and then I'll work on it, right? It doesn't mean I'm ready to let the chat GPT just send that email out, but I'll work with it. You know, some people call this kind of a centaur model, right, where it's like man and machine together. Well, we're going to do

more and more of that, but eventually this system will be able to do that as good or better than a person and in a measurable way. You know, you have to back test and see how it's done. So that begs the question, then what do we do? Right? and how do we even have analysts or what do analysts do? And I think that the job of an analyst is going to fundamentally change. It's going to move from doing that kind of analysis the way maybe a lawyer would no longer draft a contract, so to speak, to becoming a partner with their customer. And that customer might be a president. It might be a lowly case officer, such as Anthony, you know, 15 years ago or, you know, a lowly, you know, infantry lieutenant such as Aaron, you know, 15 years

ago, right? Like the customer won't just be a president anymore. It might be people in the junior and middle ranks and it will be those partners. And that's where I still believe the human part will matter because you have to build a relationship and build trust. And you need somebody at the end of the day, they operate the AI and to task it, what to do, and so forth. So I think that's sort of where this goes. And to your point, then how do we create those analysts? I think it will be more like an apprenticeship model. Rather than training, let me show you how to be an analyst. It will be, come to the meeting with me, watch what I do, learn what I do. It would be like guilds you know, back in medieval ages, right? Like, you're an apprentice. You just sit there and you watch and you

learn and you learn to become like this person. And those skills might not be how to analyze an image and pick the aircraft out of the, you know, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, with the customer, right? You're going to watch me do this with a colonel and we're going to get you up to speed so you can do it with a lieutenant and then move up over the course of a career. The third element of your account of the new world, which is this whole of society, crowdsourcing, call it what you like. This connects to what you do at your firm, too, I believe, in terms of geopolitical forecasting using AI. Help us understand with a little bit more clarity what you mean. We talk about something like citizen spies. I think that's going to make Americans uncomfortable.

You know, it's, again, back to my point about, you know, sort of liberal democracy and how we think about the world. It seems to me, Truman shutting down the OSS is a good example of this, actually. It seems to me we see spying as probably sort of necessary but very dirty work that has to be done in certain limited circumstances. But God forbid, it should be a kind of intelligence collection should be sort of an omnipresent aspect of our society. we shouldn't really want our society thinking about national security every day. That's a failure if our society is. And that seems fundamentally different than the attitude that, say, as Xi Jinping might have. What do you mean by this?

What do you do? I mean, what is today? What is the nature of the work that you do in terms of geopolitical forecasting and leveraging these new tools? Here's the unfortunate reality is we're living in a world where now there is a threat, right, where China and Russia do spy on. the American people and execute information operations against us. And the 2016 election interference by Russia happened, better or worse. And there is a huge amount of evidence that TikTok, for example, was used for information operations by China. And that's why the Trump administration and the Biden administration both agreed that we needed to remove that from China's

hand. So it's just an unfortunate reality. But it's not the first time. Right. The same thing happened after 9-11. We realized all of a sudden these enemies that had once been abroad were now here, right? And we had to learn to live with that. Or it happened with cybersecurity. 30 years ago, if you went to your average American and said, are you worried about cybersecurity? They would have been like, what is cyber mean? Right? Like, they had no idea. But now we know it's a threat. It's reality. There are hackers that hack into our computers and into our workplaces and so forth. And we as a country actually, have always not just been distressful of intelligence agencies, but government to some extent. And we've always taken it upon ourselves to defend ourselves as a community, right? And you go back to, look, Paul Revere was like just a guy, right? He was like a silversmith. He wanted to warn his local community about the British coming. And that goes, that impulse, which I believe is a strength of America, has continued. We've had neighbor Watches for decades in America. We have police.

We also have neighborhood watches. We know there's criminals and we don't necessarily want the police to be everywhere at once. We don't want a police state. So we watch ourselves, right? Well, and the same thing I believe is sort of happening now with intelligence. The reality is there's a threat. Do we want the CIA or the FBI coming and sitting on all of our laptops and phones and determining what information is true or false? Absolutely not. No way. Non-starter. Do we want,

you know, the director of operation at the CIA to determine if something should be censored? No way. So what do we then do? How do we deal with this threat? And I believe that the way to deal with this threat is to tap into this very American thing, which is that we can turn to ourselves, our community, to defend ourselves and working with the government, but to defend ourselves where it counts at home. And the way to do that is to learn some tactics, you could call them, some techniques of the intelligence officer, right? How do I, how does an intelligence officer think? Well, they never trust a single piece of information, right? They never look and say, well, I read this in a newspaper, therefore I trust it. No, they're always going to say, let me see in this other

source, whether they agree or not. Let me, let me triangulate this information. That's an example of an intelligence officer way of thinking that can be taught to all of us and will help make all of us more resilient when it comes to information operations, for example. And there's a series of these techniques that can be taught, just like we did for cybersecurity. It's not like all of us became NSA officers, right? We all just kind of learn some basic techniques, change your password, don't fall for fishing scams. We got trained in school. we got trained in our workplaces.

We even trained our children, you know. Our kids learned to change their passwords, even when they're very, very young. It didn't change how we were as a society. I think, in fact, it made us feel a little safer and a little more like we can control our own destiny. The same thing will happen with intelligence. Unfortunately, we have to realize that this is a real threat. It is what it is.

Nuclear weapons were a real threat. School shootings are a threat, right? These are real threats in life. It's unfortunate but real. We just have to learn how to deal with them to protect ourselves, and that's okay. And when we do that, we can strike this balance where we kind of maintain our rights and don't necessarily have the CIA and the FBI and so forth patrolling our information. They're doing what they do best, and we're protecting our own information.

Anthony Vinci, the book is called The Fourth Intelligence Revolution, the Future of Espionage, and the Battle to Save America. It has been a really interesting, albeit unsettling, conversation. and I'm super grateful to you making the time to come on the show. Thanks so much for having me. Really appreciate it.