The Future of Espionage & Intelligence Operations | Anthony Vinci | Ep. 388

an application into the website. And so so I did, and uh and and signed up. It is an immense amount of selection that you go through for these kinds

of jobs. The clearance, you know, where there's everything, and then lesh joint government and then when you when you want to go to training program to go to the farm midday, uh to you know, make sure personality type and you're able to handle yourself, you scrolling things like that, and UH made it through all that and UH when when went down south training.

understanding reading the history. The CIA obviously did and they sent Jawbreaker and the other teams out there, and so Rumsfeld, you know, you can kind of read said, you know, we need to stand up some some capability here, and they did.

d A is there for strategic military intelligence. You know, the these are these are missions where you know, and it involves counter terrorism, counter you know, insurgency operations of course, but it also involves you know, really traditional military intelligence like order a battle, military equipment, you know, those kinds

of things, preparing you know, preparation of the battlefield, all of that that's reporting up. It's it's not typically as tactical as you might get within the services, you know, where where you've got you know, you know, guys looking out just around their fobs and things like this, collecting

on the battlefield and so forth. But it's sort of at the more strategic level, as opposed to CIA, which is looking really broadly at American foreign policy, at geopolitics, at strategy, also at some you know, military issues, you know, especially when they're kind of the national level military issues.

And you know you're in you're in traditional counter terrorism counter insurgency operations as well. I mean you're you're looking to figure out where you know, insurgents are located, where terrorist groups are located. You're you're thinking about who's in who's in command of those terrorists are insurgence groups, and and and those sorts of military things. I think everybody was just chipping in on the same mission.

Like it's a very different type of intelligence I would think, I.

some way. You know, it was like I was like, yeah, that's in. I understood that it was important and somebody had to do it. But you know, myself and a lot of guys who signed up post ninety eleven were like, we're here to be in the game, and this is

where it is. And and you know what's interesting is that sort of started to change, you know, over time, and by the time you know, we'll get into it later, by the time I came back into the government later at NGA, that was sort of ending and gone and we'd moved back into this like these great power politics and you know, Cold War style operations again.

I'm gonna find some source who works at the embassy, you know, foreign embassy or maybe they're a you know, adversary intelligence officer, like a KGB guy, and I'm gonna try to recruit him and take people to dinner and you know, doing doing SDRs and you know, pretty cities in Europe, and that's obviously not what it was like during during you know, when you're in Baghdad or a place like this and where you know you have you're

bringing in. Now you're operating in a war zone and you have to protect yourself in the same exact way as as as as as a soldier would, as anybody would who's sort of outside the wire, right, So you're doing all of those things and you you have weapons, maybe have security outside of yourself. Usually typically a case officers operating alone in that job. Historically maybe maybe with

an interpreter, but typically alone. You learn the language yourself, and now maybe you need to operate with security or at least driver you're you're you're armed, because you need to defend yourself if there if a situation arises, you have to be in really strong communications because if something

bad happened, obviously you need to call for help. All those things are not what you picture when you picture Cold War operations, Like the last thing you want is like a radio or some piece of comms gear on you potentially because that could give away what you're doing. Right, and so you're trying to operate clandestinely, you know, in Vienna nineteen eighty six or something. But now in Iraq you're operating like a soldier in many ways, but you're trying to also be as clandestine as you can, to

stay off the radar, to stay low profile. That's just a really different environment. And it's interesting we had done that before. And when I was kind of in training, I met one of the OSS guys, the Office Strategic Services. These are like the OGS, you know, they were World War Two eras, pre CIA, pre pre special Forces, you know,

and and these guys were you know, they were badasses and they were operating behind enemy lines. There's a group called the Jedburghs, of which this guy I met was was one where they would send him behind the lines sabotage and so forth. And that was really kind of the last time we did we did that much, you know operation. I know in Vietnam they did some, but World War two is the last time where we did. You know, that's what the force was doing basically.

It's going to be fine, and uh, you know, thank god, I walked out, okay, and and and people I was working with it.

You know, you're you're just working all the time. You're you're you're just constantly running through the intelligence cycle like in real time because things are moving so fast, to like a more traditional environment in in a you know, typical world city where things slow down a lot and you start to think much more strategically long term things take a while, you know, and and for a lot of us, myself included, that that transition was hard. It was it was hard to kind of go back to that.

It didn't feel that sense of mission that you did that was just paramount when you're down range or when you're supporting somebody who is. But you know, it's two very different worlds.

diaper in your ass area looks like. So there are these problems that you know, different fits of different types of genes have, uh and the Perfect Gene strives to fix that, both the fit but also the mobility, the maneuverability. It has a stretchy fabric like they say it's like yoga ready, but you know I would say it's athletics ready.

You know, moving furniture, doing things around the house. You're not going to have that like super tight, like constricted fit that you feel like you're not gonna.

on on the subway if that's your thing. So the Perfect Gene is great for all kinds of different environments, whether it's doing handiwork around the house or on the job or athletics, and we really hope you guys will go and check them out. For a limited time, our listeners get fifteen percent off their first order plus free shipping at the Perfect Gene dot nyc. Or google the Perfect Gene and use the code House fifteen for fifteen

percent off. That's fifteen percent off for new customers at the Perfect Gene dot NYC with promo code house fifteen. After you purchase, they'll ask you where you heard about them. Please support our show and tell them we sent you.

Every ghost Bed mattress is made with premium materials, proven cooling technology, and their exclusive pro Core layer, a targeted support system that reinforces the center of the mattress where your body is the heaviest. It helps to keep your spine aligned and your back supported so you wake up ready to take on what's next. In ghost Beds, cooling materials don't just keep you from overheating, they help regulate

your temperature year round. Each mattress comes with one hundred and one night sleep trial and a twenty to twenty five year warranty, plus fast free shipping. Most orders arrive

in two to five days, so better rest is just around the corner right now. During ghost Beds Holidays sale, you can get twenty five percent off site wide for a limited time. Just go to ghostbed dot com slash house and use the promo code house at checkout. That's ghostbed dot com slash house promo code house. Upgrade your sleep with ghost bed, the makers of the coolest beds in the world. Some exclusions apply. See site for details.

And thank you ghost bed for sponsoring the show. Please check out the link down the description and support our sponsor. I mean, yeah, I don't need to ask where you were, but are you able to talk a little bit about how that functions because as I understand it, I suppose it depends on the size of the embassy. But there's some case officers and there's some analysts from the DIA.

collecting information from them. If you're a debrief or you're just debriefing say a friendly like an American person or something, and you're bringing that back and you're writing it up. At a at a literal level, you've got a you have a set of collection requirements, and you know what you need to ask about. You know, maybe one day it's how is this laser made? And the next day it is you know, where is shake so and so? And how do I find him and and And you're

you're asking those questions. You're you're you're taking notes, you're bringing that back, and you're writing it up. And then that that is the and going off to two analysts who are putting that information together ultimately combined with with other with other intelligence by all source analysts who are going to put all the pieces together and report it

up the chain. And like I said, you know, with a with d I, A or other military intelligence that might go to you know, to one of the services and to their intel unit, to like a two shop somewhere.

It might go directly. Some of this stuff, especially strategically, it's just going directly to generals, right, I mean, it's it goes right up the chain, but it's also going to feedback into the White House and so forth into the National Security Council because a lot of this stuff can be strategic and it can matter, you know, And that that was kind of a weird thing during post nine eleven, where things that seemed tactical, like where is this one person, uh, this one bad guy all of

a sudden are strategic, like it's like president level stuff, like they want to know where these guys are because you know that was the enemy. Then it's totally different today, you know, where you know, we're worried about you know, nuclear weapons again, or worried about you know, China's going to go to war.

were like gone for months at a time. But it was really only with this book where you know, with a book like this and you're you're telling your story that that all the agencies actually come in and take a look that I was able to say really anything. And so still remain a little bit cautious of where I where I draw the line, because you know, as as you know, you still want to protect those are still still in there doing doing the same job.

And when I exited that, that company, I you know, was looking to get back involved again, you know, but this time maybe just advise or something like that, and I joined. I was invited to join the advisory board at NGA, and the director at the time, Robert Cardillo, and the deputy director, Sue Gordon, were looking for somebody who knew intelligence but also knew the tech industry and new kind of Silicon Valley and really the cutting edge.

And today that's common, but you know, five ten years ago, it was actually pretty rare. There weren't that many people.

There was this real hard line like Silicon Valley people didn't want to touch national security, and national security people didn't really have a seat to do anything in Silicon Valley world. And I had happened to kind of get into that, and so I went on their advisory board and tried to help figure out how to use AI and how to use commercial data. And then they I said, I gave them kind of a strategy to execute around that, and they just called my bluff and said, why don't

you come in and execute that strategy. And so that's how I ended up going back in as a as a govy again.

can imagine it. Back in the Cold War, they would get a piece of film from a satellite and they would look at it with a magnifying glass and try to see where the Soviet bomber was. Right. That's updated now and it all goes through computers, but it's essentially the same thing, and they're trying to see, you know, is there is there a missile on the launch pad?

is a geospatial information tools. So you've got a map and you've got location and and you're appending information to that location. And geospatial intelligence is about figuring out where something is and and when when it was there or is there right and what it is and making intelligence tools out of that. So for example, you know, if there's uh, you know, thinking about like maybe there's a

missile on the launch pad, Well you've got it. You've got kind of a map of that area, and you're you could put and append the information about all the other facilities on there, who's there, how things changed over time, and create a geospatial product like that. And so that's

what that agency does. It's hugely important. Not a lot of people think about it, but it's one of the agencies that keeps everybody safe because for some of the biggest threats in the world, like unwarned launch from like a North Korea, really some of the only ways to get intelligence out of places like that is from satellites. You can't it's real hard to send a human source to North Korea, and it can be very hard to even get signals intelligence from there. So this is where

a lot of that information comes. And these analysts are coming there and keeping watch globally all the time to make sure we're not attacked, but also at the tactical level are producing targeting packages for conventional units, for special operation units and things like that that are necessary for kind of tactical operations as well.

that all comes together.

making maps. Those two groups and they successors, along with some other pieces, became n g A after I think around two thousand and three mm hmmm.

involved with requirements for future systems. So what sort of intelligence should we be able to get off of these satellites, because you can imagine the analysts want to have a say and what is the next generation of satellite is going to want to do. But they n g A then is not making the satellite themselves. That that's going

to go on to another agency. And then n g A is very very used to be very very involved and fully ran acquiring commercial satellite images and ran a big contract with Digital Globe and g o I for a long time, and then other satellite commercial satellite companies like Planet Labs and so forth. That that mission has also moved to n A ROW to actually do the acquisition.

N GA remains very involved with acquiring and developing in some places analytical systems like geospatial analytical systems and in particular computer vision and things like this, which is a form of you know, imagery analysis, right, And that's what I was very very involved with when I was at NGA.

do this, but it didn't work very well. And they were now around twenty fifteen, twenty sixteen, twenty seventeen, computer vision got much much better because they started using machine learning to do this and and I sort of realized that when I came into NGA, and and that was

really many ways my role was. Part of my role was to figure out how to do that, how to acquire those kinds of technologies because they were being developed in Silicon Valley in places like Stanford not it you know, in national labs, or or or by the military or

by the prime contractors. For the most part, how to acquire that stuff and how to bring it into the mission because you know, what was also happening is the amount of imagery was just skyrocketing, and we had more and more classified systems because they build up over time, and then all of a sudden, we had all these commercial satellites with all of their data, and now we're you can there's a limit to the number of analysts you can have in the world, right and how many

how many billets does does an agency have? And does the Congress want a fund? And so there is a there's a maximum capacity, and so how do you how

do you break through that? And how do you ensure that something doesn't slip through and that missile ends up on the launch pad and you don't just miss it, right, And so AI was a way to do that because you can identify objects in an image that you can tell if they've changed and so forth, and then automatically alert and that that was the goal, and that's what we're trying to do.

a position at like Hewert Packard or something like that. It's not something you commonly associate with an intelligence agency. Tell us about how that role came about and what it actually was for you.

wanted to do. He took you know, somebody was forward leaning and kind of a tech guy and pretty young, I mean it was thirty nine at the time, and said, not only do you have this mission, which is I want you to transform how we do intelligence, but here are all these resources, and now go and go and do it. And we built a lot of different offices

and organizations to do that. Over time. Then I transitioned into this Chief Technology Officer role, which was a piece of that which was really thinking about the big the

big technology strategies. You know, how do we you know, what kind of technologies do we need, whether to develop internally or to buy from externally, and how do we get them into the system and how do we incorporate them with existing technology systems that are typically designed by a contractor, how do we incorporate them into workflows and processes? How do we train people to use these systems? And and also really be the advisor, you know, to the

director on technology and on these kinds of technologies. And that's really what that role became about over time.

We set up an OTA consortium to make it easier to buy some of these technologies. We worked with ink Tel and things like that, which is the CIA's venture capital organization, to even invest in these companies and to try to get them to do business. I would go directly talk to a lot of these companies because again, at that time, a lot of Silicon Valley did not want to do business with an intelligence agency. They really didn't.

They did not want to be our customer. I couldn't even give them money.

so forth, I kind of knew. I had been on the other side, so I sort of knew how to communicate a little bit with the investors and with the companies, and that A big part of my role was at so to go and get this technology. But that was an obstacle, you know, to to sort of convince people. A lot of what I did was give speeches to people like that and and meet with you know, vcs and so forth, and and then but then once you get the technology. Actually it turned out that was the

easy part. Like that was the It just got harder from there, right because now, okay, I've got this cool you know, algorithm approach, this this computer vision system made by some cool Silicon Valley company where guys are working in hoodie sweatshirts, you know, at a we work and and then yeah and then uh they and then all of a sudden you realize, well, you can't just like plug that in right there there. There wasn't a lot

of a lot of cloud computing at the time. Most of what we did was still on prem It was out of servers that we own that were you know, government owned and operated, and there were the way the workflow works, the way the processes work in these agencies.

There's a lot of pieces of locked software that have been made through this really long, multi year requirements process by a single contractor, and to mod that in any way requires another year, two years, you know, to say, hey, well, I just want to take the data from the system and incorporated this other algorithm. They're like, yeah, no problem that. You know that twenty five million dollars and eighteen months we got it. Just no problem. And you're like, that's

just not going to work. And so like that integration

part becomes really really hard and threatens people, right because companies contracting companies have those contracts. They employ people in states to build that stuff, and they call their congressman if they're threatened. People who job it is to be a pro program manager at n GA are threatened because hey, I've worked my whole life to get this system in place.

You know, it's taken me years, and now you're telling me you want to remove it and put in this thing that these guys built in their garage, you know, in six months, and they've never they don't even have clearances. By the way, I don't even know if this is good, which gets into the other issue is, you know, it's got to be secure, you know, unlike a company where you know, if you get hacked at a company, it's it's it's a bad day. You know, you maybe lose

some money. You know, your product fails. You know you have an unsecure system. In an intelligence agency, you know, somebody dies right like that that that soft guy in the field now has a piece of compromised you know, map or or imagery that is dangerous, right, and so you've got to take the security piece very seriously and figure out how to you know, now, how do you ensure that this kind of technology is secure? And how

do you bring bring those companies in. You've got to get clearances now for companies where you know, the CEO has never even heard of a clearance. They don't even know what you're talking about. And you're saying, no, I need, I need to your guys to get a clearance. They need to come, you know, work in the Beltway now,

and so you've got to go through that. And then let's just say you can make it through all that right now, you've got to convince an analyst to use it because you know, remember these these people take their job very very seriously, right. I don't know if if you watch House of Dynamite, that that that movie Catherine Bigelow, and it's it's about an unwarned launch where you know nuclear missile is coming for America and it's gonna blow up and it's killed millions of people, and it's it's

it's a very intense movie. But this is a real job. People go to work every day and they worry about that day and day out. That's what an NGAA analyst is doing. They're worried somebody's going to launch a missile and it's going to hit a wahu and it's going to kill a million people. And so when you kind of walk in and you say, I know you have your system locked down, and you've been doing this for years and you've been trained to do this, you're the best at it in the world, and I know this

is the way you like to do it. But hey, I have this new technology. It's going to fundamentally change how you do it. And you know they're going to be skeptical and they're going to be like, I don't trust it. I'm not sure how it gets integrated, and not just them, by the way, the customers as well outside of the agency who are sending these reports that they're going to be like, wait, you're telling me this.

You know, this piece of intelligence was made by a machine, not by Joe, who I've been working with for the last fifteen years, who's the best imagery analyst. You got. It's made by a machine. They're not going to trust it either, and so you have to get over that. So there's a lot of of you know, working with people to build that trust. There's a lot of tests and evaluation. There's a lot of training for people of how to use these systems, ultimately recruiting people now to

help you make all these systems and integrate them. So it becomes this very big issue, and it was hard, and you know, I worked on it and lots of other people date as well. But it is starting to happen now. And I'll say one more thing, which is recently, so not this current director NGA, but the one previous.

So eight nine years after I had started down this path, the director announced publicly that they are now publishing intelligence reports in which quote unquote no human hands have touched them. So it is happening, and it only took you know, eight or nine years.

you do it? Well? You know, one of the things is I think you got to start in a place where you are working with the analysts, for example, or with the people that are going to use it. As

you're developing this stuff, as you're acquiring it. You've got to bring them in and have them be part of the process so they see it, they kind of walk down the path with you and have choice and and a lot of them when they do see it are like, oh wow, that's that's actually pretty cool, Like I would like to use that, like how do we do it? And you've got to get that, you've got to get

buy in there. And I tried to, you know, work with with those folks as as much as I could, you know, and the director of the of Analysis at that time, who was also very much trying to do this, and and we would partner on these things. There would be tension because she's you know, she's worried about her people and I'm worried about what I got to do their retention. But we tried to work together. So that's

the first part. And I think you have to do a lot of tests and evaluation, like you need to show, you need to have rigorous testing and really show this is how it's going to work. Here are the limits. And and the thing is it can't be the limit can't be like perfection. People aren't perfect either, you know, like analysts fail sometimes it's something you know. They they can be compromised as well, you know, yeah, and AI can be hacked, but you know, a person can be recruited.

Trust me, It's what I used to do for a living, and so like you're you know, you've got to you've got to try to get to a point where it's you know, how good it is and what the performance envelope is and how close it is to a person and then and then show that and share that and be open with that with with the people who are ultimately going to use it, And then I think you you also have to. My view is is try it

out in less risky missions. Some people will say they want to try AI out and like business operations, and that's where a lot of people were at that time and saying, well, yeah, we can use this for like business stuff, you know, admin HR finance. I think you have to go straight for mission because that's the thing we have to get right. But there are missions that

are less risky than others. Right, you can worry about a mission in Africa rather than saying, yeah, we're gonna entrust it to monitoring you know, Russian nuclear weapons, Like we don't have to start with Russian nuclear weapons. We can entrust it to a mission that you know, if it if it fails, it is not the literal end of the world.

in new technology. Why do you want to bring this menace back into the world.

sat on some boards. I remained interested in tech and and in particular in in technologies that I think, you know, we're either dual use where I thought, you know, these are gonna maybe they have a commercial aspect, but they also are going to help the mission, or where I felt like it was an area that we needed to compete with China on. And I think biotechnology is one of those areas, and and America really needs to compete in that area. And I think Colossal's doing that. And

I'm I'm not in the company. I'm an advisor to the company, so I don't work there on a day to day basis. But Ben Lamb, the CEO, is is very strong CEO. He's doing you know, unbelievable things like Elon Musk level craziness, like saying, hey, when I first heard it, I had the same reaction. I was like, wait, what is this Jurassic Park? Are you going to try

to like resurrect of wooly Mammoth? This is insanity. But he partnered with this guy, George Church, who is like one of the great you know, like you know scientists in this space in the world. He really he's like a legend. And he's like George is like no, no, no, this is this is a doable thing. This can happen. And yeah, it's about the it's you know, it's about the wooly mammoth and the dire wolf, and they're they're

looking at other things. But it's also about you know, animals, and we've made extinct more recently as well, and and or ones that are highly endangered and so forth. So they're doing pretty amazing work.

to happen. And so I wrote an article and Foreign Affairs about the AI aspects of this. But then I realized, look, it's even bigger than just that, and so so what I did in the book is I kind of backed up and I said, you know, let's let's look at the history of intelligence how we got here. And I really started World War two because that was really the first time we had what what we would all consider

like a true intelligence agency with all the features. And then and then, you know, we shut that down after the World War Two and then stood up the CIA and all these other agencies like d i A and NSA and so forth, and that was the Second Revolution. It was an entirely different way of doing intelligence. The you know, OSS was like cowboys, you know, just out

there trying to you know, collect and do ops. The CIA and the NSA, these were professional, you know, well financed, large global organizations up against AGB, you know, up against this other there very professional global organization. And we really

professionalize intelligence. And then you know, nine eleven change intelligence again, and all of a sudden, we weren't going up against

nation states. We weren't going up against the Soviets. We're going up against this decentralized enemy in al Qaeda and other terrorist organizations, and we had to intelligence had to adapt, and we had to start using a lot more commercial stuff, for example, using the Internet, using cell phones and things like that, having this kind of network intelligence and using

more open source information for example. That changed again, and now we're hitting this new revolution, this fourth Revolution, where AI and China are really driving that revolution.

on everybody's minds and it's changing the world. I feel like biotechnology hasn't had that like LM moment, right, There hasn't been that jat GPT. But we're seeing the signs. I mean, they made this this COVID vaccine in like months instead of years. We're seeing you know, people curing

like blindness and diseases by changing people's DNA. Where Like, there is a real business in the world that is trying to resurrect the bly mammoth and that is a real possibility, and they birthed a dire wolf actually, so like it is starting to happen in a real way. It just hasn't sort of gotten big yet. And whenever you've got a technology like that, nations are going to care and they're going to figure out a way to use it for national security. They're also going to figure

out a way to spy on it. And when I was you know, I partially put that chapter in the book is I wanted people to realize this was now in the game. This is part of the game now. And there are you know, Chinese companies collecting DNA information globally. There's a company called BGI Genomics. It's a massive biotech company out of China that collects people's DNA and they do it through COVID tests. They do it through like prenatal tests for like pregnant women thankfully not available here

but available in Europe. And they're collecting this information. This same company has contracts with the PLA, with the People's Liberation Army and they're doing work in one of those things that they're doing work on the When you read the description, it talks about population improvement, right like like, okay, where are we going with this? And you're doing this with the PLA. And when I hear population improvement, and I hear people's liberation army. I hear super soldier like,

I don't care what they say. That that's my you know, analyst brain. That's what I would say is going on there?

potentially smarter. We're not there yet, I don't from what I read in the literature where we can do those big things because they're complicated. You know, intelligence is affected by many things. But people are absolutely working on it. Companies are working on it. But you know, governments are governments, and an authoritarian government and one that's known, for example, to have in the past performed experiments on political prisoners

for example, like illegal experiments. Like you've got to assume, of course they're not going to come out and say this, but you've got to assume they're doing those kinds of experiments, and they could be used. Like you said, they can make people stronger, they can make them smarter. A lot of people think super Soldier and they think, oh, it's just like going to be in the movie and it's

going to be really fast, strong guy. Actually, what scares me way more is like you give people an extra fifty IQ points, Like that's far scarier in our world of competition, where people like that could be become intelligence handlists or inventors or compete with us through robotics and other things. Right, So they could absolutely do that. There's an even darker side, which is could you create genetic bioweapons, right, and could you target a disease to a specific ethnic group, right,

or could you target it to a specific individual. And this can seem crazy, but it's actually already happened. So in South Africa under apartheid, they tried. They had a program, the apartheid government had a program to create you know, bioweapons and essence that would affect black people and wouldn't be targeted at white people like they This is reported and it's scary. So there is a history of this happening.

And I'll say I'll say one more thing as another signal that you can see this is happening, is that when g came to San Francisco to meet with Biden a few years ago and he got up from the table after the meeting, his guys came in immediately and wipe down every surface anything he touched with his body. Immediately they were there wiping it all down. So there's no genetic residue. Like they know, they know the danger and they're watching out for it.

They just change those rules, or they can do things secretly that we just wouldn't do. We just you know, we're not going to have a lab in experiment on an ethnic group. It's just not going to do that. Whereas in China they have literal concentration camps for the wigers and easily could experiment on an entire population if

they wanted. And maybe they are. It's hard to say because it's kept secret, but so I think they and they're definitely investing heavily in the space and BGI Genomics and other companies have stolen American ip in this space

to kind of move more quickly, and then they're investing their own money and training their own scientists and having them come over here and get PhDs and go back there. So they're definitely making a push, you know, and militarily it's hard to know what they're doing because it would be classified. Of course, on our side, you know, we've continued to do research dart Buzz, continue to have you know,

biological programs for example. But I don't think that we're you know, if you ask me, I don't think that we put the level of effort into it that we should be if you just simply look at the level of the threat, and to me, bioweapons are right there under nuclear weapon as the most dangerous possible thing, and I don't believe that we spend just under nuclear weapons

amount of money on this. I think, for example, we need to do far more monitoring of diseases because people saw what could happen with a major disease with COVID. Nation states saw it. Also, terrorist groups saw it and realize like, oh wow, this is how you do damage, trillions of dollars of damage and you kill millions of people. You know, that sounds great. To a lot of enemies of America, and so we need to really get better at monitoring for this stuff and then figuring out how

to respond. And then I believe, and I kind of suggested in the book, like we may need an entire intelligence agency that just does this, it just focuses on biothreats and genetic engineering and synthetic biology and all of these things in the future. It's that important.

arrested the guy. I haven't seen any reporting of really what's happened for all, you know, who knows. Maybe he's just working in a lab now, right, Like it's hard, that's the problem in an authoritarian state. You can't know what really happened, right, But you know, I think I

think if people see that happening. I think if somebody, if they see, if they if we see another COVID like epidemic where it's very clear that it was designed, I think people are going to wake up and become very, very worried.

say there's a missile. That's not there's no rule, right, It's it's actually thinking in a sense, And and you can give it any image, you don't have to show it even it sometimes and it can pick it out. That's what's sort of new here. And chat GPT and these other LM models like Claude and so forth are are the biggest example ever I think in history of this.

Like they they produce information like a person, and they can do things that a person that we have traditionally thought only a person could do because it requires some level of like thought and reaction and adaptation, right, and so in intelligence, look, I think just broadly people are going to apply AI to do all sorts of jobs that people do. It's gonna do. It's gonna draft contracts. It already is draft contracts for lawyers. It's gonna help teach kids, like you know, kids use this in school

and learn from different AI systems. It's going to make movies like if you've gone on soorra or nano banana or something like, can make make a little movie. Same thing is going to happen in intelligence and this technology AI is going to do all of the different things and intelligence that a person does today. It's going to do what's called the intelligent cycle. It's going to help the collect information, it's going to analyze that information, it's

going to disseminate it. It's going to help plan the next information that needs to be collected and so forth. And it's going to be used on sensors like drones to autonomously fly them, remove them, and to identify what they're seeing and then to collect that information and then send it to another drone or send it to another

AI system to be analyzed. So I think what's going to happen is that all of the parts of intelligence are going to be done by AI over time and I sort of say in the book, within five years, ninety plus percent of intelligence will will be done in some way by an AI. And that also what's going to happen is intelligence is going to be made for

and buy machines. So these AI systems are going to go collect and analyze intelligence and feed it to other AI systems, and some of them are going to be like drones feeding it to drones, or AI analytical systems feeding it to other AI analytical systems, and that's really new, and we really don't even know how that's going to work quite yet, but I'm pretty confident that that is

what will happen. And people intelligence officers are going to do jobs that are very very different, Like an analyst might not just their job might not be to do traditional analysis. Is probably not going to be to read a bunch of reports and then write a new report. They're maybe going to task an AI to read those reports and write the report, and they're going to Q see it or have a vision or figure out what

questions ask. But really I think what they're going to do is partner with somebody in operations or with a general or with a president to be their interface and that's what it will be to be an intelligence officer and an analyst, not to do it in the old way. Even for a case officer, I think you're going to need to use AI systems. You know, we're still going to collect human intelligence. Ultimately, we still occasionally have to

go recruit another human being to give as information. But you're going to need AI to help you deal with all the surveillance for example.

feed into each other. And you know, you're just the guy that gets the report out of the machine, and your whole job is to like hand it to the next guy. I mean, what do you that's like going to reshape the human species? I mean, if you think about it.

died off where they just never trained another generation. So we'd like have forgotten things or like iron working, you know, like iron working that you used to be able to do in the eighteen hundreds, Like there are very very few people left in the world who can do that kind of artisanal craft. Right, So we've definitely this has happened before with technology, and I think it will happen

with AI. It's inevitable, and we're going to have to at a minimum continue to teach people some of the basics. And I think about like long division sometimes, like when I was in school in the eighties and learning long division, and I knew there were calculators, and I'm like, why are you teaching me long division? Like I'm never going to do it, and they still stubbornly taught us. And it's true, like I can't remember the last time I

did long division by hand. I just would use a calculator, but you still needed to learn it so you could understand math generally, and so we could identify mathematicians and engineers and like, you know, do you you needed those building blocks? Right? I think the same thing will have

to happen where even though the AI can do this stuff, we still have to teach people, especially children, but even adults how to do this stuff. You're still going to need to learn how to write. You're still going to need to learn how to do intelligence analysis, even if you're not going to really do it that much in your day to day job.

ground and holding it. How do you see AI interfacing with that?

Like in Ukraine, they're these autonomous drones and there are also guys in trenches, you know, digging them out with a shovel that like literally you could take a Roman legionnaire and put him in that trench and he would know what to do. You know, he'd be like, oh, I see here's a shovel. I'm gonna dig it out, and like, okay, you don't have a bow and arrow, you have a rifle, but I get it. Like bad guy over there, I'm over here, he's shooting. I'm shooting. You know, he would

get it. It still works. But then there's this drone flying over. I think that there will be more of that, and like maybe a taste of it is like Operation Spiderweb, which is this Ukraine operation where they took the drones they had to get them for deployed into like Siberia and other parts of Russia and then they kind of flew them either remotely or autonomous lead to attack, but

they needed people to get downrange. So you're still doing these like super complex you know operations to move stuff and get it there and then and then you're letting the machines do it now, you know, imagine that in more of a warfare scenario where you've got maybe special operations guys who've got to get this downrange and are using these weapons or yeah, you're you're having the drones come in, but the people are you know, kind of innately flexible in a way that AI is just not

there and machines are not and survivable in a way, right Like, you know, at the end of the day, if you have a United States marine on an island, like he could run out of ammunition and he's going to be like, oh, I see, I can cut down this like you know, stick and turn it into a

spear and continue to fight. Right Like, We're still going to need guys downrange doing this or controlling these systems or positioning them or you know, ultimately to take and occupy a piece of land, you need to put a person there, Like putting a drone there is not going to work, Like the locals are going to be like they're just going to wait for the thing to run out of batteries, so you're still going to need to

put people. I think it's going to be this mix and nobody really knows what it looks like quite yet, but we're going to kind of figure it out in Ukraine's like the preview.

project could happen, I would be surprised. That's like by definition surprise, but at some point someone is going to make one of these computers. And the biggest thing for intelligence is that such a computer could break encryption and traditional encryption, and that doesn't just mean the encryption of what we're currently transmitting. So we could start today and you know, encrypt everything in what's called post quantum encryption. That couldn't be broken by such a computer. But it's

all the old information. And they could have collected that old information, including like nuclear information for example, right like

anything that would have been encrypted and transmitted over any period of time and going back to the beginning of you know, computers really and and then they'll you know, potentially could decrypt that and that that will be you know, a major turning point. I think that that information will be out there. Do you ever see that movie Sneakers Robert Redford from the nineties, great movie? You know that

that's you know, spoiler, that's what it's about. And you know they're able to break into anything and and sort of look at anything, and like a world like that could be really scary and for intelligence would change everything because all of our old intelligence, which is still sensitive by the way, twenty thirty year old intelligence is still important you know, it could be it could be decrypted, and that would be a major issue.

and technology, and that includes on the intelligence side. And China has been performing economic espionage on America for years, right They're they're collecting information about American companies about their intellectual property, and they're passing that to their own companies to replicate a product, for example, or they're they're trying to slow down our companies in some way. They you know, they're recruiting people to collect information like insider threats within

these organizations. So they're doing economic espionage. And we have always done a little bit of economic espionage, I mean, but it's not really been a big part of it because the Soviets weren't an economic superpower, like, it wasn't that big of an issue. After the defeat of the Soviet Union. Like in the nineties, we were the superpower, we were globalizing, we weren't focused on other you know, collecting other people's information that much. Clinton. President Clinton did

ask for some economic information, but not a ton. And then during the Global War on Terror, like we were focused on counter terrorism and like they were not economic actors. So all of a sudden, now we're worried about, you know, economic issues, and I think that we need to build up the capability now to do economic intelligence ourselves, to know what Chinese companies are doing, maybe to try to

stop them from doing it. Right, what if, for example, and I'll give you an example, what if there was another global epidemic that was far more powerful than COVID and millions and millions of people were dying, and China would not share with us a vaccine, which you think it's justified to steal that vaccine. Of course, like this is what you know our country does, we'd have to

steal it. So or what if we wanted to know if their companies were making this quantum computer that would be able to get our information and all the encrypted information, including about nuclear weapons and things like this and really threaten us like you've got to. I think you're justified to go get that information. And I think presidents need that, and the Trump administration has made it clear that economic

competition is a big part of its policy. They should be provided with the information they need to make decisions, and that includes classified secret information.

government help a company. When I was writing the book, I still thought that was like for boting, like, we're not going to do that. I just wrote about the other thing, which was, well, we need to collect this economic information that's important to national security and give it to policy makers so they can make the right decisions about economic issues. Right, so they can know how far along is China with making a quantum device, how far along are they with AI or biotechnology. We need to

know that information. It's national security. And their companies, by the way, are state controlled. Oftentimes they have state owned enterprises, and even if they're not owned by the state, they can be controlled by the state, so they're not purely you know, economic actors, and we need to be able to know what they're trying to do and if they're harming American companies or American people and stop that. Right. So that's what I was thinking about, and I think

that's what we need to do. Since I've written the book, the rules are changing and our government now owns ten percent of Intel, and we are investing directly in companies, you know, that make and manufacture things that are important to the national security industrial base, the defense industrial base, like companies that make rare earths like MP materials, for example. So I can now whereas even two years ago I probably would have gotten on here and said, no way,

we draw the line at sharing that information. Now I think we might be headed in a direction where we could share that information, and we might do so, and we would be justified to do so, because it's that important that we compete in these technology spaces or in these manufacturing and supply chain spaces, and so we'd be justified to do that now. At the same time, and this is the capitalist in me and the libertarian in me, I also don't want to break our system, which is

a good system where companies compete with each other. Right, So if you're picking winners who are bad winners, you're going to get bad companies, right, and like you're not going to pick the right one, and government is not great at picking like the best technology of the best company. So I still want there to be that competition personally.

So what I would love to see is a solution where there was a way, for example, to you know, get information if it was necessary, and supply it to an American company if we had to, but to do that in a way that maintained American companies competing with each other, so it didn't like a fairweb.

open world than in the classified world. And now it's even more the case because again we're competing in these areas like economics or science and technology, where almost all of that information is open source. It's not classified, it's not made behind some you know, in some military base, it's it's at a company, in an office somewhere, and so the intelligence community needs to be able to use that information effectively, to obtain it and use it effectively

to to compete in those areas. In particular, Now there's a downside to open source and which is, well, it's open source, so the you know, your adversary knows that

it's out there and can compromise it. They could change it, right they or or if there's there's open source in the sense of like commercial satellites for example, that you know, you, you and I could go buy commercial satellite image right now, you know online, but you know, China or Russia or these other countries know that they're up there, and so

they could hide their things. They could use a cyber attack to change you know, what the what the satellite sees or something like this, And so we've got to be careful. At the same time, it's secure about using this open source information, and we still need the classified stuff, right because that's really where you get your edge, Like I want an unfair advantage, Like that's the nature of intelligence. That's the game. It's it's not like this isn't there

are no rules. I'm purposely trying to cheat. I'm purposely trying to get something that you don't know about. You didn't even know I was able to do it. I'm doing it secretly, clandestinely, right, And so we still need those classified systems. So I think we sort of need. I call it like a Barbell approach, right, Like we need the open source stuff, but we need the super classified stuff too.

is cognitive warfare, disinformation, these type of subjects. And I think back to a conversation we had with somebody about the about Russia and the war in Ukraine, and he was kind of saying, how like this is sort of like the ultimate checkmate against America because a country like Russia can feed us disinformation and feed our public disinformation all day, but here in America we have the First Amendment. You know, you can't stop people from talking, you can't

tell them don't say that. And it's like kind of the ultimate checkmate for somebody who's a dictator that they can mess with us all day and we don't have a defense.

And what you're doing in cognitive warfare is you're trying to influence you know who you're competing against, your adversary, and get them to maybe not want to fight, right, Like, that's what propaganda is often about, like get people to not want to fight or disrupt them. You know, Russia tried to disrupt the twenty sixteen election. I don't personally think they were even choosing a side. I think they

didn't care. I think they just wanted to create disruption and chaos and make Americans sort of hate each other more, right, Like that was the goal, and that's a win for them, right. And that's another form of cognitive warfare where you're not you're you're influencing people, but maybe not to an end. What's what's happening now, and what really scares me is that now you've got cognitive warfare and artificial intelligence and

propaganda and information operation. Cognitive warfare in the past used to be kind of like a one way street. You could essentially advertise, right, like you could put out a flyer or an ad or on Facebook like they did in the election hack, or you could you know, like spread rumors and things like this. Right, but now with AI, AI is like a two way street. Right when you talk to an AI, you talk to it like you ask it a question, then ask you a question back.

And and AI is becoming this arbitrar of information. It's where we're all going to go to get information in the future. And the danger is that if an AI system is owned by an adversary country like China owns deep Seek and it's still available here, or they infiltrate one of ours, they hack it in some way, which

is not as hard as you would think. There's some studies recently by Anthropic for example, like that show that it's it's actually pretty pretty straightforward to change these big lms that they could use that to go to do cognitive warfare and to influence people using these systems, and I believe influence them subtly and even go after like kids in grade school with this kind of stuff because kids are starting to use AI as well. So it's

a major threat. So we're seeing, you know, this move from these traditional propaganda like threats to the middle ground is probably like social media like threats like TikTok, which is a massive threat by the way, and should absolutely be handed over to American company or shut down because it's targeted a generation generation Z and is used. It sensors information and it influences its users, and this has been demonstrated. But we're going to move from that into

AI being like the platform to perform cognitive warfare. And that's that's incredibly dangerous and we really don't even have defenses right now, and to your earlier point, we don't even necessarily want the government to censor this kind of stuff, right So who do you have? And that's partially what I write in the book of how well, what is the solution I get into a solution?

as we've been talking about the public being propagandized. But I think right now today, as a soldier, you get a message over the radio over the assault net from your commander and he's giving you an order, and he tells you to go somewhere and do something, and you have confidence that in that information, that that is your commander. You saw the guy at work today, you know who he is, his commander. His orders are coming over an encrypted net. It's coming right to you.

to do that and trick the other guy. And you know, you look back even at like in in in in in in World War two, you know, the password could be compromised or something, so you know, people guys would do things like be like, well, who's the shortstop on the Yankees? Or like ask them questions where you know, you can ask multiple questions to be like how do

you know you know? And and even during you know, al Qaeda actually went through this because to them, like we were the technological enemy, right, and so they're like, oh my god, these guys can listen to everything we're doing. They've got drones, they can spy on my cell phone, and so they would do stuff like they would never

be like here's the operation. They'd be like, hey, you know, Mohammed, you remember that place when we were kids that we would go and you know, we would hang out with you know, so and so, and we would go climb that hill and there was a swing set. Well that's you remember the direction of that place from the village, Well,

that's the direction I want you to go. And like doesn't matter how much encryption or like sneaky people like you don't know, there's a there's like two guys in the world who know that, and it's those two guys, and so how do you So there are ways to you know, exist and this. And then the thing is, it's not just going to happen to these soldiers. I

think it's gonna happen to everybody. This is really what I get into the book is that everyone is now going to be spied on like this and have information operations committed against them, all of us and even kids. This is going to happen. They're gonna have those same doubts. And right now you already see like where you know, old folks sometimes have like their you know, their grandkids,

that somebody will fake the voice and last them for money. Well, that could also be done, by the way, by a foreign intelligence agency if they wanted, or it could just be done as a type of disruption information operation if you did it a lot, just to get people concerned and scared, for example.

or American intelligence is hanging by a thread. It needs to be saved, I mean, never mind reformed, Like, can we save this thing? Can we resurrect these relationships?

What do you do in that kind of scenario? And you don't necessarily want an intelligence agency to help you there, because the last thing we want is a CIA to come in and start censoring information because they're like, oh, this is disinformation, so I'm in a censor and it's like, well no, actually, even if it is disinformation, even if you know for a fact, that was created by a gru you know, Russian intelligence unit, I still like, somebody still has a right to see it, like it's crazy,

but that's true and that's a fact, and so you can't censor this stuff. So actually, I think shutting down the form align the counterform Malign Influence Center for example at Odie and I was like a good idea, Like I don't think it was particularly harmful, but I can imagine this scenario where it could become harmful, you know, if any side decided they wanted to use that to

manipulate information. So nobody wants that, but you still have to protect yourself against these threats, right, And I think the way to protect ourselves is that everybody now has to become that resilient endpoint, right, And you have to get trained to think like an intelligence officer and have some level of caution and say, oh, that could be you know, a piece of disinformation. This could be actually somebody just trying to collect information so they can model

how I think. And this AI could use that to influence me over time. Right, And then there are ways that you can be taught to, like intelligence officers are to deal with that doubt, right, Like one way you deal with it is triangulating information. You just never trust one source, right, Like the most intelligence officers would never trust a single source, no matter how good you thought

it was. It could be a satellite you built with your own hands, and you'd be like, I'm still have some level of doubt here, right, Anything could happen, and so you use multiple sources, right, or you assess the risk of Oh, you look at a piece of technology and say who owns it? And you trace that back and you're like, oh, it is owned by China, should

you know, be careful about it. So I think that we have to start looking at people as the endpoint almost in like cyber security terms, and make them resilient at the individual citizen level, where you're protecting yourself, you're protecting your family and your community, and that's separate from

an intelligence agency. And I almost think about it as like police versus neighborhood watch, right, Like we we still have police and we still need those guys to go out and shut down the next Internet Research Agency IRA threat, right, But we also need people who are part of the community, looking out for each other and looking out for ourselves.

And so when I think about what's going on in the intelligence community today, like you know, we can argue about whether the you know, there's changes here or there. I actually think we need to actually think outside of

the agencies entirely and make those agencies better. They should use AI and do all these things, but we should be building as civil society, as a group of citizens our own you know, intelligence function to protect ourselves, because that's the only way to protect yourself from some of these threats.

I think we're going to need that gene editing, fifty additional IQ points for each person in America, to navigate all of that.

against cyber threats, right. I think we can do the same thing for espionage threats and information operation threats and teach them very basic skills starting in grade school in like an age appropriate way in grade school, but ultimately by high school and college and in work, like in a real way, here are some basic things you can do.

Don't trust just any piece of information you see, right, You need to think about where it came from, what's the source, triangulate it with something like if it looks weird, it probably is weird, right Like, And hey, don't trust any piece of technology that you're given. Like we've been taught as a society to just trust technology because historically it's just been a consumer good that was made by

another American. That's not the case anymore. Like apps and so forth come from people who don't have our best intentions at heart, right, and they're not just trying to make money, And so you need to start and look

and say, Okay, I don't just simply trust technology. I think like an intelligence officer, I'm going to assess where it's from and what the threats might be and give it a second and those kinds of things they those are These are basic life skills that I do think can be taught at a literal level in grade school and in high school and at work and and so forth. But I don't even think they necessarily have to even

be taught by the government. They can be you know, taught by families and you know, and and parents can teach their kids and and and so forth. Right like that, it's not rocket science, Like like you're your your you know, your trust me, your average case officer is not a rocket scientist. Like these are people who like you know, are just like good at talking and like taking people to dinner. Like this is this does not have to be you know, super super complicated, but like there's just

a basic way to think. And it doesn't mean being paranoid all the time. Just means being basically cautious the way we are with crime. Right Like you you know, if you have if you have a child, you teach your child like from an early age, like, hey, we're in a bad neighborhood. Just be careful, hold my hand, don't leave stuff out in the car if it's parked overnight, you know, because if somebody might break in. Like we teach these things. I mean, we teach kids active shooter drills. Unfortunately,

like they understand there are threats in the world. This is another threat and let's you know, go out and show people how to deal with this threat.

what is to me obvious AI slop. It's just somebody had AI generate a picture through it on Facebook, And now these people are sixty seventy years old losing their minds about it. And it's like, I see when I see that, I'm like, I'm not sure how you knows as a society, how we recover from sort of like

our feet have come off the ground. You mentioned you mentioned Baudriard in your book, and you know how there was a time where, you know, maybe in the nineties when he was writing that stuff, it didn't seem totally real or relevant. But now here we are in twenty twenty five, and it really does feel like we're living in that world.

you're just not sure what the trust. But I'll tell you, what I really worry about is actually not those kind of call them big fake things, like obvious fake things.

Like what I really worry about is when it's like really subtle, right, like when you just change and adjective that someone uses, or even like the tone of their voice, that subtly influences how somebody thinks about something, right, And it'd be so hard to spot that, right if you're just changing single words or tone, right, and and but doing that those subtle changes continuously and over time, like you're talking to an AI and it never says something

blatant like Tiana mins Square never happened, the Weakers are free people, Like it never says anything like that. It just always gives you like a slightly skewed answer, right, like you know, like you know Tianamen Square. Instead of saying Tianaman Square was a terrible human rights abuse, it says Tianaman Square could have been done better, you know, something like that, right, like it, like it. It just makes these subtle change or even more subtle than that,

right over time, and that influences people over time. And that's what like our generation I think is going to fall for, like our parents generation is falling for like the blatant, fake ones, but like I think we're going to fall for the subtle stuff where like you can't even see it and you're like, oh, this looks like a New York Times story or a Wall Street Journal story or a news back story, whatever your preferred like media outlet is, and you're like, and everything it adds

on the site where I read it through chat GPT, but there's just subtle words changed. And that that's to me the scariest thing and probably where all this goes.

see the weaker stuff here, just less. And then what they've shown these groups is that it's working because the longer they show this, the longer you are on TikTok, the more you use it, the more benevolent view of the human rights record of the Chinese Communist Party you will have. Right. And so it's like just these subtle things have an effect, and when you look at it across a generation like and millions of people, it has

enough of effect. For example, that could lead to more of a not being elected who wanted to defend Taiwan in a war. And so over time, as there are more and more voters in this group and just enough of them have been skewed in this direction, America all of a sudden, China doesn't have to worry about is defending Taiwan because they have convinced an entire generation that we shouldn't defend it in the first place.

that directly in government and become an intelligence officer. And by the way, you don't need to study you know, international relations and speak five languages anymore. Maybe you're a technologist or you have you have an MBA, and right like, we need MBAs because we're doing economic intelligence, right like, so there's still room to serve and help it. But even if you're outside of intelligence, like you're working at a tech company or something, you can still help that company.

Like building these dual use technologies matters. It helps, like we need them. We need people who care about this stuff, and we also need people to just know what's going on in intelligence so that you can kind of be part of the system, maybe become like a citizen intelligence officer, like I said, to defend yourself and your family and your community, but also to know what our government's doing. Right. And like every time there's this intelligence revolution, there's been

a backlash. We had the Church Commission, we had Edward Snowden, Like, inevitably intelligence will overstep the bounds. That just seems to be a truism unfortunately. But by paying attention, bye, by focusing on what's going on and like, you can help to ensure that it does the right things. We still need to transform, We still need to you know, touch

the chalk line, and that's what intelligence does sometimes. But you know, you can play a role in elect the right people and make the right choices, and that's important. And so that's I was, you know, I hoped I sort of ended the book on that, and I think it's sort of the most important takeaway, Anthony.

and whatever's topical or current on the high Side. So it's another way for us to get the information out to you as social media algorithms are pretty iffy and you never really.

up the link. The link will also be down the description if you're looking for it there, and that's Teamhouse Podcast, dot kit k I, t Kilo India, Tango dot com, backslash joint S,