"Mapping Military Power" with Michael Hilliard

You're listening to a stage talk titled Mapping Military Power. This week we were joined by geopolitical and defence analyst Michael Hilliard from the Oxus Society. Michael spoke about their recent mapping exercise collecting years of data to pinpoint the scale of military power per region in Central Asia. My call was also joined shortly on stage during the talk by Mitchell Pullman, an OSINT researcher who helped with the mapping project. Within the talk, we discussed all sorts,

from drones to Chinese influence. You can find links to all the resources mentioned in the talk in the podcast description. This talk was hosted by me, Charlotte Mar, on Thursday the 6th of November, 2025, in the Bellingcat Discord server. Let's get started. Okay, welcome to another stage talk. Today, we're talking about military forces. We'll be looking at the central question. How do you map an army and its resources? What features represent military power? Here to enlighten us

is Michael Hilliard. Michael is a Director of Defence and Security Analysis at the Oxford Society for Central Asian Affairs. where he leads the new and expansive Armed Forces of Central Asia

project. Drawing on his experiences as a policy advisor and defence analyst, a former conflict journalist and the host of the geopolitical podcast The Red Line and Spotlight on Central Asia, by the way, you must listen to them and I'll pop the links to those in the chat later, Michael can answer complex security questions using data. Today he's going to unpack their latest report

on the military force of Uzbekistan. a region known as a leading military power in Central Asia, as well as highlight the useful military units and map tracker created as part of the larger armed forces of Central Asia project. By the end of the talk, we should have a better understanding of the military might of Central Asia and the power dynamics at play there, as well as how we could develop a similar mapping

exercise in other regions of the world. As we talk, please make sure to add your questions in the chat box via the message bubble icon. in the top right corner of your screen, as I just mentioned. And please note, within your question, if you do not want me to read your username out, as this is going on podcast platforms. Okay. Over to you, Michael. I'm going to mute and give you the floor. You're far too kind to

me, Charlie. It's great to be here again. I absolutely love the last time I was here and I am here today. I'm going to share my screens with you a bit, but obviously we're here for the podcast. Let's talk a bit about the project. I've been heading up and effectively talk about what Becca's done a bit. Obviously we have our second chapter of the Armed Forces of Central Asia report, effectively giving what is probably the most comprehensive analysis of the five Central Asian republics.

We're releasing each chapter alongside a bunch of maps. So this is the second chapter or the public version of that chapter. And so far we've located about 5 ,400 military and security locations across Central Asia. So quite a lot of work on that one. Now, part of the re -prompt we got here today is effectively looking at how to measure military and security power. I think that's an interesting question. So there are a lot of outlets online who do, and for lack of a better word,

try and put a number on firepower. And I think it's very difficult to do that most of the time. militaries really aren't easily to quantify on how good they are at what they do. I'm not going to throw a GPF under the bus. I'm just, you know, this is kind of the one I see quoted the most in public studies. I want to take a look and kind of talk about the problem with trying to quantify things on a really solid number. Looking at three militaries. And effectively here we

have Australia, Indonesia, and Uzbekistan. And you can see here the national GDP, Australia is much larger. But so is the military budget. Most countries do tend to kind of go to that way, where the military budget goes up and down depending on how rich the country is. Bigger budget, more money to play with, understandable. But when you dive into sort of more military analysis, you start to see things like this.

You'll start to see effectively, if you're not looking at my screen, where Indonesia, even though it's got a much smaller military budget, has like 400 ,000 personnel in its active service. Whereas Australia's only got about 58 and Uzbekistan's only got about 48 and then it gets bigger once you put paramilitary in. But that doesn't make sense comparing what we just talked about, which is military budgets being, you know, Australia

being the much largest one. And then again, it changes again when you start looking at things like tanks per person, you know, as an example, Uzbekistan actually has effectively 73 personnel for every tank, whereas Australia has about 481 and Indonesia's got about 2 ,900 people per tank, which is a little off on the ratios. So even when you look at military budgets, it's hard to quantify how much that translates to, I guess, the ability of that military and what it needs

to do. There's quite a lot of reasons for that. We're also, if we look at the percentage of the active force in each military, whilst Uzbekistan and Indonesia have most of their military, 75 % nearly, put towards, let's call it the active ground forces. I'm only talking about the actual military here. The Indonesians have very little within their air force and the Uzbeks, understandably, have much less within their navy or naval forces. Again, because militaries are kind of made for

purpose. And this is why it's really difficult to quantify exact numbers on what a military is capable of. Australia, it's built its entire military around doing a few really quite niche tasks. We are pretty good at going out and putting a battalion somewhere in the South Pacific or in the South China Sea, holding that ground. When we insert places, we're pretty good and we have a lot of equipment and technology with

us. But if you were to ask Australia to plonk on a field or you were to ask Australia to run a long mountain warfare campaign, we might not be very good in that. Indonesia, if you look at that as a military effectively is much, you know, it doesn't have the kind of air force that a lot of other countries have. It doesn't have the massive, uh, tank forces, you know, that less even as Bekistan has, which is a much smaller

GDP and budget. But again, it's designed as a military that can quickly respond anywhere around the country. And again, because it's an archipelago, anywhere across that massive archipelago, very quickly, you know, whether it's crushing riots or responding to a threat. knowing that it will probably be a while before they get resupplied. People design their militaries to deal with the

situation they're in. And it's why I really do enjoy looking at militaries because you do get a good look at long -term planning and long -term thinking. For Uzbekistan, I think that is the most obvious out of them. And what I've got on screen for people who are watching, but you can also check it out in the report, is where the

major brigade level stuff is in Uzbekistan. What you can see pretty quickly is the Most of your tank units, your kind of real heavy, I'm thinking old kind of Soviet style stuff is right there down the south, right near the border with Afghanistan. They view that as the conventional threat. It's also the approaches that you'd use to go toward Tajikistan, we'll talk a bit about that later.

Whereas your rapid response units are often out near the capital, Tashkent, or in the Fagana Valley, partly to do with the fact and when we interviewed lots of Uzbek military officers and personnel. They kept bringing up events in 2005 in an incident called Andijan. Effectively, what happened is there was an uprising, a very small uprising at the beginning, who actually managed to take a military base by surprise because they were just not very well coordinated at the time.

The success there gave them a bit of a boost. They went on and took a MIA outpost. They went on and took more and more outposts in the town. Eventually, they had taken most of the administrative block in the city and then eventually the military comes back and roars back into town a few days later and thousands will die afterwards. There is a huge reaction to that following the 2005 events that Uzbekistan has decided that one of its major priorities right now is to make sure

it can quickly respond to things. And it's something that doesn't matter how high or low through the military, you keep hearing this, had we been able to respond quickly while there's 30 people, well, less people would have died, which I see the... very morbid logic behind it, but it's interesting. Again, this is a military that's built for quickly responding to crises. The conventional stuff's down toward the border with Afghanistan, because it's obviously a slight conventional

threat. And most of your rocket or heavy artillery stuff is actually near the center of the country, which means you can effectively pivot at where it needs to go. And again, these are logical

decisions that kind of make sense. When we look at kind of Uzbekistan, I guess, in the context of the five Central Asian republics, that being Kazakhstan, Uzbekistan, Kyrgyzstan, Tajikistan, Turkmenistan, almost every graph you are going to look like, you know, is effectively going to look a little bit like this, but on screen, imagine you're at a one to 10, 10 being the best, one being the worst, Kazakhstan is always going to be kind of a 10, Uzbekistan is usually going

to be, let's call it an eight or a nine, Turkmenistan would be about a four or a five, and then Kyrgyzstan, Tajikistan are going to be down in the ones and twos. Not saying they're bad, but just I'm trying to make it as easy as possible. This here is military spending. And again, that tracks. We saw that GDP matches military spending, except in Central Asia's case where, whilst Uzbekistan tends to spend much more than you'd expect as a share of regional military spending, Kazakhstan

tends to spend much less. But it's also because Kazakhstan is so much richer than the other states that it can effectively afford to spend 1 % of GDP or a bit above. and still be the dominant player effectively. Whereas Uzbekistan is traditionally spent toward the 4 % mark and starts to dip back down toward two as things start to stabilize. But again, I think military, even if we're looking for a way of quantifying things, military spending

doesn't even do that either. Because not only does it get a bit wonky when you start getting to purchasing power, it's obviously cheaper to keep the Russians armed and sort of in the field because they spend less per soldier than it would be, let's say, compared to the Americans. But at the same time, you also have what gets included in that. Military research, for instance, gets included quite often within the US DOD budget, but in China that often doesn't. It's a separate

budget. And that starts to make things really wonky. Again, we have what was Bekistan with quite a large military and is the largest by numbers, but does that mean they actually have the best military? I'd argue probably not, particularly when we start to look at things like paramilitary. So these are... If people who really know what they're looking at, let's call MIA forces. Effectively, they're not going to have the heavy artillery, they're not going to have the heavy tanks. They're

mostly designed for internal operations. They're better trained at riot control, better shields, that kind of stuff compared to your standard military. And Uzbekistan actually has the lowest out of the five Central Asian Republics ratio of paramilitary to military. In other words, if you combine that number together, 63 % sit

within the conventional military. with 36 % sitting within border forces, MIA forces, emergency services forces, which means they have quite a heavy, still somewhat Soviet military sitting at the core of this. And that's a big problem because when you have a massive numbers military, it does help with unemployment. It does help with a whole bunch of other factors, but it also really hurts your ability to... So if you spend money, because you're spending way too much money on

wages and pensions. If we go back to nominal GDP, this is what I was talking about, where even though Uzbekistan and Kazakhstan are somewhat similar military power wise, the GDP by Kazakhstan is just so much higher in the region and growing at a much higher rate. And there's a whole bunch of reasons for that. Even if we look at, as you can see from the pie chart, but for people, Kazakhstan alone makes up about 60 % or 57%. of the entire

region's GDP. So it's not unusual to expect that they would be the core military power, even though they haven't got the largest military in the region and have got less tanks or whatnot. Part of the other thing you want to look at when assessing military capabilities is who their suppliers

are and how over -reliant they are. And I guess you'd almost on a scale of things, look to someone like Uzbekistan on nearly one end of the scale where Russia or the USSR supplied let's call it 85 -ish percent of all of their equipment. And I guess this is an interesting one for me, I guess. It's almost a drinking game for Central Asia analysts. We see all these headlines coming out of Turkey selling this or China selling this. And that's true, but it still doesn't make up

anyone near the maturity of what they use. And part of this is just simple economics. For Uzbekistan, if let's say you're the defense commander and you went, hey, what are we going to do about tanks? And they go, well, we could buy an American tank for $8 million, but we need new mechanics. It's going to cost $8 million. We need new manuals. We're going to train the guys and we could get sanctioned. If we have problems with Americans,

we need to buy new gearboxes and yada yada. If you go with the Russians though, you can probably spend 20, $30 ,000 on a modernization kit, slap it in and call it a day. So you do find that if they don't need to spend the capital, they tend to just slap modernization kits and new coats of paint on Russian equipment, which is why Russia is still pretty dominant for the kind

of basics and consumables. That changed a little bit over 2020 through 2023, as Russia effectively diverted more of its industry toward the war effort and, you know, what was left for the export market was competed for by India and whatnot. But, you know, it was back in time really just starting to roll back and Russia is taking a bigger part in their procurement as we speak.

I guess what the interesting thing though is when it isn't a decision between new tank and modernization kit and, you know, equipment versus equipment. that's where the really interesting part is happening. And that's where you ask them to see China and Turkey and a few other places. Because when it comes down to, well, I need to buy a Russian drone or a Turkish drone, they tend to go and not with Russia. They do have

some Russian drones. They do have some Russian bits and pieces, but yeah, it is a testament to, we think Russia is pretty good for basic consumer. It's, you know, basic bullets, basic gearboxes. They're not so good for the high tech stuff at the moment. I think they're slightly distracted elsewhere. The other part you want to look at, and I get a lot of people don't include this when they start to look at military power,

is the ability to be self -sustainable. If you're Australia, we can kind of keep some of it going ourselves. We're not great for that, but we can. When you're, let's say, the central agents here, almost everything they have is imported in. For Uzbekistan, as a classic example, when they have their planes, which are getting quite old, needing repairs, they can't do anything by basic repairs in their own facilities. They end up sending them off to Belarus or Russia to get that done.

Which if you can imagine, let's say you're in a conflict with Tajikistan or one of your other regional players. Do you think Belarus is going to be hop, skip and jump to want to fix your planes? Not really. And you kind of have to take that into account when you're starting to plan and war game out a number of these scenarios. Even you know, something like here, which is the Churchill aircraft repair plant just north

of the Capitol, north east of the Capitol. This has been pretty under repaired for a while and is suffering quite badly. And frankly, you know, even though a lot of these States would probably benefit from, you know, combining their money and getting this back up and running like it was during the USSR times, kind of doesn't get that because no state trusts each other enough to do that. Now, and this is again, we'll weigh

into military calculations here. And again, this is one of the reasons that I think it's difficult to kind of put a number on, you know, exact number of how powerful a military is. Now, when you're looking at mapping bases, there are a few things you can look for. So actually, I want to kind of, before I get into the mapping stuff, let's go right through this and I'll come back to that.

The two things I found really interesting in this study is effectively, no, lots of things I found interesting, but one of the most interesting things is looking at purpose -built. So as part of this, we ran a bunch of war games, getting a bunch of military and security experts in and sat down and worked out how each guy would respond, what the situation is, what they're capable of, and effectively what the outcomes are likely

to be. Now, the two we ran for the Uzbek chapter, where one was an ISKP, so Islamic State Khorasan Province, effectively gathering up on the Afghan side of the river and launching a, using kind of rubber boats and dinghies, an attack into the southern city of Termez where the Uzbek -Afghan friendship bridges. Now, this was so difficult for the ISKP team that we actually need to give them an absurd amount of concessions to even get them off the ground. We need to say that

there was no riverine patrols that night. There were no surveillance that night. People were asleep when they weren't watching at the watchtowers. They took the massive base just over the bridge here. by surprise and pretty much everyone surrendered immediately. All stuff that even in the start of the chat we get is absurd and not going to happen because every other time we didn't give these concessions, the enemy forces would kind of build up on the other side of the river and

be immediately eliminated. And it just shows you that if Bekazdan has thought about this, Bekazdan knows this is a threat and has built up so much in the way of defensive materials and force down here that it's really good at doing that. They pretty quickly pushed the ISKP forces back. And even in this absurd scenario where they get a lot further than they would in reality, the whole thing ends in, I think it's about 24 or 36 hours. They're meant to do

that. Then we ran another scenario. And this is what I think surprised a lot more people is an Uzbek invasion of Tajikistan. Now, Uzbekistan being the largest military in Central Asia, Tajikistan being the smallest. Particularly with the fact that the Tajik capital is only 40km away from the Uzbek border, you'd think, well, this is an easy fight, Uzbekistan should win this, not a problem. And we're modeling off effectively

water disputes. They decide that they want to go for an invasion and take the northern capital of Khujand and Dushanbe at once. Again, they've got a massive military, it is in the realm of possibility. But immediately logistics becomes just so much of a problem. You are effectively marching your troops over long distances with very little in the way of ability to keep them

supplied. And whilst we kind of had to give them an artificial parameter as well, saying that the Russians wouldn't fire on them if their Russians weren't fired on first. And the Uzbeks were able to build up prior to the invasion completely by surprise, which again, I think is not possible. Day one goes fairly well as expected, Uzbekistan can pack a real punch on day one. They can even get in the north in better punch and they can really kind of get right to the outskirts of

the major city. On the southern front, Dushanbe, the capital of Tajikistan, they can get right to the edge. But once you start getting into your day seven -ish of the scenario, things really go badly. Where A, the Uzbeks aren't trained to do urban warfare very much anymore. It's not really what they're aiming to do. B, the logistics are just not there. They don't have the mobile refueling stations. They don't have the logistics

trucks. They don't have the communications staff and the dedicated communications and logistics units. Their stuff is still coming from hundreds of kilometers away in Termez at this point. They don't have the suitable railway units to get this all up and running. And you end up having this situation where they just end up losing way too many men. trying to fight this urban campaign without trying to maneuver themselves

away from the Russians as well. So the end result is effectively that the Uzbeks almost inevitably, because their top priority will be to make sure the administration remains stable, is they pull back. They go, it's not worth our time. Let's pull back. And that's a pretty bad scenario. If you've gone in, you lost 20 ,000, you're probably less than that. You've lost, I think it's about 8 ,000 personnel. And you've gained effectively little to nothing. So it ends up being a pretty

disastrous scenario for the Uzbeks. And again, a lot of people would take that away and say, well, obviously the Uzbek military is a weak military and on paper, their score is higher than the Tajiks. They should have won this. But it's just because they're not really designed to do that. You know, this is a military that is mostly designed to respond quickly to internal uprisings or pressures, respond to border skirmishes

where the Olympics are really good. If it's a short, sharp campaign, they're really good at that. But if it's a prolonged logistics heavy campaign, not really. And again, it comes down to that kind of what we talked about earlier with the tanks. If you're a defense commander and you say, you know, sir, what should we buy? Should we buy all the stuff we need to go invade

Tajikistan? Well, probably not. If you're not thinking you're going to be doing that, you put that money towards communications equipment and pensions and everything else you need to. And again, that can give me a bit more easy sleep at night knowing that none of these militaries are really preparing or even halfway prepared to launch massive invasions of the other. So if you're trying to measure military power, I

guess, which was the kind of prompt here. I don't know if you can, you can kind of rank things first to fifth in this region, but only because it's so blindingly obvious in order they're going to go in on almost every single metric. But that doesn't mean you can actually expect easy victories because their numbers are higher. Tajikistan, when we wargamed their invasion of Kyrgyzstan went really badly and they ended up pulling back

on I think day 10 or something. Kyrgyzstan and Venetian Tajikistan, higher number, still can't get past even the first, you know, can get past the first major city, but can't make it further. You know, this is the trouble we're trying to rank military power, I guess. So I'm going to obviously turn it back to Charlie and she's going to have some questions. But yeah, to, I guess, be an overall summary is it's very, very difficult to put hard numbers on things. Yeah. And I haven't

mentioned railroads yet. I'm sure someone will point that out, but Lots to talk about with the Uzbek chapter here. Yeah. Someone has just mentioned railroads, as you said it. Perhaps we can dive into that in a second. But to start with, I wanted to ask, and maybe this is an obvious question, but this is a lot of research, a lot of data. It's fascinating to see the practicalities of it through your war game scenarios. What were

your main sources of data for? this particular Uzbekistan report and how long and how many resources does it take to build a report like this as part of the Armed Forces of Central Asia project? This has effectively been a two year project. It's been massive to put together. And it's been combing through everything from, you know, government reports, you know, through NGO reports, through to weapons inspection reports, through to social

media posts. You know, the wonders of social media have made it not only very easy to find certain bases, but also how to gain a lot of context clues on what's there and what's not. When you're trying to find a unit, the amount of units we can't pin to a certain location, all of a sudden we'll find a local primary school did a excursion out to the base and took a bunch of photos from within the base. a random Facebook

or Telegram group is just insane. So it's effectively talking with lots and lots of experts, people who know this region quite well, talking with a lot or reading a lot of these reports and piecing together thousands and thousands of points of data to effectively try and have a pretty good data set to run with. And then once you get that, you can start to piece together other pieces.

And the fortunate thing is that a lot of these militaries are still somewhat similarly structured and based as what they were during the Soviet era, which does give you a baseline to start with. But generally, yes, it's just been, it's been two years of an entire research team kind of putting, pulling together a huge amount of data points. And how difficult is it to map kind of independent power? Like how much influence, for example, do foreign allies have on military

power in central Asia? You mentioned some of Russia's influence. Yeah. How does that come into play when you're looking at writing a report

on an individual country's power? This is some of the most difficult and I think, you know, it happens to everyone all the time where I think everyone has different, you know, we all do our math and everyone will do different scores and everyone will kind of have a different approach on this, but it's what we weight differently I think that has, is where people tend to differ.

you know, I found particularly working with different members of the team is that whilst I tend to put a lot more weight on logistics and economics to drive things, then other team members tend to think more about cultural dynamics and tend to think more about, you know, how the industry will work or various different factors that do drive their kind of what is important and what's not. When it comes to, I guess, the Russia question, Russia is definitely the kind of the military

power in the region. And I think it's weird to say, because when you go to the region, Russia is culturally losing a bit of a step. You know, it's only been in the last couple of years, I've gone, you know, flown into these countries and I've spoken Russian to them. And I get back, hey, we speak Uzbek here. I go, yeah, well, I'm sorry. You know, my bad. This is the cultural shift. But when it comes to the military and security services, a lot of those guys were trained

in Russia. They're still very friendly with Russia. All their manuals are often in Russian, particularly with the very Zinni command. Their primary language might even be Russian. It's also a interesting kind of tidbit is when you do a lot of training exercises that as much as China is starting to come back in and do a little bit more, the soldiers distrust the Chinese a fair bit and they don't

drink with them. They don't have as much fraternization with them, particularly at the sort of NCO level

and above. So Russia is... Yeah, still very much the supply power and the guy supplying the majority of the military equipment here, but also is still kind of the influencer when it comes to the higher ends of security and military forces and probably will be for at least the next 20 years or until that generation of the guys that are kind of coming into the service now reach the top levels of power, which again, it's a bit of a delay.

Yeah. So the other one you're talking about is how you can measure what power is in a base. So I got a few examples and obviously I have to describe the people listening, but you know, when you look at bigger basis, you, this, you know, it's a kind of a concept of when you have a really small business to give an example, a bit of an analogy here, if it's just you and one of the guy, that's what you need. But the moment you get to kind of 10 people, well, now

you're kind of need an accountant. And then once you get to 50 people, you probably need a HR person. You need an office manager and, you know, you just kind of need all these extra supplementary things. The bigger the base gets or bigger the company gets in our analogy. You know, I'm looking at a screen here, but this is quite a large base. This is out in Samui in Kazakhstan. And once you get to a base of this size, you start to need additional motor pools. You start to need

probably dedicated communications units. You start to need a dedicated explosive storage warehouse. It's just outside of town. Those are kind of things that you don't need for the smaller basis. So, you know, if we go to these, some of the smaller ones, let's call a brigade size. Yes, they do have large motor pools because they have to keep the vehicles somewhere. They do have dedicated barracks because of the people living there and you can get a good idea on how many

people tend to live there. They do have usually the parade grounds, which are quite common across these guys, but they don't have the kind of massive fuel and explosives dumps that you see in much of the larger base. So, it's probably going to be smaller and less capable. If we look at, let's say, going down another rung to border outposts, you know, the one on the right for people who have the visual here is actually the Chinese

base in Tajikistan. And as much as people have made a massive deal about it over the years, when you actually look at particularly the construction photos when it was originally built, you know, there's not a massive supplies of fuel here. There's not massive supplies of vehicles here either, even on the surface. And it just shows that whilst yes, it's a, you know, an important data point to take into account. This is not a base you'd launch a massive invasion with.

This is effectively a base you use to long -term guard a road. And so if they don't have the ability to kind of sustain operations by themselves in a good way for a long time, it gives you an idea on what their base's purpose is probably going to be. But yeah, I'll take another question before I ramble too much. I guess this is a good point as well to mention, re -mention the map that you have got going that tracks kind of military

vases. Please do have a check that out. I have put it in the chat, but it will also be in the podcast description as well as all of these links. And it's on screen for those in the audience now. Michael, quickly, do you want to just quickly do an overview of this? I've also got somebody who worked on the project who would like to jump in, Mitchell Polman. They've asked if they could

join the stage as well to help. So perhaps while you're talking through the map, I can invite them on and they can chat a little bit as well, if that's okay with you. Sure, go ahead. Yeah, obviously the map I'm looking at at the moment is got a little more data than the public versions of these maps that we sell. But the map here, you can see that Uzbekistan I think has got about 1400 pins by itself. Effectively, what we've done is tried to sort it into different categories

so you get a bit of a snapshot. As you can probably see here, let's take it down to the larger units, which are brigade and above. Again, for anyone military and security oriented in the audience, I'm going to use brigade and massive air quotes for this region of the world. Effectively, the idea of the map is that you have this central resource that you can effectively go... Oh, I wonder what's in, actually, let's go to Termez because we were talking about Termez as an example

earlier on. And I can just turn on a bunch of pins and I have a really good idea of what is in the city and where everything is. Because for playing our war games, it's really useful to know where everything is, but also generally it's just good to know the capabilities of each garrison. For instance, here, you can see up here in the north of the city is where the air defense batteries are and all the radar batteries that are effectively monitoring the south here.

You know, here's the naval port. You can see there was a green pin. It's part of the riverine flotillas. And it's just, it's just pouring together oodles and oodles and oodles of data. I mean, Mitch, you were working on this project as well. I mean, you can attest on just the sheer quantity of data you have to kind of pull together for a map like this. Oh yes. Can you hear me? Loud and clear, far away. Okay. Yeah. You've pretty

much got it covered, Michael. I just wanted to say that it was actually surprisingly easy to track down some of this information. It wasn't all that long ago that Uzbekistan was being referred to as the North Korea of Central Asia, and I was a little surprised. You could actually even find addresses and street addresses and telephone numbers of military units in the equivalent of an Uzbekistan phone book. It's right there online. It's all in Uzbek, of course, but it's there.

Sometimes they even list, you know, the individual units that are at a military base. So it was kind of surprisingly easy. I'd say that it was easy for some of them. In the early stage of the project, there was the low hanging fruits, other guys in the phone books. But when it comes to more sensitive areas, particularly when it comes to things like air defense batteries, that's

where things get much more difficult. Unlike the Russians tend to be a little more consistent with their military unit numbering, the Central Asians are really scattershot with it. Yeah, I was going to say that was sort of the bigger challenge was the numbers. And names of units changed all the time and keeping track of who was where and what was who was a bit of a challenge

for us, I think. Yeah. I mean, this is part of the project is obviously putting together massive data piles, which is far beyond what goes in the public version of these, of effectively everything we know about, let's say, you know, 11506. And it will be, hey, here's three reports we have that talk about it. Here is, you know, photos of the base that have been taken from a couple

of YouTube videos that we've got. And, you know, here's the size of the motor pools, which should indicate that there's this many things here, you know, putting all that together and effectively trying to build a bit of a concept of what's there and what they have capable. Again, I also used, really odd at times. Sorry. I also used, there were online, Uzbek military journals, newspapers that are online and going through them, you could oftentimes find, you know, good information or

at least clues. You know, even photos sometimes of, you know, that went to unit locations and things of that nature. And again, a lot of the time you have to just build these massive data piles and dumps because you're trying to piece together multiple different parts of information. You know, you know, There's a commander who is from this base. You know, he's done a news interview at this facility, which you can geolocate to

here. You don't know what branch he's on, but then you find a different report, you know, which effectively shows that commander not at that base, but the patch he's wearing would indicate he's from this unit. And you just have to have all these links together to kind of combine a little data later on. You know, sometimes it can be that difficult. Other times it can be even more difficult. Other times I remember looking at some drone footage coming out of, I think

it was Uzbekistan. It was Uzbekistan. I'm trying to figure out where is this drone footage from and you're sitting down with a cup of tea thinking you have to geolocate it and then pull down the file name and it had Fagana Airport written in it. So sometimes it does get a little easier than others. But I think we'll take more questions because we've got quite a lot of questions coming through. Yes, thank you. ever be professional.

Michael, thank you. And thank you, Mitch, for jumping on and giving us some insight into the open source research that you were doing there. Just a little PSA, obviously, this is the Bellingcat community, so we have to be careful. If you are working with personal data, please be careful, particularly when it involves militaries and make sure you follow procedures. But thank you so much, Mitch, for coming on and chatting. I've got a bunch more questions to throw at Michael,

though, if that's all right. So, so many. I wanted to start at the top because some people have been waiting a while. Fade asks, how do you see the advent of commercial drones, which you've just mentioned, in warfare improving slash hindering logistics in the future? Could hex copters be used to supply at the unit level? So that's a really great question and it's the million dollar question I get asked more than anything else

about this region. Drones can hinder and hurt and it really depends on what kind of a drone it is and what it can do. Again, people kind of think drones and I think they tend to think of the kind of bi -racked RTB -2 stuff when a lot of drones are mostly just there to either be a exploding munition or they're there just

to do reconnaissance. And if let's say, I am, you know, pull up and barb talk, I'm going to pull up and move up to Dushan, but if I'm, let's say monitoring that single road we have to go on to kind of approach the capital, you know, having drones Uh, monitoring that area is going to be great. It means I can effectively keep

a close eye on what's coming through. And I know that, Hey, you know, there's three tanks that have just come through this area, but also there's eight fuel trucks and you can attack much more easily into the, in this, the soft targets. And again, it comes down to that classic problem of a tank without fuel is just a really weird car. Um, you know, If you, if you can attack the logistics and the soft points, it makes any

war scenario much more difficult. And this is in our war gaming of sort of a Tajik invasion. The Uzbek team had massive problems with effectively the Tajiks just sitting in the foothills launching, you know, lots of small kind of needling drone attacks in the logistics line because it is a really narrow choke point to get toward the capital. When sort of hex copters, hex copters are great

for reconnaissance. They're good for munitions, but when you're trying to supply a front, you were talking massive quantities between tonnage most of the time, because soldiers just burn through lots and lots and lots of equipment. And particularly once you start to get the bigger unit levels, you know, whether it be everything from medical supplies to blood, to ammunition, to fuel, to food, to, you know, whatever you need, that's probably too difficult to try and

do by Higgs Copter at the moment. So at this point, it's making it much more difficult and you have to spend a lot more of your resources and troops protecting your rear areas. But drones are evolving at a pretty rapid pace. So I'm never going to say never. Okay, coming in quick and fast. Silverknife, one of our lovely moderators,

asked earlier about significant shifts. You mentioned at the beginning, I mentioned at the beginning as well, that Uzbekistan was seen as this big military power, but Kazakhstan on paper, to a point, has kind of overtaken that. Have you seen any significant shifts in structure, deployment and composition as you've been going through these reports? Absolutely. And I think this is the really fascinating part of these militaries is whilst they all kind of imagine you're looking

at a bunch of cakes. everyone's kind of got the same sponge cake. They're all somewhat similar inheritance from the Soviet Union, but where they've changed the flavoring tells you a massive amount about what their priorities are and how they want to structure things. Uzbekistan has, let's say, its ground forces and its air and air defense forces under the ministry of defense,

very standard to everyone else. Its internal troops under the MIA, pretty much like everyone else, and its emergency services get a separate department, which is like everyone else except Turkmenistan. But then you have things like the the National Guard and the Presidential Security

Service, which are fairly new. It was under the previous leader, Kareemov, who effectively ran the country almost as close as North Korea, who had so much power concentrated in what would have been the KGB back in the day, the State Security Service or SGB back then. The State Security Service at that point controlled the Frontier Service, which is the border guards, the internal troops, a big chunk of the Special Forces, the National Guard and the Presidential

Security Service. But now that Karimov has died and the head of the SGB at the time, Ian Lotov has been moved on. A lot of the powers have been ripped out from that agency and distributed to other agencies. And I think watching which agencies get extra resources and which ones don't give you a really good insight into the kind of the

internal power dynamics going on. That as an example, the National Guard, which was originally part of the security agencies, originally was kind of, was going to go to the Ministry of Defense. And then it was a bit, do I trust the Ministry of Defense? So it became a separate unit and was given, you know, would protect the president

at one point. And then all of a sudden, had the president ripped out the president's security service from the national guard and made an own separate unit that can investigate anyone as well. So it's those kinds of changes that seem quite subtle and an administrative on paper, but you know, actually tell you quite a lot of what's going on behind the scenes. Yeah, let's go to the next question. Yeah, that answers your question. Absolutely. Just with that in mind,

have you tried to address the temporal? dimension with these projects, because obviously the reports themselves are PDF documents, which kind of sets it in tone. Is that kind of where the military unit map comes into play? Where is that going to be updated live? Have you thought about the kind of time sensitive nature of a lot of this information? Yeah, so for our paid version of this effectively, you know, when a event happens.

So like, for instance, there was just a shootout on the border between Kyrgyzstan and Uzbekistan. We can identify that pin quickly and kind of pin the story there and say, Hey, look, this is what happened. This is the pin you're looking for. This is the result of that. But at the same time, I also wanted a version, a public version of this, which doesn't give everything, but does allow, particularly for the OSING community, rather than spending a million years like I have

and just staring at mountains of bitumen. um, you know, you can quickly identify which bases, which, and know where the pinch points are going to be. Uh, so yeah, we will be updating, uh, as we learn new locations, as things change, as things open up, I'll keep adding them to the both map, both the public and private. Uh, but also, you know, right now, obviously we've done all the mapping for all five countries, but as each chapter comes out, we will be releasing

that country's pins as well. So the next one out is Kyrgyzstan. So that should be exciting. Amazing. And I think someone's in the chat already put the purchase link in there, which is awesome. Thank you for sharing that. We've had quite a few questions on how you overcome deceptions and what are the key characteristics of deceptions

usually seen in Central Asia. We also had a follow up question on that in terms of have we seen any interesting evidence of these powers making efforts to insulate themselves against Chinese espionage as well as we're talking about hidden details. I mean, Chinese espionage is a great question. And I think Tajikistan is the one you probably talk about more for China. There was a few years ago, about almost a decade ago now,

I'll show you the whole thing. China effectively announced this base out here, the PAP facility, which is technically not a military because people are on police, but we're going to get into that another time. And everyone thought that was kind of going to be China's first step into the region to be a military player in the same way the Russians have, because Russia has quite large military

deployments out here. In fact, the largest military base outside of Russia is in Tajikistan, apart from what's in Ukraine at the moment, but that's a whole other thing. The Chinese though, At the same time, we're dealing with problems in Gwadar in the south of Pakistan, where effectively what was happening is they put more personnel in to run the port. Those personnel would come under attack because the locals would take jobs or

for a whole myriad of reasons. And then they put more security personnel in to protect those personnel to which that would just exacerbate the problem and you end up having a terrible cycle. So what's happened is the Chinese have looked at the headaches they've got in Pakistan and then looked at Tajikistan and said, I don't want any of that. I'm just going to throw money

at you. I will pay you Tajikistan to just put on extra border outposts, build extra border facilities, put on extra border guards because I don't want to deal with this headache. And whilst that has somewhat solved a small problem for China, it also creates other problems. So that for instance, Tajikistan has incentives to somewhat exaggerate border problems at times.

Where looking at some of these shootouts that we see reported in the media that Tajikistan talks to China and says, look, look what's happening, we need more money. And then we call locals in the area and we find out it's kind of one farmer who was drunk and trying to shoot some goats. At the same time, you also have problems like what we've seen just in the last couple of weeks in particular, is out in kind of an area called Satagor, which I've put up on the map for people

who are watching live. You know, you have effectively a gold mine on the Afghan side, which is funded by China and a gold mine on the Tajik side, which is also funded by China. And because the water keeps getting diverted and other water problems, the two sides are now shooting at each other. But China is also funding security services on both sides. At the same time, China has just built massive border guard training facilities

just north of here as well. So China is somewhat, simultaneously, I guess, very involved in almost every angle of the border problems. not in a kind of boots on the ground way that I think people were anticipating or looking for when this issue first came up. So it is a super interesting dynamic when it comes to China trying to deceive them. I mean, they will exchange some information, but depends what they will exchange will depend

on what sort of information it is. But there's also just a You know, again, it comes down to that security services problem of, you know, a lot of these spies and a lot of the mid -level officers don't trust the Chinese very much. The whole bunch of reasons, and I do find it interesting when you talk to particularly Tajiks, you know, you kind of ask them questions like, hey, China just built this massive highway for you. That's

pretty great. I mean, this is effectively, you know, people were dying on this fire all the time. This has cut your commute time by an hour. How do you feel about it? And the response will often be something like, well, they built a big enough tanks, didn't they? which is just absurd. China's not going to do that, but it just shows you this kind of underlying dynamic that quite often comes up in the region. All of that in

mind. And there's some funny comments in the chat right now about China playing both sides. Is there any concern when you're doing these military infrastructure mapping that the information could be misused? How do you deal with... the problem of malicious actors when you're working on these, particularly when you're doing war games and exercises like that. We're very careful in what we do put public and what we don't as well as who we do contract with. It's why you

can't just buy the data. You effectively have to be kind of vetted before we, you know, what level of data we see. I mean, if it's just kind of mapping stuff, we're a bit more open about that. But if it's coming down to like how many tanks for each base, that's a much more thorough process. Most of this stuff is going to be particularly in the public realm, not knowing, but it will be fairly easy. It's not going to make a massive difference to any actor. It will for Osinter

in keeping tabs on what's going on. And we've also made a decision to keep black sides off this, mostly for my... I have to go to the region every now and again, and I prefer not to spend the rest of my life in this big prism. But yes, we've had obviously lots of conversations about what does get included and what doesn't. And it's why we had to glaze over a bunch of very, very specific details. We had a question earlier about concern saying, should we worry? Do you

believe a real invasion is imminent? Dushanbi, for example, that you mentioned earlier. When you're doing these war games scenarios, have you come across any that have been likely or have caused concern? No. So this is an, it's an interesting question. We start pretty much most of the scenarios saying we do not think this is going to happen, but there's two problems.

One, when you brief quite often diplomats or non -military people, they often might not understand what you're trying to tell them unless they've got lines on the map, which, you know, having the war games to kind of put them in front of them and go, hey, look, here's what the likely scenario plays out, makes everyone easier to understand. But two, it's also a way of, you know, I feel actually much more comfortable and confident that, you know, this region is relatively

stable because of these wargaming. Because A, it just shows that none of these guys are really prepared to fight a long campaign. None of these guys are gearing up for a long campaign and none of them have shown really any interest in preparing for a long -term campaign. And that's probably

a good sign. I think almost all these militaries are, you know, imagining that they may end up in a border skirmish or they may end up in a kind of three -day, you know, you go 10 Ks back from the border and I don't think you'd notice, but generally it's been fairly, particularly on a conventional point, you know, campaign averse,

which is good to see. That doesn't mean that there is no chance of things going wrong, particularly when it comes to Kyrgyzstan and Tajikistan or Afghanistan, Tajikistan, but at least on a sort of massive conventional scale. No, I think that's a lot of things would have to go wrong before we get to that point. So no need to panic everyone. We can all have a deep sigh of relief. I've just put the... Sorry, the other point I will add quickly is they're also all incentivized to not

go to war. And this is one of the interesting things that I've seen from both sides is that everyone wants to seem like they're stable because there's more tourism, there's more added advantages. But when I do consulting for particularly energy companies and mineral companies that are looking at Central Asia as an option, the classic story I continue to see is, hey, we're currently mining X or Y in Sahel or in Central Africa. We're looking to diversify. We're looking to move our operations

somewhere else. And we're looking at Central Asia because we know that they are more stable, more reasonable. They don't tend to go to war and that I could build a mine. And because of the way that mining works, that quite often you lose money for the first 10 years, but you make a bunch on the back 20, long -term Central Asia looks a lot more stable. And this is where the money is coming from. It is coming into the region.

image of stability. So everyone is driven not only in themselves to keep things stable, but also even if let's say when Tajikistan and Kyrgyzstan have their border skirmishes, you can guarantee everyone in the region is calling them saying, please shut up. We're trying to be stable here. Can we all just knock it off a bit? So everyone is making significant efforts to try and make sure this region doesn't break out into a war.

And Mitchell just backed that up in the chat saying all the nations are very much integrating their economies and settling their border disputes now. All five presidents are now here in Washington together today. Thank you for that little addition there. We have so many questions in the chat about, there's a lot of people wanting access to your resources now and asking whether you have special accommodations for independent contractors working with academic or CIVSOC institutions,

for example. Do you have any? That is possible. I would obviously suggest you can email me and my details are all on those websites and we can have a chat about it then. That'd be the best way to get in contact. Fabulous. So you heard it here first. Please email the dedicated email with your application. to get access to the paid version. But it's great to see so many people interested in really delving into the data here.

I was just going to say, it's been fascinating to watch all the Central Asia people come out of the woodwork and see more interest in the region as time goes on. I think this is part of the reason I got into this project in the first place is that every other one of my friends who works on the Africa desks or in Europe or whatever, particularly in the South China Sea areas, they all have all of very similar resources they can go to, to look up where things are and

understand very quickly what bases exist. Central Asia just doesn't have anything. So it's finally good to have kind of a central resource I can put in the third screen, remember where each base is. Yes. And if you are in the Discord server and you're interested in this particular region, we do have a space for you to chat about it. It's called Crocuses Central Asia, and I've just linked it in the chat now and I'll put it in the Discord. I'll put it in this stage talk description

as well on the podcast. So please do hop in there and continue the conversation from today within there. That is the space for you. I know your expertise is in Central Asia, but you mentioned that your team also works across lots of different regions. And at the very beginning, In your speech, you were talking about how it's quite unique in Central Asia that you're able to kind of rank the militaries. How difficult is it to kind of take this model and look at other regions of

the world in regards to military force? What would be your advice if somebody is looking to do this in a different region? And what were the kind of key factors that you had to think about when approaching this particular project? I think the reason that it's a little easier to do here is because of the geography, because they are very similar economic stylings, they're

very similar geographic realities. If you're trying to, let's say, rank Australia versus Indonesia, and you were to say, you could either rank it as, hey, what are they capable of if we put them in a giant flat field with no terrain and just plopped all their military resources on either end of the field and said, go boys. And yeah, you could probably work out something of that variety, but I think that's a little unrealistic.

You know, if you're looking at something like Australia versus Indonesia, Australia could absolutely take some of the small islands, no problem. The Australian Navy is really, really capable, but the Australian ground forces are quite small because we don't anticipate that we're going to be fighting massive land wars anytime soon. So again, it just gets, you know, unless you're dealing with specific cases or you're trying to, you know, rank countries on their ability

to do a specific task. So let's say, if I was to try and figure out if Uzbekistan invaded Kazakhstan or Kazakhstan invaded Uzbekistan, who could get further? That would probably be Kazakhstan. But does that mean there are better military? It's tough to say. And again, because there's also going to be geographic realities that ruin a lot of this. Again, you put them in a plain field, Uzbekistan wins against Tajikistan every day

of the week. But because of the fact that there's such a narrow choke point approaching Dushanbe, that logistics aren't built up for this, that a lot of the, you have to cross certain river points that are quickly going to be occupied. It just makes it much more costly for Uzbekistan to go on the offensive and gives massive advantages that you wouldn't get if you were Tajikistan, if it was just kind of a plain map. I either recommend you try and rank things on a specific

metric. If you're looking at Air Force projection capabilities or supply capabilities, or how many tons a day can you put into the field? That's something you could try and measure. Overall, it gets a bit difficult. And again, I think another example of this is everyone, when they try and measure tanks, looks at the stats on the chart and says, oh, well, this tank can fire further than this tank. And I think it's a little bit

deceptive. When you look at the American tanks, everyone points to them and goes, oh, they're so fuel intensive. And that's correct. They are. But if you're the Americans and you have the American logistical capabilities, you're allowed to be fuel intensive. They built the tanks around doing that. You look at a South Korean tank and

you go, oh, it's so slow. But then you realize that the tanks in South Korea are mostly built to kind of hunker down and take on waves of advancing Paul Hyam and North Koreans, which that tank is kind of perfect to do that. Again, militaries build things for purposes and that will obviously change your rankings and how capable things are. And trying to match a, you know, take the American tank, for instance, and give it to Uzbekistan where they have a pretty average logistical capability.

It's going to effectively drain Uzbek fuel reserves very, very quickly because it's just not made for their purpose. Absolutely. And a really good example there. Obviously you spoke to us previously too. I've just put the link to your previous talk in there. So people can also listen back to that on researching Central Asia. We've only got a couple minutes. I want to ask you one last question. And this is based on the fact that we've had quite a few stage talks recently on

paramilitaries. And you mentioned the, you had that amazing graph earlier about the disproportionate percentage of paramilitaries in the different regions. Is it more difficult to map paramilitary power than traditional military power? And how do you kind of go about doing that? Well, I think it is more difficult to try and map paramilitary stuff because it's much softer capabilities.

You know, it's easy to map, try to look at the amount of tanks or look at the amount of vehicles or look at the amount of logistical capabilities someone has because a lot of it's, you know,

things you can point to. Whereas with paramilitaries quite often, how good they are at cracking down on a population or how quick they are at, you know, riot control, that kind of stuff is much softer and you really have to spend more time watching military exercises or watching, you know, trying to get your hands on video of these guys doing training exercises or interview people

at how quick it took them to do that. Or, you know, smuggled out videos of, you know, let's say Turkmenistan of the internal security services cracking down on a city, how quickly they can shut down the internet, that kind of stuff. it becomes a lot harder to measure and you're usually running off much smaller data sets, which will obviously complicate things going forward. But there are things you can piece together. It just becomes a bit harder, I guess, particularly if

you're trying to put a score on things. Thank you, Michael, for your amazing insight once again. As I mentioned, we have had previous talks on paramilitaries very recently. We had the New America group come in and talk about mapping militia networks. Recently, it's a fascinating discussion. Do listen back to that if you're interested in these kinds of conflict networks. I tell you, you would have had Candace and Ben on for that one. We did. Yes, exactly. They are

fantastic. I can't recommend particularly Candace's work more heavily. It's amazing. But in the meantime, please do check out these fantastic resources that we've shared today on Central Asia, particularly the new Uzbekistan report. And thank you so much for joining us. Michael, thanks for coming in and talking about your work. And we'll be back in two weeks time with another stage talk. Thank

you all. Thanks again for having me. And if anyone has any questions, feel free to email me and I'm always happy to try and answer any questions people have. But it's been an absolute pleasure to be here. and hopefully I'll be back when we release the next chapter of this thing in a while's time with Kyrgyzstan. Absolutely, we'll be itching at the bit to have you back on. Thank you for

listening to the stage talk. If you'd like to catch a stage talk live where you can ask the guest questions, join the Bellingcat Discord server by visiting www .discord .gg slash Bellingcat. The music you've heard is titled Dawn by Newer Self and is courtesy of Artlist.