Bringing integrated data to life

Speaker 1: Welcome to The World of Intelligence, a podcast for you to discover the latest analysis of global military and security trends within the open- source defense intelligence community. Now onto the episode with your host, Harry Kemsley.

Harry Kemsley OBE: Hello, and welcome to the first podcast this year from Janes. Sean, thank you for joining me again.

Sean Corbett CB MBE: Very welcome.

Harry Kemsley OBE: It's been a few weeks since we last spoke, there's been a lot going on. And I thought that we might start the year, Sean, by looking at the stuff that we discussed at the back end of last year, all the various topics, all the various theoretical things that we discussed, and maybe start to bring them into a slightly more practical focus. We have discussed the value of open- source intelligence. We've talked about why agencies perhaps are reluctant for a variety of reasons, good and bad, to actually take open- source intelligence on. But we've also, in the similar period of time, been working really, really hard inside Janes, our own intelligence agency, at looking at how we might better do the things that we do in terms of open- source intelligence. And one of the areas that's come up from our previous discussions and recent experience from Janes, is the ability that we now have to first in interconnect all of our intelligence, and how that is also helping with the integration of other people's intelligence with the change intelligence content. I thought we would get started with this shift from the somewhat theoretical to the somewhat more practical and applied, by looking at this topic of the value of interconnecting your intelligence. And then the ability to integrate other open and close source classified material to the sources that Janes brings and other people bring. I thought we would start there, Sean. I think you're familiar with what we've been doing at Janes. I know you and I have talked about some of this off camera as it were. What would be useful I think to get us started is just a brief reminder between us about the sort of things that came up about the values of open source intelligence overall, so let's just talk about that briefly in terms of what we remember from there. And then, why does interconnecting intelligence make it even more powerful, even more useful? Let's get started with the fundamentals first. Sean?

Sean Corbett CB MBE: Yeah, sure. I think it's a great approach actually, because last year I think we got into the psyche, not just us, but obviously a lot of organizations, got really into the value, the potential value of open- source intelligence. And there's a lot of really good think pieces out there. I think the message is in there. The next step now though is bringing it to life, in terms of what is the value and how do we optimize that value? I always used to think when I was working within the Defense Intelligence Agency, when I was talking to analysts, in the back of my mind was," What's in it for you? You're very busy. You've got lots and lots to do. So how do we help now within the open- source intelligence?"

Harry Kemsley OBE: Yeah.

Sean Corbett CB MBE: And that's two, as we've spoken about many times, it's two elements really, isn't it. It's the actual data itself and the richness and the strength, the applicability of the data, but it's also making it as usable as you possibly can. And those things are of course deeply connected. But there's no point in having single piece of intelligence or information that is relevant on a particular challenge if you can't then manage it, curate it, wrangle it, as we now know, great word, and then use it to actually deliver what you need to. I think transferring from that theoretical approach, which we have done a few times to be fair, into the practical, okay, how can we use it? What does it mean? And the interconnected piece I think is really important. You know that when I overheard the word or the phrase all- source analysis, it makes me flinch because nobody uses all- source analysis. It's multi- source analysis. I mean, almost by definition, you can't actually do because you don't know what you don't know. But you can't have an analysis today that doesn't include open- source if you want to call it a all- source analysis. So how do you bring that in? How do you interconnect it into using the Janes vernacular and then bring it to life?

Harry Kemsley OBE: Yeah, so I think that's where we're going to go. For me, my recollections of the various conversations we had last year about the value of open- source, it was as much about saving time for the analyst, where they could rely upon what they've received from the open- source environment as intelligence. It was about that provision of context that is sometimes omitted from the more exquisite, perhaps somewhat drinking straw view of the world you get from classified. And it was also about the gap- fill the priming of the pump, the fundamental intelligence. That foundational intelligence I think is what we called it in the end, where you can step into an arena, into an operating environment, you can step into a discussion around an equipment system for example, and not need to do all the homework that you would otherwise need to do because you can pick it up. Which goes back right to the beginning in terms of saving time. Let's then talk about when I take all of that intelligence and I start to interconnect it. And for Janes, that really means everything we've got now interconnected. So when I pick up a single piece of information from Janes, all that we know about it is available as you pull it. It all comes with, it's all connected, it's all interconnected, to kind of phrase. Why would that in itself be valuable, Sean? What's your view of why that is intrinsically valuable to the analyst?

Sean Corbett CB MBE: It's because you're trying to consider every facet of it, and it does depend on the nature of the question of course. Very few people are going to say," Right, we've seen this bit of kit on the battlefield. Tell me all about its dimensions and how big it is," and all the rest of it. They will want to know what its threat is, what can it do to me. But to understand that you've then got to say," Okay, well, it's going to be associated with this formation or even this other type of equipment." And that means that that is this threat. So it's joining it all up, and that can be from very tactical to strategic quite quickly. And I did actually look at one of your... Because I thought let's bring to life with a use case and some of your entire videos that you've been putting out there. I focused on one of them, a place called Rogachovo Air Base, excuse my Russian, but the Novaya Zemlya archipelago. It's basically on Russian's Arctic flank in the Baltic Sea, middle of nowhere, nasty place that you really wouldn't want to be posted. But I thought I'd focus on that just to try and bring a use case to life. So if that helps, I can start to bring it from one piece of intelligence and expand it from there, if that would help.

Harry Kemsley OBE: Yeah, that would be useful. And for the listener, what we'll do at the end of this podcast, Sean, is we'll put a link in for them to go and look at the video you were talking about for them to understand some of the specifics that you'll describe. Thank you.

Sean Corbett CB MBE: Okay. I looked at this place on Google Earth and it literally is a single runway in the middle of nowhere, very cold, very inhospitable. But there was a Janes report actually, open- source report, that talked about, in January'21, the establishment of a MiG-31, so Foxhound deployment, to establish what you and I would know as a QRA capability, quick- reaction alert capability. Now that sounds, well, so what, and that's really where you start looking at it. So the MiG- 31 is quite an old aircraft, probably goes back to the 1970s, but it has has a significant number of upgrades. It's still quite capable, as you may even remember, not that I'm saying that you were around in 1979. But very, very fast supersonic, particularly if you match it with a new, pretty capable air- launched hypersonic missile, the Dagger it's known, Kh-47. That gives a real capability. Now, then you start looking, okay, so why would they potentially deploy that there? This base used to be a staging point for strategic aviation when they're were doing their exercise for the nuclear triad. But it's generally, there's no permanent ORBAT there, so why would they do that? You then start looking into the intelligence, and then you start seeing that, well, actually prior to that, in July'19, their strategic sound batteries, the S- 300s or SA-20s, I don't know them, the Gargoyle, were replaced with the S- 400s and upgrade SA- 20 Growler. Which have increased range and mobility, far more sophisticated radar and commander control. Then you start thinking," It's not just an exercise deployment, there's something going on here." Then you start doing the digging. And if you interconnect just even those two, you start looking at... That represents a significant increase in the Russians A2/ AD, anti- access/ area denial capability, in the Barents Sea region. Now that then starts to think," Well, this is more than just either tactical or even some operational stuff going on." You start then doing even more research and you find out that in that whole domain, including the maritime domain, they've increased their presence of nuclear- powered submarines, frigate landing ships, service- to-air missile systems, anti- submarine warfare, all of that sort of stuff. At the same time, there's quite a big building program going on. If you then look at the command and control, which is always a really good sign that either stuff's happening or they've had to recast themselves. Without being cynical, don't organizations always rename their commands and develop them when it's all going horribly wrong? In this case, I don't think that is true. What was quite a small command, in early 2021, they came under what they then renamed the Arctic Joint Strategic Command, which basically started to bring all forces under command, rather than just either the Northern Fleet or independent. And then that was again in'21 consolidated with other commands to become the Northern Military District, which has put it on a power of other military districts, with an expanded role of a command arms strategic formation. Now, if you start looking at that, suddenly you're talking about a lead capability with increased, not only presence and capability. And so the punchline of this really, and I apologize for going on a little bit long, is I've gone from a single aircraft deployment into a realization that Russian is getting serious about the High North, as I would call it, or the Arctic.

Harry Kemsley OBE: Right.

Sean Corbett CB MBE: And so it allows the analyst to just go straight through all that stuff and do it quickly. And I did this the long way, I'm old. That took, just that little bit of research, and I recognize it's superficial, took me probably a day and a half. When really, as an analyst, what I should be looking at is the threat capability, plus intent, plus opportunity. If you can automatically do the capability there and then that gives you, the analyst, the time to go," Right, what does this mean? How important is it? What does this mean to us? And what is the threat?" So that really is the bottom line. So this is what I'm thinking in my mind about one element at least of interconnected intelligence. You got one piece of kit, you connect it to others, and you start connecting it to formations and even other domains. And you come up with a," Wow, this is significant."

Harry Kemsley OBE: Yeah, that's a really good example. If I just abstract out of that the bits that, now with Janes Intara we've interconnected, the Rogachovo Air Base analysis you've just done would start with the platform, the weapon system. Then with the events data that's now being interconnected with all the other data we've got, you would start to identify the arrival of the shipping and the changes of command you talked about. So the OrBAC would automatically start to be displayed to you in its changes. And very quickly therefore, by simply picking up the new aircraft at that location has arrived, you would move through the various elements of the Janes content in Jane's Intara. And you would immediately get to the point where you have all that foundational intelligence, so that you can return quickly to the analysis that you've done and the deductions and insight you've created from that analysis. You took a day and a half. It'd be fascinating to do that same process with Intara. Which, given that it's already interconnecting and you're not weaving in and out of various data sources, various data content, you would find it all interconnected for you automatically. I'm sure that would be an interesting comparison. A day and a half by a very experienced analyst, put the same problem in front of an inexperienced analyst, see if they come up with the same sort of conclusions. Or at least the same foundation intelligence much, much more quickly than you're able to do with the various sources available to you. So that's a fascinating example. Now, before we step from that, one of the things that has certainly become apparent to all of us is that when you look at all the publicly available information, the commercially available information, and then of course the classified available information, these three massive buckets of information. Whereas we may not ever get to the panacea of all- source for the reason you said earlier, even multi- source now is becoming a utterly enormous task for the analyst, just to manage the information, to synthesize it down to something that allows them to see the foundational floor. How do you think, in that context of the publicly, commercially and classified available information, interconnecting intelligence and then being able to integrate it, what's the advantage there? Why would that be of advantage to an analyst?

Sean Corbett CB MBE: It's a really important point actually, because you are getting that... And it's the word that you use quite a lot and it's really been brought to live, we mentioned it as well, it's that context.

Harry Kemsley OBE: Yeah.

Sean Corbett CB MBE: Contextualizing the information by bringing all most multiple sources in. And some of the stuff that I was looking at, there were actually Russian government announcements, some of it was a little bit of imagery. I don't just mean commercial satellite imagery, I mean, stuff on the ground. All that sort of stuff that an analyst, right now working within the intelligence community, probably wouldn't consider. Now, that's not to say that they're not going to come to the same conclusion, but they might not come to quite the same conclusion and certainly not in the same time period. But I absolutely recognize, and nobody is saying, I always reiterate this, no one is saying that open- source intelligence will or should replace the classified stuff.

Harry Kemsley OBE: Sure.

Sean Corbett CB MBE: Because there is some really rich stuff. There's no way on this earth that commercial world would be able to replicate. Although there are times when I do wonder how much you could get out of open- source if you did the full spectrum of it, but that's a discussion for another day. So that contextualization is really important, and the gap filling. And I go back to the fact that intelligence is called intelligence, not information because there are gaps, that's the whole premise. One of the explicit tasks for an analyst is to try and fill in the gaps.

Harry Kemsley OBE: Yeah, for sure. crosstalk-

Sean Corbett CB MBE: Sorry. I was just going to say with... And you're not always going to have that level of assurance, another word that we use quite often, that you want. But by layering it and have as much of it as you can, you can cross reference, you can validate and come up with a better idea of the solution.

Harry Kemsley OBE: Yeah. You've actually gone exactly where I was about to jump in there, that cross- referencing. The ability to map one set of data onto another set of data so that you know you're looking at the same thing, is one of the many problems you've got when you're trying integrate data. When you have, as Janes Intara now has, a stable and standardized data model that allows you to be confident, the definition of that piece of equipment, that all that, whatever it is, is consistent. Mapping data together so you can get a consolidated foundational picture, is where you can really begin reduce that significant data management and data wrangling, to use that word again, burden that's on the analyst. They haven't just got to find the data and verify it, they've also got to collate it and they've also got to integrate it. And that's another area that certainly Janes has seen a great deal of success for itself. And now increasingly for our customers, in terms of the ability with Janes Intara or to do that integration,

Sean Corbett CB MBE: There is a sort of subtle element to that as well, and what I found in doing this exercise, was that you were going to get multiple versions of the truth.

Harry Kemsley OBE: Right.

Sean Corbett CB MBE: Looking at the command and control piece, it was really quite confusing because there'd been so many changes over so many period of time. There were so much written on it, obviously in the open- source domain, some of which was clearly messaging, but some of it was good stuff, that there are slight nuances and slight differences. And actually getting to the ground truth in terms of what command transferred into what, and then who was subordinate to what, was completely difficult and I didn't get there. Well, not fully there anyway. But by interconnecting it, you can come up with that wiring diagram, which we all love in the military. Very, very fast, without having going necessarily through that," Is this more important than that? Or does that cross- refer to this or is it contradictory?" That really helps in the timeliness for the crosstalk.

Harry Kemsley OBE: Yeah. And as you say, when you've got a piece of equipment and you've got an ORBAT, and maybe you've got a organization that creates said equipment, the company that builds it, when you have an intelligence team like Janes does, that is responsible for managing ORBAT intelligence or equipment intelligence, et cetera, all the different intelligence groups, they've all got to be convinced about how they interconnect. You've all got to find a way of making sure that when I introduce this piece of equipment in this location, which it hasn't been in before, and the ORBAT is changing, all of that interconnectivity that is required demands those teams within Janes do the interconnections reliably and with a standard and stabilized way, which is our data model. That is why, when you pick up said initial piece of information, the aircraft at a location, and you pull then immediately into the associated capabilities like weapon systems, and then the events data bringing together all the other arriving naval equipment and the changes in C2, command and control organization, all of that being interconnected demands that it is stable and standardized and usable, quote assured, unquote. And that is what Janes Intara is doing for our customers. Now because time's going to evaporate on us if we're not careful, let me move on to a slightly different part of why Janes Intara and his ability to interconnect is so important, which is something we've touched on but I would just dig into it a bit further. I think there was another scenario within the videos, I think it's a Gulf maritime activity, where you start to see the benefits of interconnecting, not just Janes content, but other third party open- source intelligence. And you can start to create very sophisticated indicators and warnings with quite a substantial amount of situational awareness information as well. Have you seen any of those videos, Sean, where we've started to get crosstalk-

Sean Corbett CB MBE: Yeah, I have actually.

Harry Kemsley OBE: Yeah.

Sean Corbett CB MBE: I did have a quick look at that. And what that brought out to me, as much as anything, is that, and I wouldn't yet advocate using the sort of information we've taken out in real time.

Harry Kemsley OBE: For sure.

Sean Corbett CB MBE: But in terms of planning and in terms of dependencies. So when you're starting to look at commercial shipping routes, for example, for ports that are dual use, for flows of, very topical right now, energy. All that helps to bring that picture into so the analysts, once again, can see very quickly about what's going on and come up with a so- what for it.

Harry Kemsley OBE: I agree with that. But I think the piece that I was particularly interested in is the fact that when the analyst pulls the Janes data through Janes Intara, and they may already have on their system HawkEye or other intelligence feeds, the ability to map it together gives them insights at that foundational level that allows them then to determine that," Actually there's something here I need and go and look at." The great amount of information that's now available in the open- source environment, and it's of course not just Janes. But the fact that Janes Intara can enable you to bring that in, co- locate it on a map even, and just see how these things are starting to merge, does give a level of situational awareness from an open- source perspective that is very, very compelling. Now, if I do that with a commercial source outside of the vault door, and I bring it all inside the vault and I superimpose upon it the classified, then I think it becomes extremely powerful. And certainly our early forays with customers, in the earlier adopter work that we're doing with a lot of these customers, is starting to show exactly that, the ability to build a very sophisticated foundational intelligence set.

Sean Corbett CB MBE: You've touched on some really good points there that is probably a conversation for another day, but looking at the multi- source in the commercial domain, you're absolutely right. You know my feelings on that. You've mentioned things like HawkEye 360. You've got some of the commercial satellite imagery, the electro optical and the radar now, which is really good. You've got the RF stuff, but you've also got the almost documentary stuff as well-

Harry Kemsley OBE: Yeah, yeah.

Sean Corbett CB MBE: ...that's just written in there. You bring that all together. That is really good multi- source. And what that does is that layers, it validates all the rest of it. The key for me, and the key question which I'm not sure we've got there yet, is at what stage do you integrate it into the secret sauce that is the really good stuff. Do you do it on a single- source basis? Commercial SIGINT as I call it, the RN stuff and that sort of stuff. Do you bring it into the SIGINT world to give them a consolidated SIGINT view and all that brings, and they can equal it with the imagery and the open- source intelligence, operational stuff. Do you do it at the source level and then bring it together? Or do we, as a commercial community, bring it all together and say," Hey, we've got the answer for you. Now integrate that into your answer." And that gets into some really quite difficult things in terms of policy, rivalries. I do, again, not being rude to my colleagues because I love the organizations, but most single- source organizations now, intelligence organizations, are not single- source at all. They bring in other pieces of intelligence. We know they do and they're a little bit competitive. So what would they bring in to able to bring their stuff to life and make it more real? That's a huge question that I guess we can have a philosophical debate on it and we probably do need a debate on it. But I guess the more things are integrated, the more we start using cloud- based computing, the more that discussion becomes less relevant. And as a side, I think for the intelligence community, that's something they're going to have to wrestle with sooner rather than later actually, is that is it still relevant, and I can't believe I'm saying this, to have those delineations of the SIGINT, the IMINT, all those sort of stuff. Now of course there is, because there are special techniques, there are special collection capabilities where you have to have that expertise. But at what stage do you bring it together to come up with that integrated approach?

Harry Kemsley OBE: That is a philosophical question, but it's worth just exploring just a touch because for me today, you go back to things we discussed before Christmas, where we talked about cultural bias and cultural resistance to the open- source environment. Because it's all just noise, it's all just unhelpful noise. However, when you move into the open- source intelligence realm where you're dealing with highly relevant, accurate, and I think timely intelligence, open- source intelligence, being that if it is assured, then it is appropriate to bring it in, I would suggest, at quite an early stage. Now, perhaps given the cultural barriers that we've talked about in the past, we ought to be no more ambitious than to say to the analyst," When you are considering a novel problem or even an existing one, if it's novel, you can get a foundational set of intelligence that you don't have to go and build for yourself." It saves a huge amount of time just doing that. But if you already have some foundational intelligence, it can also give you a very strong triangulation, a confirmation, of what you're looking at being true and, or accurate. One of the things that I've seen done many times in that gap analysis that you're doing, trying to understand what you know and don't know, is filling the gaps can be done very quickly from open- source. If it's assured, if it's high quality. If it's not, then yes, it helps, but it's never going to be done at a high confidence level and therefore you are not going to save as much time. So it's important you go for an assured source. But to my mind, Sean, in short, I think you would do bring it in early and allow it to be foundational, rather than to bring it in later and then have all the complications of integrating it into a more complicated picture you've already created.

Sean Corbett CB MBE: I would agree with that 100% for now. Back to the trust issue. Because at this end of the day, our data needs to be trusted. Commercial data needs to be trusted before it's ever going to be brought into it. And that's another philosophical debate for another day, because some of the best assessments I've seen are based on low confidence data, almost by definition that we don't know what's going on. So you need somebody who's very clever and knows their business to bring them together and come up with," On balance, I think this." Now that's not a game that we want to get into yet because we are really about the foundational level stuff. The actual data, the efficacy of the data is very, very important to us. Do I see a time when we might be feeding stuff in? Yeah, absolutely. That is not assured. But I think, no, I know and I know you're working on it, we would have to characterize that in some sort of way that,"Look, we haven't got a high confidence level in this, but actually it's worth looking at anyway." Now that's some way down the line at this stage and I'm advocating that for now.

Harry Kemsley OBE: Now I think the likelihood of Janes putting out raw data is not near term, but putting out a data stream that is less assured is certainly something we're looking at. Because we know that there are things that in the context of open- source are valuable if they're available to you earlier. Because that low assurance data feed, information feed, to the analyst at just the right time, who has the background, who has the exquisite capabilities at their fingertips as well, that might just be a tipper that really moves them forward. And it's not that Janes will always only want to be right rather than first, it's about being right and helpful about optimizing what we do to be more available to the analyst. Even if it means at the early stages of what we're feeding them, that it's a low level of assurance. And that's certainly something we're looking at. It would be remiss of me given our roots, going right back to 19th century, when you think about Janes use as it was then at the basis of naval war gaming, for me to not mention the other scenario that I know seen on the entire videos, which is around the operational tactical planning, the red teaming work that's been done with Intara. And the reason I bring that up is because one of the other aspects of Intara, Sean, as you know, is the fact that it is agnostic to your tool set. It doesn't matter to Janes Intara that you've got one particular set of analytical tools or the other, it will be integrable. And the scenario I'm talking about, which you might have seen, is the South China Sea scenario, where we're looking at a series of threats and prioritizing those threats as a red teaming environment around a naval vessel. Have you seen that video, Sean?

Sean Corbett CB MBE: Yeah, I have. I have actually. And although it is a real time scenario, it is a training synthetic environment.

Harry Kemsley OBE: Correct.

Sean Corbett CB MBE: And that is perfect for the Janes data. Because Janes data, that fundamental stuff, everybody values it and uses it because it is the standard. No question about that. So by feeding that in where there's no assumptions, there's no tweaking should we say of it, you can do very, very objective," Right, what is the..." Back to my threat equals capability plus intent plus opportunity. If that capability's there, you can then start," Okay. So what have we got to counter that? What do we need to focus on? What is the greatest threat," which comes into the intent. So no, it's a good example actually.

Harry Kemsley OBE: And it's also one that's on, I think it's ArcGIS, the Esri geospatial platform. Again, just to make this point that Janes Intara is ambivalent about the tool set you've bought already and installed, it will work within it. That's why the way its been designed. Sean, I can see that time is starting to tick away from us, so let me just bring this back to where we started. We came into the new year, having had lots of really interesting conversations, many of them arguably somewhat theoretical. The feedback we've had from people who have sent us feedback is they would like us to get more applied, more specific to real world examples. And we've done that today by looking at how Janes Intara is allowing Janes to do a series of things that it couldn't do for itself and for its customers, in terms of how it interconnects and integrates and works in different tool sets. The Rogachovo Air Base case study, the maritime activity in the Gulf case study, and the other one I mentioned, the South China Sea, they're all available as videos. We'll put a link in the podcast notes. If you were to walk away from this conversation, Sean, with just one thing that you wanted the audience to understand about the power of interconnected intelligence and the ability to integrate, what would that one thing for you? Given that you've done some work recently on for a day and a half, what's the one thing that would for you be the takeaway for interconnected intelligence?

Sean Corbett CB MBE: I think it's the applicability of it actually. You've gone from one small tactical thing. You can very, very quickly go up to the strategic by using the tools that are available. So you don't need to do that mandraulic stuff all the time to get to where you should be doing. So in the Rogachovo case, and I missed my punchline really, which was that, so what does all that mean? Well, that means that the Russians are, for obvious reasons, increasingly concerned, active in the Arctic because of the trade route that are opening and all the rest of it, so we need to pay attention to that. What that interconnected intelligence did was allow me quickly to go from, or, and it would've been significantly quicker if I hadn't done it the manual way, from one specific event to," Right, what does this mean?"

Harry Kemsley OBE: Yeah, perfect. And so fundamentally it's a huge amount of time saved, but the time's only saved, which is my takeaway, if the information and intelligent using is assured and high quality. Good. Well, in coming weeks and months, Sean, we have a range of fascinating podcasts in front of us, which will be increasingly applied and real. I think the next one we were planning to do, Sean, is it the violent extremist organizations and open- source contribution to?

Sean Corbett CB MBE: Yes, that's right. Yeah, we've got a, a distinguished colleague, who briefs at West Point, actually talking about violent extremist organizations and how open- source intelligence can help to look at that particularly challenging problem set.

Harry Kemsley OBE: Yeah, very good. Sean, as ever, thank you for your time today. Great to speak with you as always, and I look forward to the next time. Thanks for your time.

Sean Corbett CB MBE: Thank you.

Speaker 1: Thanks for joining us this week on The World of Intelligence. Make sure to visit our website janes. com/ podcast, where you can subscribe to the show on Apple Podcasts, Spotify, or Google Podcasts so you'll never miss an episode.