Is it time to redefine OSINT? - part two

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: Hello, and welcome back. For those of you who listened to part one of this podcast, you'll know that we're about to pick up the second part now. Thank you for listening. Sean, I can see you're leaning forward. Before we do that, though, I'm just going to ask you, Joe and David, to start thinking about, okay, so if we accept that the concept of OSINT as a separate taxonomy and lexicon of words around the world of intelligence, how do we start to steer this oil tanker back towards where it should be? How do we start to get the community to start correcting back to where you think we should be? So, while you're thinking about that, because that, by the way, will be as much a cultural discussion as anything else because there is a religion out there called OSINT now, but come back to that later. Sean, your thoughts.

Sean: Yeah, it's really coming back on the training and education bit, and I understand why they are recruiting OSINTers and there's cultural, but it's also process reasons for that. So, just as any INTer, so as a SIGINTer, so I used to look after linguists, you've got to be trained pretty specifically to be a good linguist. So, that's a specialist training piece. IMINTer, I am one. GINTer, as they call them now. You've got to be pretty well- trained to do that as well. And I would say for an OSINTer, then you've got to be trained on certain ways of scraping the data that's out there in whatever means. But the fact that it is publicly available but might be hidden, which is back to your point, Joe, that you have to be trained to do that. So, it does require a certain specialization if you like. And the other thing that we haven't talked about and we haven't got time to talk about is the whole classification rigmarole, is there are people that you could employ, and James I know does employed, who are absolutely brilliant at what they do, incredibly insightful. And they both got that balance between the technical understanding about how to manipulate the data, which is the other piece, and then, doing the so what out to it. Now, they, because of their background, would never ever be employed in intelligence organization. They just wouldn't get the clearances. So, that gives you a little window about, okay, I do understand what you're saying, but you can't just exclude them, say, " Right, we're not going to employ OSINTers," because you then, as the IC does anyway, you exclude a whole raft of people who could add huge value. Now, just to bring this a little bit towards ahead is that we had a very good discussion recently in the States, but where it's become, or it's been said to us that they are about to within ODNI, produce or certainly work on providing a paper of standards for OSINT. Now, you could say, " Hang on a minute, why is it not just IC standards?" But it is going to be IC standards for OSINT. So, that's a start that really is in terms of recognizing that there needs to be some sort of formalized tradecraft and acceptance of the open source area of what we do, but we've got to formalize it and make it as usable as the classified stuff. Because you're absolutely right. I do sometimes wonder in the human world, whether they do the same checks and balances as some of the journalists, for instance, and not all journalists are the same, to actually get to the ground truth. Harry's...

Harry Kemsley: I'm wincing only because... It doesn't matter, keep going. inaudible.

Sean: Yeah, okay. So, I guess my punchline is that I accept fully that we need to get beyond the discussion of OSINT. But back to the practicalities of it, and I guess this is my punchline actually, Harry, is that if you look at the progress that's been made just in the last 18 months in terms of even talking about open sources to be used in the intelligence community, because although it has been used for everybody, no one really talks about it and it's always a bit dirty, and we don't say, " We've actually used OSINT," even though the people do. So, that discussion has gone from there. We'll use it if we have to and for the intelligence sharing, which is probably another whole podcast for another day, all the way to how do we make open sources the data of first resort to use, which is effectively all the same? That discussion is live, but when you've got big organization that's talking about, okay, how do we standardize it, how do we make it legal and all the rest of it? So, again, I was just reading inaudible there's new intelligence community policy framework for commercially available information published in May this year, which provides direction to the IC, what they can and can't use and the legalities of it. Now, that's really positive, which we couldn't have had a year ago either. So, I guess it's a long way of saying that, " We'll get there by osmosis," I think. If we try and push things and do the bigger bang approach, it never works in the IC for all the reasons that we've said both in this episode and before. Culturally, policy, ownership, money, all those things is that we're not going to go dead air overnight, right? Forget OSINT, doesn't matter. We'll just do it all this way. But we are driving to a place where the use of publicly and commercially- available information is normalized, and I think that's actually happening now.

Harry Kemsley: Yeah, okay. I agree with all that. I think what I've heard then, and I'm going to bring us back to that question I posed before Sean spoke, Joe and David, what do we need to do if we need to correct something? What do we need to do to start correcting it? I think what we've said in this conversation to date is that if the concept of OSINT has been useful to focus the mind, to get people to accept that there is real value out there in public sources that previously wasn't available to them because of technology or whatever, then if that is becoming now redundant or counterproductive, how do we start to correct it? I also note the point, David, that you made about the economic reality. If students are coming to master's programs because they want to have it on their resume, their CV, that they've got the skill that employers are putting in their job adverts, that's a reality too, right? If I want a job and it's an OSINT job working for Morgan Stanley to assess risk, I want to show that I know how to do that from a master's program that I've done at King's or wherever. However, if we now focus in on the federal environment, the government environment, the agencies, which is where ultimately, national security matters are addressed, is there a sufficient concern in this concept of OSINT that we need to correct it? And if so, how do we start to do that? And Joe, in fact, no, let me come to you first, David, how do we start to correct this?

David Geo: Yeah, well, I know Joe's going to mention Wittgenstein again, so I thought I'd just get my more basic thought out of the way first. But actually, I wanted to just talk briefly about a paper that Joe and I wrote a couple of years ago during the pandemic, because I think there were a lot of lessons in that paper that the pandemic was the... No one is going to address a problem in the government absent a crisis. And the pandemic was a crisis. And I'll just give you a little origin sort of story on it. Remember when everybody was having Zoom happy hours and early on in the pandemic? And I'm like, " Well, yeah, I was happy not seeing you in person, but now, I have to see you on the screen, but whatever." And so, my wife and I were having a glass of wine on a, I don't know, Tuesday night with a friend who's pretty senior at CIA. And she and her husband dialed in from their vacation cottage in West Virginia. And I said, " What are you guys doing in West Virginia? It's Tuesday night. Don't you have to work?" And she said, " Hey, it's the pandemic. Can't come in. Can't work." And I don't know if this is a family- friendly podcast or not.

Harry Kemsley: It is.

David Geo: I was like, " Heavens to Betsy, no." Right? I think I might've said something different because just because you can't access classified systems doesn't mean you can't perform the function of intelligence, which is to narrow the cone of uncertainty for policymakers or decision makers to aid in statecraft and foreign policy development. And this whole, look, if I can't access the high side, then there's nothing I can do, just drove me up the wall sideways. And so, what do I do? I get on the horn with Joe and I was like, " Look, I'm mad." And Joe was like, " Okay, let's talk about it." So, we came up with this paper asking if intelligence analysts could telework. So, to me, this was the real way to crack open the egg of appreciating OSINT because OSINT can come in and save the day. Because if you have most policymakers, their questions could probably be answered to a pretty good, if not perfect degree, simply using unclassified resources.

Harry Kemsley: I'd say Jane's completely agrees with that, by the way, just putting it out there.

David Geo: All right, well, there you go. I want my free T- shirt now. It says, " Jane's on it." So, what do we recommend in the telework paper? We say, " Scope your intelligence requirement. Do it in the SCIF if you have to. But then go away. Go to the library, go home and use other resources to generate the answer to that question. And then, after you've been away for, I don't know, two days a week, however long it takes you to get that first draft, and then, you can come into the SCIF, fire up your secret systems and answer that question using classified sources." And if they are complementary, well, that's wonderful, right? You have a good check. You have a warm fuzzy. But if the secret sources and the unclassified sources are saying something different in your analysis, oh boy, right? Then, we have a problem. Good to answer that problem now before we go gallivanting, acting on this sort of intelligence. And so, I'll turn it over to Joe, but I thought that the first way to get into it really was by embracing telework and just getting out of the SCIF and showing... All right never mind. I'm not going to turn it over to Joe. I got a nice message from a friend of mine who used to teach at the National Intelligence University in the States, and their whole shtick is that they can have education in the classified realm, so they can use different case studies, et cetera. And he wrote to me during the pandemic, and he said, " Dave," he said, " I didn't believe that you could do intelligence education outside of the SCIF because I really thought that we needed these classified vignettes and all that sort of thing and these records and these files." And he said, " But we got told to go home and teach your exact same syllabus except using unclassified cases." And so, he said, " I think you're 99% right. There's still a couple things that I wish we could get in the SCIF for," but he said, " Yeah," he said, " I didn't believe it before the pandemic, and I was forced to teach intelligence studies in an unclassified manner." And the point is, is that how do you reorient the ship or whatever you want to say? It starts with the crisis, in this case, telework, that I hope we don't lose the momentum that we had during the last couple of years.

Harry Kemsley: Thank you, David. And Joe, I'm coming to you in just literally one second, but I think now, what I'm starting to see is a picture, which I'm going to hand over to you, Joe, to think to answer to is it is a continuum. There is intelligence value in the open source environment. It is going to be important for a variety of reasons, context, early warnings or sometimes, your only source of information to act upon. And therefore, if an organization for cultural reasons, lack of education, lack... They must be indoctrinated if necessary on the value of intelligence value derived from open sources. I'll avoid the OSINT, but I'm just going to call it that for now. So, if that's the case, Joe, what have we got left to fix? What is in front of us now in terms of correcting our use of language or correcting our concept or correcting our trade craft? What is it? Or is it all three?

Joe Hatfield: It's a bit of all, all three, but I will say since Dave mentioned our telework paper, I want to be clear to the audience that our thesis within that paper, it does not fly in the face of everyone's practical observation as especially somebody like me who spent almost all of my career doing military intelligence supporting combat operations. There are certain facts out there that you simply cannot acquire via open sources. You are not going to find the location of a Russian submarine through open sources in most circumstances, right? But we may have some exquisite sensor somewhere that could help us do that and so on. So, we are not at the strategic level. That's really where Dave's comments about it being open source information being equally or sometimes, in some cases, even better than some of the classified stuff. That's where that really comes into play. But when you get down into the tactical level, when you start talking about troop movements and things like that, that could very well be a case where open source may falter. Although, as we've seen in the Ukraine war with Russia, we've seen that in some cases, people on Twitter are tweeting out locations of Russian divisions and things quicker than people are in the SCIF who have access to these sources. But just to be clear that our view in the telework paper is not at all inconsistent with the value of classified intelligence. We're not trying to devalue at all. But to get to the broader point, Harry, that you asked, I do agree, obviously, with Dave's points since we did co- author that paper, that as kind of a, I guess I would say it as a half measure, as a transitionary phase, telework is a way to take the crack addict out of the crack house and recover a bit. You don't want to ask the crack addict to rehab in the crack house. It's a terrible analogy, sorry. But in case the listener has-

Harry Kemsley: Listen.

Joe Hatfield: Yeah. But in the grand schema thing, and this gets back to Sean's deeper question or a deeper point, I think as consistent with my thesis, I think that what we need to see as our goal is that HUMINT tradecraft and the teaching of HUMINT tradecraft or IMINT or whatever, MASINT or anything else, these tradecrafts ought to include, okay, there are going to be sources that are going to be unclassified. How are they kept separate? How are they, in terms of, as Sean mentioned, in terms of handling, protection of sources and methods? But it ought to be part it. So, the tradecraft manuals, the actual CIA at the farm, at NGA teaching GEOINT to young GEOINT analyst. They ought to be able to have a similar tradecraft doctrine, book, set of procedures, et cetera. And then, it forks where it needs to fork, both being GEOINT that is special handled and GEOINT that is not, and I think that that actually reinforces the special handling requirements that we already have within the classified world. All of us in the classified world are read into sub compartments or special categories that are handled via different systems. Only certain people are read in. There is a paradigm of thought for all of these people to, okay, in this circumstance, because of the origin of that sensor or because of the delicacy of that human source or because of whatever, we have to handle it with special handling caveats. These are marked on the classification lines, on the top of reports. There is a paradigm for special handling within the intelligence community. That's the whole point of compartmentalization. To me, open sources ought to have that kind of thing where, hey, this is openly available, so it's handled differently than this other stuff. I think that fits well within. And so, we can get rid of the notion of open source, which is incoherent philosophically. It's got us into all kinds of problems. I think it's served a very valuable tool. It's brought the attention of the entire intelligence community to this vast array of information. Now, we need to dehomogenize this massive thing called OSINT, bring it back home where it really belonged. It served its purpose, it needs to come back home. And yes, as Sean said, we need to within the HUMINT or the SIGINT or whatever, within those disciplines, of course there's going to have to be special handling distinctions between sensitive sources that require protection and those that are openly available. But by doing this back in that way, we don't fall into the pitfall of hiring people that I'm an expert in OSINT. What does that even mean? We don't have situations where we are not applying HUMINT method, tradecraft methodology to human sources that just happen to be acquired by the New York Times or whatever. And so, you get a lot of philosophical and practical synthesis and methodological simplicity by adopting the proposal that I argue in my paper.

Harry Kemsley: Very good. Now, we've been running over these two episodes now for a chunk. So, I'm going to start to draw stumps, a reference to a part of a game in the UK called Cricket for those of you that aren't familiar with that phrase. Pull stumps means it's coming to the end of the game. So, Sean, I'm going to come to you first, David, Joe there afterwards. I want the audience to have one takeaway from these two episodes, and that's the one takeaway you want to keep absolutely locked in their mind from this whole discussion. I'll go last, which probably means that the two or three of you'll come up with two or three things I was going to say, but that's okay. So, Sean, for the benefit of the audience, what's your one takeaway you want them to understand in answer to or in reference to this point, there is no such thing as OSINT? Sean.

Sean: I would say that in terms of terminology, it doesn't matter what you call it. You call spot for all I care, but we have to regularize, normalize, integrate and use all sources of information, data out there to come up with the so what and what if, the narrowing of cone of uncertainty, which I just love that, by the way, because that's what it's all about. We owe it to the decision makers and the people that we are working for to say, " Okay, based on everything I know, based on some really good tradecraft and my experience and everything else, this is what I think because of what." And give them the confidence that based on everything that's available, I'll take that. So, that's my takeaway.

Harry Kemsley: Thank you, Sean. David.

David Geo: Yeah, let's go back to the crack addict and the crack house. That's where I'd like to leave people, because you need that forcing function. You have to put them in rehab because the SCIF, we have this cult of secrecy and we love it. And as Joe says, there are exquisite systems and exquisite capabilities, or I'm a former case officer. There's exquisite sources that are absolutely irreplaceable for intent, just for one example. But I'm very hesitant to treat OSINT as this other thing that's not on your team. Okay. We want the IMINTer, we want the SIGINTer, we want the HUMINT analyst. So, I'll put them on the team. And then, the OSINT guy shows up and we're like, " Oh boy." Or OSINT is itself, puts on its jersey and shows up on the team. Or we're like, " Well, we don't really recognize you as playing the same sport that the rest of us are playing. Why don't you go play over there?" And that assumption, mistaken as I think it is, and I think as Joe thinks it is, I think the US government is pretty close to at least considering making another agency to have, I don't know, the 57th US intelligence agency this time dealing with OSINT. And we have to bring it into the all source tent, if you like. And so, I am a little concerned that we're on the verge of making a mistake. And where I would leave people with, because I'm a lover of cakes, I think that right now, the ratios might be a little bit off, but if you just picture a cake, most of the cake is secrecy that the US government provides, and I think the UK government is the same. And then, the frosting is the unclassified insights that just get lovingly sprinkled on top. And what you have to do is, I think we have to get our heads around the idea that actually, you can answer most intelligence problems with the open sources or the unclassified sources as the majority of the cake, the fluffy bit. And then, using the exquisite stuff, those exquisite sensors that Joe was talking about, or that exquisite source in the Kremlin or whatever as the special sauce, as the magic sauce that the taxpayer rightly can say, " Hey, for$ 80 billion a year, whatever it is, I want some magic sauce to really provide the best insight we can to, ultimately, the customers that we serve."

Harry Kemsley: Very good. I'm loving these analogies, by the way, gents, the crack house, the cake sprinkles, the bugs under the ground, et cetera. So, Joe, on that vein, your final takeaway for the audience?

Joe Hatfield: Well, my final takeaway is there is no such thing as open source intelligence. We need to stop pretending that there are no... It's a junk drawer concept. We need to stop pretending that there aren't knives and forks in the junk drawer. We need to retake... They might be shaped a little differently, but they belong in the cutlery drawer. They belong with the other knives and forks. They ought to be used by the same people that use those knives and forks. And we have to stop creating agencies that are experts in the junk drawer and analysts that are somehow experts in everything from the hammer to the rubber band to the paper clip. We need to understand that, yes, by creating this idea of OSINT, we've paid attention. Because that junk drawer's gotten full of a lot of cool stuff. And we have directed the attention of intelligence people to the contents of the junk drawer. But that doesn't mean that... Once we've opened it and looked around, we need to make sure that we can bring all the power and expertise and tradecraft that we typically bring in all these other drawers to the contents of that drawer. And I think that's best done by emptying the drawer, putting it back into the categories and getting rid of this notion of OSINT, which I think causes all kinds of problems, both philosophically as well as practically.

Harry Kemsley: Thank you, Joe. I'm going to give the audience one further takeaway, which I think captures most of what's been said. I'm taken back to the early part of this century when the world seemed to want to focus on this thing called effects- based operations. I don't know if you guys remember that. Everything was about understanding the effect you could generate, and you had to analyze effects you were having on the battlefield, which for me, was absolutely junk drawer material because nobody could actually understand the effect you were about to have. You could potentially do it in retrospect, but from the examination of effects came a whole series of analysis, of kinetic, non- kinetic activities and so on, which was useful. And for me, what this conversation feels like is very much like back in the early days of my time in operations when everyone was talking about effects- based operations, but nobody could do it, not really. And yet, from that, we've now got a whole series of studies that have become very helpful to targeting processes and related matters that are about kinetic and non- kinetic activities, information, et cetera, et cetera. So, for me, the takeaway is get the stuff out of the junk drawer, get the termites out from underneath the ground, out of the crack house, et cetera, et cetera, all those things we've said and get them back in where they need it. I have really, really enjoyed this conversation and the fact that we've given two episodes to it tells you how much. I think we probably, gents, could do this for the next three or four hours. So, if the audience is up for it, go and get a coffee, we'll continue and just keep going, but maybe we'll do that on another day. Let me draw stumps by saying Dr. Joe Hadfield, thank you. Thank you, thank you. And Dr. David Geo, thank you as well for your contribution to what for me, has been an absolutely scintillating conversation. Sean, thank you as well for your contribution. One of those things about moderating these kind of things for you and I, Sean, is it's great to just sort of pull the pin, as it were, and let the grenade roll and boom, away it went. That's down to you, Joe, and that's down to you, David. Thank you so much for joining this conversation. It's been fantastic. Thank you.

Speaker 6: Thanks, guys. Thank you.

Harry Kemsley: For our listeners, before we say, " Goodbye," properly, and I will let you say, " Goodbye," and I'm just Joe and David, I am very, very keen that now we've started to get questions and requests from the audience, that continues. We've had two great guests on today who've given us some fantastic insights on this concept of OSINT. What are your thoughts? What kind of questions didn't we answer today that you would liked us to? Joe, David, thank you so much for joining this podcast.

Joe Hatfield: Thank you, Harry. Thank you, Sean.

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.