An interview with Bellingcat founder Eliot Higgins

Speaker 1: Hello, and welcome to The World of Intelligence, an open- source intelligence podcast brought to you by the Janes Intelligence Unit. For more information on how we can help with OSINT training and development, go to janes. com/ osinttraining. We have with us today, Eliot Higgins from Bellingcat.

Eliot Higgins: Hello.

Speaker 1: Hi Elliot, thanks for joining. Thanks for taking the time to join. And I've mentioned to you already that I was hoping to get you onto this podcast for a while and every time I've sort of looked at your Twitter feed or the outputs from Bellingcat, I've noted how busy you seem to be. And especially over the last 12 or 18 months, it just has been incredibly, a busy period for you. And so I almost hesitated contacting you and asking you to come on because I didn't want to sort of interrupt what you were doing because you've been putting out some great stuff as well. It would be great just to get from your perspective, a little bit of information, especially for those of the people listening who have perhaps heard of Bellingcat, but maybe don't know much about it. I mean, I'm sure they'll be lots of people listening to our podcasts who'd know all about Bellingcat and know what you guys are putting out, and I'm probably more interest in getting those sort of behind the scenes take from you of how you guys do what you do, but for those who aren't familiar with it, it would be great just if you could kick off with a bit of an introduction to yourself and Bellingcat, and describe what Bellingcat is and what you do.

Eliot Higgins: Yes. So Bellingcat is, it's a bit hard to define. We're often called a open- source investigation collective, which basically is a nice way of saying that we do a lot of open- source investigation and we have a team of volunteers and a team of staff members. I think one thing that makes Bellingcat somewhat unusual is that we do a lot of, our work with crowdsourcing. So often because we have a kind of big audience it's kind of online and interested in our work, we can kind of crowdsource a lot of our work as well. So we have the advantage of being able to do the kind of open- source investigation side of things, but also take advantage of this kind of big resource we have by putting stuff out there and seeing what people come back with. It launched in 2014, just before MH17 was shot down in Eastern Ukraine and that ended up being more of our first, really big stories. And back then Bellingcat was just kind of me, about 50,000 pounds I crowdfunded and now we're a fully rested county in the Netherlands with 19 staff members and a team of volunteers. So we've expanded quite dramatically from where we started off with.

Speaker 1: I mean, it's been an incredible story. And maybe prior to Bellingcat you could give us a bit of a taste of what you were doing yourself in terms of your background in this sort of field of open- source intelligence. I think you've described it very neatly as well in terms of what Bellingcat does as a collective and I'll come back to that if I may in a bit, but yeah, if you could give us a sense of what you were working on before, because you'd built up quite a profile in the open-source intelligence space even prior to Bellingcat and it would be maybe an interesting insight for some of our listeners to know a bit more about what you did prior to that.

Well, back in 2011 I was kind of working in a completely different area. So I was doing kind of admin and finance work for various companies. But I was very interested in all this information that was coming from the conflict in Libya and I was kind of spending time online arguing with people about it. And a lot of it came down to whether or not stuff could be verified to be true. But back then, this is 2011 and you didn't have hords of people geo-locating videos and photographs like you do now. So I just kind of looked at one of the videos and thought well, maybe you can see these things on a satellite image and looked at Google maps and hey presto, there it was. And I kind of stumbled into kind of what we now call geo-location, but back then it was kind of just me on the internet kind of doing stuff. And then in 2012, I started a blog called the Brown Moses blog, which is named after a Frank Zappa song I'd used as an online pseudonym for a while. And I kind of just started it just so I had somewhere to put my thoughts on these kinds of videos I was finding and about the same time Syria was kind of escalating as a conflict. So I started looking at videos coming from there and over time I kind of built my knowledge of what these videos were showing. I taught myself and in consultation with arms experts eventually about all the different weapons that were being used. So in that first year, I kind of focused a lot on the arms and munitions in the conflicts, but then in early 2013, my first really big story that reached kind of more of the mainstream was when at 2013 I identified some very unusual weapons that started appearing in the hands of the rebels in the videos they post online. And that led to the New York Times picking that up and discovering that it was a size of a secret Saudi smuggling operation to the rebels in the South of Syria. It was the first time that there was kind of really solid evidence of that. And the fact that it was video evidence that the rebels themselves had posted online of what was meant to be a top secret smuggling operation, I think caught a lot of people's attention. And then over 2013 and the first half of 2014, I did more looking at things like the August 21st, 2013 Sarin attack in Damascus, kind of other stories to do of Syria. And then in 2014, I wanted to launch a new website and that became Bellingcat.

Speaker 1: It's amazing. And I think what's been particularly incredible in the work that you've done has been the ability, I think, to break stories. And I think that's obviously what's gathered and garnered a lot more attention for you from the mainstream media and publications like the New York Times, et cetera. But I should also congratulate you. I mean, you won so many awards last year I lost count of the number of awards Bellingcat was winning. But what I find particularly impactful about the work you've been doing over the years and particularly since Bellingcat began, has been the ability I think to broaden out the audience for open- source intelligence and the kind of people that are interested in it. I mean, as you're aware from the Janes perspective, we approach it very much from sort of traditional defense perspective of open- source intelligence being a compliment to other types of intelligence. But it not perhaps being seen in the way that you've presented it, I think in conjunction with and looping back to what you mentioned earlier about crowdsourcing, the kinds of people you can get involved in doing investigative research. So for me I find what I'm looking at and reading the output that you're producing, it's almost hard to define sometimes or it's hard to put it into one box or sort of some of the traditional definitions of open- source intelligence we might've had because you're crossing so many boundaries into journalism and into crowdsourcing. How have you found that? How have you found that going from sort of being that person on your own doing that research to then getting so many other people involved and having that support and input from such a variety of experts on the investigations that you've done?

Eliot Higgins: I think that kind of field has grown in the last, I would say eight or nine years from about 2011 onwards, is something that in a way is, they do open- source investigations and kind of that's what I was doing, but really I think the people who were involved with that didn't come from the traditional OSINT community. So they kind of adopted the language of that community to describe their own work, where sometimes it really didn't fit that well, because we aren't trying to produce an intelligence product at the end of this process, we're trying to produce a whole range of products. You look at the work of the team of the New York Times who are doing a lot of this work now, that is for journalism, but I think most people in the OSINT community would recognize those technique, these are something they would use. We've spoken a lot with various people who've been involved with this field and there is a kind of community now from a kind of human rights perspective who are involved with bodies like the International Criminal Court inaudible on Syria, who are kind of reviewing the language that we use to describe what we're doing, because we don't want to describe it really as OSINT because it then puts the kind of weight on the intelligence side of it. So I think the term we've all settled on as a group is online open- source investigation, which can be part of OSINT, but it focuses more on the fact that this is opensource, it's online. And we try and make that definition quite clear. So that community has in a way grown separately from the more traditional OSINT community, even though there's an obvious amount of overlap, but from there you get people from a whole range of different backgrounds. And when I started the Brown Moses blog, I was kind of, because I was just like some random person on the internet, I wasn't seen as belonging to any particular area of expertise. I wasn't someone who was considered a journalist or an OSINT person, I was just blogging on the internet with a funny name blog. So I think from there, for me, it's always been about actually connecting with these different communities and actually connecting them to each other, because I think the big part of this work is having a network and it's not just a network of experts talking to each other, but also a network of just normal people. Like we have our followers on Twitter who sometimes get involved. I mean, that's been very powerful with the Europol, Trace an Object to Stop Child Abuse campaign, where Europol asked members of the public to identify objects they had taken from abuse imagery, and they were sharing this on social media, but because Bellingcat has such a large audience I thought we can amplify this. And thanks to that, it was very successful.

Speaker 1: That's a really interesting example. Because we've been delivering open- source intelligence training at Janes for over 10 years or so, but mainly into sort of government and military agencies, those kinds of bodies who want to learn more about how to make use of open- source information, which if they're not using, then they potentially are missing out on things which could give them interesting insight into what's happening in different parts of the world. What I think is always slightly difficult for them is, I mean, as an example, we get people coming along to our training courses and they'll say something like this to me, which will be, " Oh, have you seen the really cool stuff that Bellingcat do?" And I say, " Yeah, yeah, they do some amazing investigations." And they'll say, " Can you teach us to do that?" And I always say, " I'd love to, it's not an individual task, it involves a network. It involves crowdsourcing. It involves quite often pulling together expertise." And I think it's really encouraging that you're able to use that capability, that network, that outreach to help in cases like the one you mentioned from Europol where those investigations are so potentially wide ranging and difficult for any kind of single body to handle that you're bringing together all of that attention and expert understanding and knowledge to have an impact, it's fantastic work. Did you have them reaching out to you or did you go to them and say, actually, we can help you with this?

Eliot Higgins: Well, initially, we just saw what was on social media and we started sharing it, but we've also been using a platform called Check, which is produced by a organization called Meedan. And it allows you to basically share kind of one open- source object to photograph a video, whatever it may be, and then set a verification task for it and then we were putting these items on there and kind of directing people there to include their comments. So we kind of structured it a bit more so people who did get involved, see what people are saying without having to rely on Twitter as a medium for that, which really isn't ideal for this. What happened then is about nine months into it, Europol reached out to us directly and asked to kind of talk to us about it. And they were kind of very pleased with the results because it had increased the tips they were getting, I think it has now led to like several arrests and children being rescued as well, so it's had an actual impact. But they wanted kind of advice on, to discuss how they could do this better in a way. And they said, " well, could we thank Bellingcat directly?" And I said, " Don't do that because then you get all the nutters who hate Bellingcat kind of going after you for it," which is a sad fact of life. But one thing we're doing now with Bellingcat is we're working to redesign our website. We're hoping, the aim is at the moment is to have a volunteer section where we can be a lot more organized about getting people involved, because we always have people saying, how can I get involved? And it's very difficult to get people involved when... trust is a big issue. It's like we just can't invite every single person in to become part of a big Slack team because we don't know who's who, so it takes a while to build that tracking. And especially now we've had so much attention from the Russian Federation in particular, that's particularly difficult. But we want a place where people can go and actually contribute. So by having this volunteer section, we'll have a place for them to go and alongside that, well, we've got this very popular document that's online at the moment, a list of basically all the open- source investigation tools we can find, organized, but it's a Google doc and we want to actually move that on to our new website. So if you read about an article and you see a tool being used, you'll be able to click on that tool at the end of the article and it'll take you to a page where there's lots of examples of the tools being used, case studies, guides and external links to make it easier for people to find a news open- source investigation related tools, because I think really with open- source investigation, it's about having a toolbox and knowing how all the tools work. And then when you approach a problem, you have a variety of ways of approaching it. And when we do our training workshops, that's what we try and teach. We usually start with quite basic stuff, but the point is that we're trying to build people's understanding of every tool. So when we start setting more and more complex tasks as part of these workshops, they actually kind of say, oh, I can use this, this and this, and then come up with a solution.

Speaker 1: It's really interesting. You mentioned tools and it's something we've discussed on previous podcast episodes is, when we are delivering training, you do get people coming along and saying, oh, can you just show us all the good tools? And it'd be great to hear your thoughts on this, but I think the way that you're presenting it and the way that you're presenting case studies on the Bellingcat website is really useful because it's not just about the tools themselves, you've got to be able to put it into a workflow and think about when you can use those tools and when they're going to be useful, when they'll be appropriate and which ones to pick. And I think that's often the challenge, it's... And do you find that in terms of either the training you deliver or working with others at Bellingcat, that as much as you can have lots of great tools out there, sometimes there's not always the understanding of when and where you might use them?

Eliot Higgins: Yeah. I think as well that often people come to us and they might have a tool they've designed and they are trying to design a tool, but I've noticed less between this, but a tool that kind of tries to do every open- source thing. It says, okay, this is going to help you verify an image. It'll look up 20 search engines and do all this stuff and then no one ends up really using them because often when you're doing an investigation you know from experience which are the most effective tools to find your route to the solution. And if one tool doesn't work, there's probably others that could do something similar. So we'd published a couple of weeks ago an article about the different search engines and their reverse image search capabilities-

Speaker 1: Oh, yes.

Eliot Higgins: ... because mostpeople know about Google reversing research, but there's also Bing, we showed off and Yandex reverse image search. And we made the point that for all the tests we did, the Yandex image search tends to give the best results, but that's probably the one that's known less by everyone, but this is at all, it doesn't say thing, it just does it say slightly differently. So it's educating people about those tools and saying, there's more than one way of doing this. And we do that for our workshops. We do that through writing these case studies, we do that sometimes by doing the investigations and explaining step- by- step how we did it as well.

Speaker 1: No, I think that's really beneficial. In terms of the work that you're now doing with Bellingcat obviously, as you mentioned, you've got a growing team and you've got more stable sort of funding sources, et cetera. The investigations you've done have garnered a lot of attention, obviously MH17 was key, I think. The investigation that Bellingcat produced was vital for actually getting everyone to accept what happened, because you did such a great job of putting together the evidence and presenting it. And you've hinted a little bit at some of the challenges that's caused you in terms of being targeted as a result, but what's the overall strategic direction for Bellingcat? I mean, do you have a sort of set of topics or subjects that you're going to be particularly focused on either now or going forward, or is it a case of picking up whatever cases seem interesting at the time?

Eliot Higgins: It's kind of in two areas that are related at the moment, we're looking at ways to kind of expand the community of people who are doing this both keen amateurs and kind of more experts. And that can be through training this volunteer section we're launching. We are looking as well at working with universities to train students, also universities to doing local investigations. And we have they started that off in the Netherlands with a hope that if they can engage with local issues or realize they can have real impact on the world by doing this kind of investigation. So there's kind of this community growth and that expansion that we're working on. The other area we're looking into is because we're now often approached by people interested in how open- source investigations Bellingcat has done, or just open- source investigation in general can be used for justice and accountability. We're looking at the processes that we've used to do investigations and improving them, because one thing I've discovered through doing the MH17 stuff is we did that when we launched Bellingcat, it was basically me. We didn't have any real process for archiving or saving content. And now three or four years later, I started getting, people reaching out from... lawyers who were involved with court against Russia who are now saying, can you make a submission about what you found, which we're happy to do, but the problem is we're discovering a lot of links that we just discovered back in 2014, of course, dead now. So what we've done now is we've developed a new process working with the Global Legal Action Network to basically archive and investigate material up to a standard where we could basically submit our investigations directly to a court along with archive material. So with that, we've been working a lot with the Syrian Archive who've renamed themselves for Mnemonic. So the Syrian Archive is a platform that has been archiving videos from Syria, but they've expanded into other areas. We've been working with Hunchly, which is a handy open- source investigation tool for recording your activity. So that's the kind of archiving side of it. And then we're looking at the whole process of how we investigate and seeing how we can improve different elements of that. And we've already been working on investigations related to Saudi airstrikes in Yemen, which we've been publishing on yemen. bellingcat. com. And they are using this new process for archiving investigation, but they've also been submitted as part of the current government consultation on arms export agreements with Saudi Arabia. And our hope is that we can have more examples where we do this kind of research and then it can kind of be used in a courtroom situation or a legal situation so we can actually see how it's kind of attacked, because it's very important that we continue to refine the process to make it as good as possible, but also we don't overdo it and go to the point where it actually just takes forever to do even a simple investigation, because we have so much of, so many parts of this process. So it's kind of setting a minimum standard for our work that can be applied not only to Bellingcat, but we can also export to different organizations.

Speaker 1: I think that's a really interesting initiative, and I think there's potential there to create probably something that would be seen as something of an industry standard, if we can call it that. Because I think that's a question that comes up a lot for us as well as, how do we record and capture things that would be useful both from an archiving perspective, because obviously as you say that there's a lot of stuff that we will look at in our research, which is sort of here today and gone tomorrow because a lot of online content can be very ephemeral. But there's also then that perspective of actually when somebody does come back to have a look at it later, how do you show them that content? Or how do you potentially use it evidentially as well? So yeah, I think that that's something that I think will be necessary for everybody involved in open- source intelligence to adopt, those levels of rigor in terms of archiving and retrieving information. As your sort of, or as you have moved in the last five years or so into that area of presenting more information as evidence, obviously verification has been a massive challenge and that's been something you've been involved in right from the beginning. How has that developed in terms of as a challenge, has it become more difficult? Do you feel it's gotten easier in some ways, or is it a bit of both?

Eliot Higgins: It's one of these things that the better you get at doing verification and kind of the more sure you want to be about verification, and that the more time you spend doing it, even if you get really good at it. So I mean, now when we investigate an incident, we have a much better understanding of all the kind of potential angles we can approach it from and how people who are kind of going to attack it are going to approach it. And having spent... I kind of start doing this because of my online activity, so I was always very aware of how people will attack your work, even in a very disingenuous manner, but you still have to kind of be prepared for that. And in a way, I think from a very early point, I was always very careful about making statements that I could back up with direct evidence, or if I couldn't be sure of something, I would be very clear that I wasn't sure of it and that's kind of what we try and get across in our writing as well. But it really, what happens now, we kind of have a process we've broken down into three steps and it's similar to the intelligence process. But I came up with this idea before I knew what that was. We call it, identify, verify and amplify. So we identify information either we're kind of actively searching for it or maybe in a more passive way with people sending stuff. We verify it, which is a big part of it. And then the amplification stage comes. So because we verify the information we've identified, we can use it in a variety of different products. Like with MH17, we wrote dozens of articles. We wrote lengthy reports. We've been involved with television programs about it, we've produced the six part podcast series that we've done that goes into in great depth, but it's all based off the same verified information that we've identified earlier. And kind of my approach is that there can be multiple ways to tell the same story or explain the events in a multiple way. So with MH17, we did the podcast five years in that we've revisited it by doing a second series of the podcasts that focuses on the BBC Africa Eye investigation. And that's me over killing. It was a video from Cameroon where two women and two children were marched off the road by soldiers and executed. It's a horrific video, but the BBC did this investigation with Amnesty and Bellingcat, and now we're doing a podcast series because the trials have come up and our third season, we'll look at the activities of Russian intelligence unit 29155 in Europe, which is something we've been doing a lot of writing on as well. So that takes all the stuff we've already been doing and it kind of repackages it into a new format that can reach a wider audience and get them kind of drawn into, hopefully open- source investigation themselves.

Speaker 1: It's interesting, the amplification aspect. I mean, and those are great case studies. We had, Ben Strick from the BBC on a previous podcast episode talking about the Cameroon investigation and just, it really, it'll show the art of the possible when it comes to using information, which is open- source, which is out there. Do at Bellingcat, are you sort of conscious that what you're doing is very specialized though in many ways, in that you do need to perhaps find and bring into the network and bring into the organization, people with the right skills and expertise who can do that? And it's not always easy to find perhaps generalist researchers who can do some of the, particularly the more detailed image verification or video verification, and do you find it hard or have you found it hard to find the right kind of skills or is that something you're finding easier now that you've got it branch down into training and you're helping encourage the promotion of those skills?

Eliot Higgins: Well, even in quite early on when I was doing the Brown Moses blog, I was kind of in touch with a lot of, kind of experts because they saw what I was publishing. And I was pretty much the only person who was bothering to look at this stuff. So they kind of asked me about it. And then I started talking to them and it was kind of very early on. It was a lot of kind of arms and ammunition's experts, especially linked to the kind of big human rights NGOs, but I kind of learned from them and I found cool stuff for them to look at. So then, as I've kind of gone on, I've built more of these relationships. So now when we need kind of expert advice, we have a whole pool of people we can use for that. Sometimes we reach out to new organizations because we're doing something new. So when we're looking into these GRU officers that we started identifying, we started using voice analysis, forensic voice analysis. So there, we reached out to a few universities where they do that so we could get the export overview. Same when we started doing, finding all these IDs of all these sports nutrition salesmen, who were hanging around Salisbury when Skripal was poisoned. We spoke to a university who could do facial recognition to compare the photograph we had from the ID, we had acquired their real ID and their fake identity to have at least that kind of extra level. So as we kind of learn more and kind of find out about these techniques, we try and pull them in more into our investigations. Sometimes some investigations can be very simple. I mean, the PS752 investigation we did a week ago, it was actually quite a simple investigation because we had already really done the same kind of investigation with MH17, so we kind of knew what we were going to be looking for. So with MH17, one of the first things we did is kind of piece together the photographs of the wreckage to see if there was any clear signs of like a warhead damage. So we knew with PS752 that we need to look at similar images, but then kind of things moved along more quickly because then this image of the remains of a tour anti- aircraft missile came on, which unfortunately wasn't geo- locatable. And then this video that appeared to show the aircraft being hit by a missile and that we could geo- locate, and that all of a sudden it became this huge story, which really, it was a very simple geolocation that we did. It's just, it was such a big story at the time and it got a massive amount of pickup.

Speaker 1: No, indeed. Yeah. So it's interesting that you've built that up over time and you've got those links to the right kinds of experts and you're able to reach out to them and get those positive responses and that help. You mentioned a little bit earlier, you sort of touched on some of the more negative attention, I suppose, that you've received since starting up some of these investigations. Has that hampered any of your current investigations in terms of either trying to crowdsource information or being drawn to information which you're then looking at and thinking, hang on, this is almost sort of deliberately fake, it's almost been designed for us to find, has that occurred at all?

Eliot Higgins: I think by default we're suspicious of anything we find, we always try and verify something from multiple angles and understand it as part of a network of information. So I'm never too worried about fake information being introduced because we can identify that pretty quickly. And it's hard to make it really genuinely good fake and people have tried to do that. I mean, the most absurd example is of course when the Russian MOD used a screenshot from a computer game as evidence that the US was helping ISIS. But this is the level of people we're dealing with. So it makes it rather easier when you're dealing with people who are massively incompetent. And it's like, when we're dealing with Russia, they have kind of the same pattern of behavior again and again, very rarely do the Russians actually come up with their own ideas when they're having their various press conferences and announcements. They usually steal it from somewhere else. It's like we had this OPCW meeting at the UN's Security Council on Monday of this week where Russia basically read out his claims, why the OPCW couldn't investigate staff, that they were biased and they were just repeating claims made by someone else. And those claims has already been addressed and debunked by us. So it kind of makes it easy for us when they have that pattern of behavior. I mean, what's also, I find very interesting is a lot of people don't really appreciate this kind of alternative media ecosystems that exist out there. We've interacted a lot, I think as a way to describe it with the kind of media ecosystem that is kind of focused on conflict, particularly Syria. They tend to be made up of kind of conspiratorially minded anti- imperialist types, but it's always the same network of individuals and websites that are sharing the same articles. So within this kind of echo chamber, they have this very kind of vibrant ecosystem, which they, I think they think has more impact than it really does, but that's something that then I think some people aren't as online as other people notice, they might be more academic I've found, and they think they must be Russian trolls. It must be a Russian troll factory because they can't understand why people, supposedly real people would believe stuff that's so patently absurd, but they really do believe this stuff because they have a whole echo chamber that is constantly reinforcing these messages and that's kind of who you tend to have more interactions of. And then of course, Russia or whoever needs a bit of amplification uses those individuals and sticks them on Russia Today, or brings them to the UN to talk about why the White Helmets are all Al- Qaeda and those kinds of things. And I think it was best reflected in a report that, a research called Kate Starbird dates on the White Helmets where she kind of really did a good job of mapping these networks out. And I'd really recommend that to anyone who really wants to understand how this kind of alternative media, echo chamber kind of operates.

Speaker 1: That's a great tip. I think everybody who's involved in doing open- source intelligence, open- source research really needs to have a good understanding of how these kinds of disinformation operations are working and to have an idea of how to spot and verify content to the extent that we can do it quickly or to build up the expertise, to do it in more detail. Essentially you touched on sort of the disinformation aspects of amplification. And when you talk about amplification in your own process, is it a competitive sort of process of amplification that you're having to compete against these other sort of voices that are trying to put out counter narratives that are baseless?

Eliot Higgins: I think there is some times, I mean, often you have to recognize that some of the people who are making the most noise are getting the least attention. And sometimes it's the people themselves who are promoting these counterclaims who are actually doing the most damage to them. The OPCW thing, being a good example. I think the two leakers, this Ian Henderson and this mysterious Alex figure, did the worst possible thing they could do to their reputations by first approaching this kind of working group on media and propaganda on Syria, whatever they call themselves, which I think anyone who knows about chemical weapons in Syria probably knows about them and probably has a very low opinion of them unless they're part of this kind of alternative media echo chamber. So when people are going, oh, why won't the mainstream media pick this up is because the leakers have chosen to go to the least credible people possible in the eyes of the mainstream media. So in a way they can kind of completely undermine their own messages by the fact of who they are. But it is something that I think you kind of have to pick your battles because especially now Bellingcat is more well- known by engaging with arguments on topics that are ridiculous or easily debunked, so you can actually just amplify those bits of misinformation to the audience who may be observing that and you might not want to do that. So I think sometimes you need to kind of pick your online battles when it comes to countering disinformation.

Speaker 1: That's really interesting. I think you've touched there on the OPCW case and for anyone who's not familiar with that, it might be useful just to sort of briefly explain what's happened there. Because I've sort of followed it a little bit and we've covered chemical weapons incidents in some of the research that my team have done over the past year looking at Syria. But yeah, did you want to just sort of briefly explain that example of trying to counter somebody else's disinformation?

Eliot Higgins: Yeah. So over the past year or two, leakers, former staff members at the OPCW have published documents that they claim undermine the conclusions of the OPCW fact- finding mission on the Douma chemical attack. One set of documents was engineering poor by someone called Ian Henderson which made various claims where he basically said it can't have been a chemical attack. Then this guy, Alex is a pseudonym. He several months later publishes emails from the OPCW which he claims shows a coverup. This was reported by Peter Hitchens in The Mail. I think Robert Fisk did some stuff on it. So it kind of more than other cases, it kind of went a bit more into the mainstream, although it really didn't. The problem was these documents are very complicated and unless you have a real in- depth understanding of what the OPCW has published and other details, you wouldn't really understand for example, some of the concerns raised by these two figures were actually addressed in the final OPCW report. And there's kind of lots of issues. We've done articles on Bellingcat where we've really gone in depth into both of these links and kind of looking into the claims they're making. And there's significant issues with both, but the people who want to discredit the OPCW, who want to discredit the claims that the chemical weapons have been used in Syria don't really care. They're more about the impression that something dodgy is going on at the OPCW rather than the facts of what's actually going on at the OPCW, because their agenda is to attack the OPCW. And like we had, this week we've had Russia at the UN Security Council bringing these claims, using them to attack the OPCW at the UN Security Council. But within that, the claims they're making are ones that we've kind of addressed and debunked, but it doesn't matter. It's the sound of the noise they're making, which is important to them, not actually what those noises mean.

Speaker 1: Yeah, it sounds like a typical sort of example of when their trying to basically use bluff and bluster to sort of cause people to doubt the OPCW when there's nothing there to it. And it just sort of reinforces the case as you say that actually there isn't a case against it and they sort of shoot themselves in the foot ultimately, but it does enough, I suppose, in the short term to create or to serve their agenda, so it almost doesn't matter to them the quality of the disinformation they're putting out, it's just putting out... it's the quantity almost really, isn't it?

Eliot Higgins: Up to an extent, yeah. I mean, often, this stuff bounces around the kind of alternative media ecosystem and has no real impact, but this was a rare example where, because I think in particularly what Peter Hitchens wrote on The Mail was he was kind of course playing Seymour Hersh about there being a, this is just like the dodgy dossier that led to the Iraq war. It's not at all, but it makes it a really good headline when you stick that on a website. And Peter Hitchens knowledge of chemical weapon used in Syria, like much of these people is extremely limited. It usually only refers to the kind of really big well- known chemical attacks, not the dozens, if not hundreds of attacks that occurred with no one really taking any notice of them, because they aren't really engaged with the topic in any serious way. They're just full of bluster, they want some nice headlines. They want to make their point and they want to attack the OPCW. But ultimately it's not about the facts, it's about just the appearance. It's just so they can convince people of something that really isn't true, although I'm sure in many cases they've convinced themselves it's true because they haven't really looked into it in a way that they should have done, and they might not even be equipped to do that or even aware they should be doing it in the first place.

Speaker 1: No, for sure. And so in terms of the... And sort of moving on to think about the kind of work that you're doing at Bellingcat, the types of information that you're using and the research you're able to do, one of the things that, or topics that we've been picking up in some of our podcast episodes has been around the question of, is it getting harder? Are you finding that as more and more of the people you might want to investigate become aware of what you can do with their information that they're starting to get more security conscious, they're starting to not post as much information online? And in relation to that, are you finding that some of the online platforms where we might find useful information, from our perspective at Janes certainly we found that some of the social media platforms, et cetera, are getting harder to access in many ways and that there's a stronger move towards privacy of online information now. So again, is that something that's affecting your work at Bellingcat?

Eliot Higgins: I think the biggest effect does come more from the social media companies, rather than the users themselves, because we can look at a whole range of subject, but we focus a lot on Syria and Russia. We look into other areas and there, these are often in countries where people have never heard of open- source investigation and they do not know if they should be, not publishing their entire lives on their social media accounts. But what we found is actually the social media companies, as they've tried to crack down on the abuse of their platforms, either through the way data is being used or through violent content and extremist content being shared, it actually makes it more difficult for us to do investigations. And one thing that's been quite interesting for me is how the ICC has become more interested in using open- source. And at the same time, the sources they might be using would get shut down before they could actually open an investigation and preserve those sources. So in a way it makes the kind of responsibility of the kind of almost, maybe you could call them first responder organizations like Bellingcat to archive that material as they find it even more crucial to those of investigations. And especially with the ICC's relationship with the US, getting information from the social media companies that are based there, they can't demand this information, they can ask nicely, but ultimately it's down to the social media companies, what they share with organizations like the ICC. So I think it then raises the question of, organizations who are doing open- source investigation, they have a responsibility to archive this information, how do you make it accessible to those justice and accounts busy bodies? And that's kind of another area that we're looking into at the moment.

Speaker 1: Okay. So yeah, the sort of story and that information is definitely becoming more important, I think for everyone all around. What's your feeling in terms of where the information space is going, in terms of future challenges, future developments? And also, what do you see coming next for Bellingcat? What are your plans for, you mentioned the podcast sort of series you've got coming up, but what else is on the agenda for Bellingcat over the next sort of months and maybe even years to come?

Eliot Higgins: It's hard to predict things with tools, but one thing we're particularly looking into is now we're doing all this archiving and collecting all these videos and adding metadata to it, how do we make these data sets more useful? Is there ways to use machine learning to identify objects in them? And that's something that the Syrian Archive has been doing with cluster munitions, forensic architecture. They did a project on Ukraine where they used a system to identify tanks in video footage. And I think we're going to see more and more interesting ways of taking these vast amounts of video and photographic archives and turning it into... using that data in different ways. So I think that's kind of where we're focusing on at the time, at development and technology side. As for the activities of Bellingcat, I mean, we've got more series of our podcast plans. I'm working on a book at the moment that should be out either late this year or early next year if I finish it.

Speaker 1: Oh, fantastic. What's the book called?

Eliot Higgins: Yes, so the book is going to be kind of about myself and Bellingcat and how open- source has kind of developed from my perspective, aware that my perspective is not the only perspective, but it's kind of really digging into it and talking about some of our investigations in a lot more depth and how everything's kind of developed and particularly focusing on those areas I've mentioned around online communities for good and bad.

Speaker 1: It sounds fantastic.

Eliot Higgins: Yeah. And we're also working on a couple of secret projects, very major-

Speaker 1: Interesting.

Eliot Higgins: Not related to investigation, it's for other things, but you'll hear about those soon enough. Yeah, and we're continuing our investigations. We've still got lots and lots to do on the unit type 29155 that was linked to the Skripal poisoning and other stuff in Europe. So I expect we'll see a lot more on that. We're working on MH17 investigation since the joint investigation team announced a call for more details on some of the suspects, so we're digging into that as well. So we've got plenty to keep ourselves busy and we're hoping we can do more work on our Yemen project and continue to develop that new process. And hopefully by the end of the year, we'll be able to start actually packaging it up and deploying it to other organizations who might want to use it.

Speaker 1: Oh, so in terms of that, just to pick up on that one, so when you say packaging it up, how do you mean, what would it look like and what would they do with it?

Eliot Higgins: Well, I mean, one thing we want to do is make sure whatever we produce is either free or cheap because we want to share it with organizations who probably don't have much of a budget. I mean, we don't have a massive budget, but we've probably got more than some of the organizations who are working on Syria at the moment. So it's kind of laying out the process, both the methodology for investigation and the kind of more technical side of things when it comes to the archiving of the material. And then we've been working, or a company called Benetech has been working with the Syrian Archive to create an indexing system that will index these videos in a way that makes it easiest for people to figure out who has the same videos without actually looking at the content themselves so people can be secure with their own content and not have to share it with every single person in the world before someone can use it. So that is also kind of very crucial to kind of developing this process where we can then start setting up people. If there was like 20 organizations who had archives for Yemen, we could kind of help them use this new process and help index their archives and make it useful information for those people doing investigations. Because really it's making in a way, taking the video that someone may have filmed on the grounds in the moment and then turning it into something that's useful and discoverable by the kind of justice and accountabilities body, the kind of people perhaps who are calling this stuff want to find.

Speaker 1: Yeah, I doubt.

Eliot Higgins: And YouTube is not the platform for doing that, so we have to come up with something else.

Speaker 1: Interesting. So I mean, I think you've touched on an important point there in terms of not just getting all of the data, but actually making it useful for people and actually allowing others to share it and do work with it and building on the value that you sort of have within that information. I mean, that's a huge challenge and it's something that obviously we've wrestled with at Janes for years, in terms of, as you have more technical developments in the way that you can access and categorize data and use it and link it together, there's much more the people expect you to be able to do with it. There's more than you want to do with it and it's actually a much more difficult challenge than many people appreciate. So I look forward to sort of hearing more about that as that project develops and as you start to roll things out to other organizations, seeing what comes of it. What's I think really fascinating from our perspective as well is looking at some of the work that you're doing, learning from some of those case studies as well and trying to also share and provide any thoughts that we can. Anything that comes up in terms of investigations you're working on at Bellingcat, we'll be observing closely and again, if there's anything that we can do at Janes to help, then do get in touch. And yeah, we'll keep watching with fascination. Was there any other stuff that you had on your mind that you wanted to sort of talk about in terms of things that Bellingcat are doing or the sort of direction you're going in?

Eliot Higgins: I think I've covered everything in the last 40 minutes or so. I think I've got most recalled.

Speaker 1: Yeah, no, no, it's been fascinating talking to you. I mean, I have one last question really, which was relating to some of the investigative work that you do, where you are sort of getting crowdsourced information or sometimes tip offs. And do you find that Bellingcat now is at a stage where you're at a point where you can attract and receive information that is really difficult for others to get ahold of and it almost goes beyond open- source information because it's not necessarily stuff that anybody's going to be able to find, it's just stuff which you're attracting or being sent either as tip offs or crowdsourced information?

Eliot Higgins: I mean, we do get a lot more people kind of emailing us, I'd say about 99% of those are not kind of relevant to our work. It might be people asking you about like, because we did MH17, everyone wants to know about MH370, which crashed somewhere in the middle of the Pacific, from what I understand, which is not an easy one for us to do, because there's not loads of fishermen taking selfies with planes crashing in the background.

Speaker 1: Sure.

Eliot Higgins: It's just a huge ocean. So we can't do work with that. We also get kind of people who have some really interesting ideas, but aren't the kind of ideas that we want to kind of look into, like where you can find dragons and who are the illuminati, that kind of stuff. But there is some stuff that gets sent to us and the big issue we have at the moment is because we are at absolute full capacity at the moment. It's very difficult to take on other work and I hate turning away people who are saying, oh, we have a really good, interesting investigation to do, but we just don't have the resources to do that. So that's always difficult to do when we turn them away. What's quite cool though, is because so many people are interested in open- source investigation know our work, it's like with PS752, immediately people started seeing those videos and photographs, even though we didn't ask for them, but we had loads and loads of this coming through. And lots of, I think people heard Ukraine plane crash and they associated that with Bellingcat, so they started sending us lots of stuff, but that meant then we could kind of engage with that and get a lot of these videos as soon as they appeared online, as soon as they were shared. And with PS752 as well, a lot was being shared on Telegram and then re- shared on Twitter. So then people were finding stuff on Telegram and then sending it to us directly. So that was actually something that was, in a way it kind of made this kind of human search engine for us that we didn't even have to type anything into it, they just went up and did it for us. But that's also something that I think gives us advantage over a lot of other organizations that we're so well known for this particular kind of work and doing crowdsourcing that people will send us stuff. And sometimes it can be really, really useful and lead us to kind of big discoveries, but with the Iran shoot down PS752, the video that shows the apparent moment of the shoot down, someone just send that to me, like literally nothing in the tweet, just the video. And I didn't even bother watching it for 10 minutes and I was like looking at other tweets because we're getting loads of them. And then I watched it and I was like, what's this? And then I saw what happened. I was like, oh God, you better start looking into this one. What was great there? Because we had already done the work looking for other stuff in the area, we almost immediately recognized the area that this video was shot in, and it allowed us to geo- locate it very, very quickly, which meant we could get it out there before pretty much anyone else could. And also, because we were able to collaborate with the New York Times on that, we were able to kind of help them do that and look into some of the other videos they were looking into and they could produce their own thing on that. But yeah, being able to get stuff out really, really quickly like that, I think helps when you have a lot of competing narratives as well, can help kind of clarify what's going on very quickly.

Speaker 1: It's really interesting. And timeliness I guess is an issue there as well in terms of, that rush to try and get on top of what's going on in a developing situation, something that where the information is changing quickly, from a Bellingcat perspective, are you trying to get stuff or assessments out there as quickly as possible? Or would you prefer to sort of take your time, wait until you've got all of the information in that you could get evaluated, et cetera, and put out something which you know is more reliable?

Eliot Higgins: It depends. It's like that video of the plane shoot down, there we realized how significant it was and we were able to geo- locate it really quickly. So we kind of did a Twitter thread about it and that kind of got a lot of interest, but I said, we should break this up and put it out there, put it on the website, just so there's like a permanent link somewhere for people to find. So in that case, that was definitely one where it was valuable. In other investigations, especially ones that are looking at incidents that have happened some time ago, it's just worth doing the whole investigation. But it's really about looking at how interested people are about certain information. I mean, sometimes it doesn't even have to be an actual piece of investigative work. With following the Christchurch massacre, there was a manifesto part online. So this is one of these alt- right mass shooters. And there was a manifesto put online, and this shooting happened overnight because it was in New Zealand. One of my team members, Robert Evans, who was in the US, he saw it was happening and he's specialized in this particular area. And he recognized that the manifesto is actually written in a way as basically an in- joke for HN or a series of in- jokes. And the whole idea of the manifesto is basically to trick the media into reporting about the manifesto in a serious way and would have been the shooter's like final joke against the mainstream society that he was railing against. But what happened, because he recognized that he wrote an article, we got at like 7: 00 AM explaining what this manifesto was actually about. And that ended up being hugely influential in the way it was actually reported because lots of journalists saw that, it was shared very, very widely, very quickly. It was a very well read article. And then the reporting on the manifesto basically refer to our work on it and said, this is especially a trap for journalists and it basically stopped the kind of mass shooter having his final, the last laugh with what happened, rather than tricking the media, it was pointed out exactly what he was doing. And because we could get it out first thing in the morning in Europe, we'd gone through the European media before reaching the US media, if you also reacted to our article in a positive way.

Speaker 1: That's really interesting, yeah. No, it's been a really fascinating discussion and thanks for covering so much ground in terms of the work you've done up to now and what's happening at Bellingcat, and giving us some insight into how you go about doing the things you do. I suppose one of the challenges you've got still is perhaps managing expectations where when people see what's possible and what's feasible in some of those cases, they almost expect you to do the same with so many other cases too. Is that something you're also trying to sort of deal with in terms of saying to people, actually, these techniques aren't going to work every single time?

Eliot Higgins: Yeah. The thing is we only publish our successes inaudible, so it looks like we're getting it right every time, but there's plenty of stuff we do where we just can't figure it out. And I think people kind of see what we're doing and they think, well, if they can geo- locate this, they can geo- locate this photograph. And so often that works, it's not a magic wand that we can wave at stuff to find solutions.

Speaker 1: That's reassuring because for a moment, I thought you guys were making it look easy, but no, no. It's been great, it's been great to talk to you. Thanks for sharing all of those insights and your thoughts on open- source intelligence. And yeah, thanks for joining us on the podcast and hopefully we'll speak to you again at some point in the future. And we'll await the other podcast you've got coming out and the book when it eventually hits as well.

Eliot Higgins: Thanks for having me on.

Speaker 1: No problem, thanks Eliot. Please leave a rating on Apple podcasts or on your preferred podcast listening platform. And for more information on how we can help with OSINT training and development, go to janes. com/ osinttraining.