Autopsy of a story: the art and science of fact-checking

[music plays]

NINA LAMPARSKI

The way my team works, I have, like a – I have local reporters spread across the entire continent and we all work together really closely. So when it’s an election in one country, that person doesn’t work in an isolated manner, they work together with the other fact checkers, even if they’re based elsewhere, because this information knows no boundaries. So there’s a lot of sort of mudslinging that might happen in countries adjacent to Kenya, for example. So collaboration was key.

[music plays]

MATT ASHBURN

Welcome to NeedleStack, the podcast for professional online research. I’m your host, Matt Ashburn, and yes, you can fact check me on that.

JEFF PHILLIPS

And I’m Jeff Phillips, tech industry veteran and curious to a fault. Today we are kicking off our series on factchecking and debunking. So with the US midterms just a couple of months away and there will no doubt be a lot of attention on fact checking candidate and pundit claims, debunking, the latest conspiracy theories and I’m sure just more fun to come during the elections. So to get us in a fact checking mindset when we’re dealing with online research, we’re joined today by a very special guest, Nina Lamparski. And, now, Nina has spent nearly 20 years in the world of journalism, including as a journalist for the BBC. Her experience has spanned every continent, but what’s interesting is that in her current role, she leads AFP’s award-winning digital verification team in Africa. Nina, welcome to the show!

NINA LAMPARSKI

Thank you very much. Thanks for having me. You make me sound very illustrious. I like how you presented that. But yeah, it’s pleasure to be here.

JEFF PHILLIPS

It is very illustrious. We’re super excited. So let’s start with this digital verification team. That’s what kind of connected us. Could you tell us a little bit about how you arrived in that role and what the purpose of that team is?

NINA LAMPARSKI

Yeah, so AFP, just for those who may not know this, is actually short for Agence France-Presse. It’s one of the world’s top news agencies. And so, we have a very longstanding history of uncovering news for clients who include a massive bunch of media companies around the world. And the reason we launched the factchecking department goes back to 2016, really to your election in the US and with the arrival of Donald Trump, when we saw sort of a previously, just this incredible amount of disinformation hitting social media. And in France, we had a election following year and we were kind of worried that similar things would be happening here, and we had a far right candidate with certain ties to Russia and we just felt that it would be a good thing to monitor how this election would unfold. So what happened was AFP hired its very first fact checker in 2017 and then created an alliance with a whole bunch of other media outlets, both French ones but also international ones, and they all fact checked presidential campaign speeches, basically. So all the candidates running in that race were being put under in the spotlight and everything they said was being debunked by a bunch of reporters. So that’s kind of how it started. My own – and since then this was in 2017, now in 2022, we’ve got over 130 people working in the fact check department, all kept very busy on literally every continent – and I started working in my current roles three years ago. So as you pointed out, I’ve been like a journalist for a long time. I was a foreign correspondent in various places. I’ve covered a whole bunch of topics, but primarily politics – a lot of different elections and the migrant crisis, the Iran talks. So I have a personal interest in kind of the rhetoric around the news story. I also have an interest in sort of, I would say the far right, the far left. And so when this role came up to grow the fact checking team in Africa, I thought it was a really interesting sort of approach to media and looking at news stories, not so much telling you what something is, which is what a traditional news story will do, but rather looking at what something isn’t, although it claims to be that. So that’s kind of a fact checking lead us and what differentiates it from a normal new story. So that’s kind of it.

MATT ASHBURN

That’s really fascinating, Nina. Thank you for sharing that. I’m wondering, as you’re standing up a team like this, maybe it’s a new capability, right, in some organizations – what’s the process you go through to train them to develop that investigative trade craft to make sure that the investigations meet the standards of an organization like AFP?

NINA LAMPARSKI

So all the major news fact checking organizations around the world are part of the IFCN – that’s the International Fact Checking network – and we are all signatories to their charter, which, you know, obliges you to be transparent in your funding, your sources, your reporting, be neutral. And so in order to join them, you have to go through a training process and essentially ensure that all of your reporters know how to use fact checking tools, know how to verify visuals. There’s a lot of stuff out there, a really great thing about fact checking is that most – well, I want to say a really big majority – of the tools are free. Some of them, the more pushed ones, you will have to pay for like satellite images. If you want to get really high quality images, you’ll have to pay for that service. But unless you’re, like, a military hardware expert who needs to see the ground really up close, you can get by with like free software, for example, Google Maps or Google Earth. Like, that’s totally fine. So, yeah. So I guess we’re not some of the fact check organizations out there. Like, you may have heard of Bellingcat who are kind of leaders in the field. They do really amazing things and sort of uncover they work on mass grave investigations, for example, things like that, whereas we kind of focus just on a broader – just anything that is in the news, essentially. And so we really look at anything verifying, like a doctored photo or video, all the way to political claims that could spark violence on the ground. So yes, it’s a pretty broad canvas that we have, but we don’t do what we don’t particularly do is long term deep investigations that we balance. So we have a slightly different planet, but that’s kind of what we do. So you sign up to the ICN and then you have to obviously every year you get to this monitored as well, and they verify that you are doing a good job, basically that you’re up to standard.

MATT ASHBURN

And one of the major world topics right now, obviously, is the conflict in Ukraine. And we’ve seen a marked increase in the attention given to amateur sleuths and researchers out there on the Internet. Lots of really incredible information that we really haven’t had that level of visibility before in a conflict like this, so it’s pretty interesting to see. I’m wondering, as a journalist, what’s your relationship or view of these amateur or independent researchers? Is there a place for them in your work, or is there a different view on that?

NINA LAMPARSKI

No, I think there’s definitely a place for them. As you point out, I think a lot of amateur researchers, especially for the Ukraine conflict, for example, they are military hardware experts. They take an interest in it. Some of them are ex army, others have an intelligence, security intelligence background. I mean, I work with that in the same way that I work with any source. I never take just one source for granted. I would not run anything unless it’s an AFP correspondent. I think, generally speaking, we definitely look at more than two or three sources. But yeah, I guess the better ones end up having quite a good reputation. I certainly follow OSINT experts online. We had a lot of Ukraine disinformation in Africa. There’s a very strong, in some countries, a very strong anti-NATO sentiment, so they take a strong – and then pro-Putin sentiment as well. So we saw an astonishing amount of this information. Some of that definitely involved like images of weaponry and so on. Then you have to kind of go and look. You do look at what amateur researchers do. But as I said, I think it’s just kind of you have to just follow your gut and make sure that, how do we say, don’t trust anyone. But if you have two or three sources and they all kind of match up and you kind of can cross-reference it with your own reporters on the ground, then there’s a pretty good chance. And with other experts, like renowned experts, then there’s a pretty good chance that it’s true. So, yeah, so it’s all part of a bigger ecosystem. I wouldn’t just rely on one person, but I definitely think it’s great to have them for sure.

JEFF PHILLIPS

That makes a lot of sense. And yes, we can’t trust, I don’t know what to trust anymore when I’m reading things. Yes, sadly. Well, we mentioned at the top of the episode that the US midterms are fast approaching. So are there any challenges of fact checking during election cycles or any trends that you’re seeing in elections from around the world that demand fact checking or make it more complicated?

NINA LAMPARSKI

Yeah, I think when I say don’t trust anyone, like, definitely don’t trust anyone in elections, I mean, I haven’t worked on US elections, but I do follow them very closely. I have, obviously, colleagues who have debunked candidates over there. From our experience on African elections, it’s definitely the case that both camps or several sides, more than certainly one side, pushes this information. One trend that we witnessed in, for example, the recent Kenyan election was politicians hiring influencers. Influences are really hot obviously. You know, they’ve got huge following and so they had these sort of these influencer troll factories, some of these little armies of online sort of social media warriors who kind of go out and like in the very targeted campaigns just simultaneously spread a lot of, on the one hand, disinformation against rival candidates, but then also sort of really kind of nice stories about the candidate. And so when you have things like that happening on a large scale, it’s obviously quite difficult to sort of fact check that and sort of keep on top of the speed of it. So that’s one thing. And then I also think what’s really useful in election time is you should know your stats. You need to have, like, a team of people who know exactly what the figures are when it comes to unemployment, what a candidate’s achievements have been in the past. Candidates love to throw around statistics, percentages, and so the more you know your background on that, the better and the quicker you can react to the fact checking what they’re saying.

MATT ASHBURN

Yeah. Speaking of elections and world events, there’s no doubt in my mind that the recent Kenyan election probably was a hot topic for your team. Can you talk a bit about some of the work that went into that recent event and some of the fact checking that took place there?

NINA LAMPARSKI

Yeah, so the way my team works, I have local reporters spread across the entire continent and we all work together really closely. So when it’s an election in one country, that person doesn’t work in an isolated manner, they work together with the other fact checkers, even if they’re based elsewhere, because misinformation knows no boundaries. So there’s a lot of sort of mudsling that might happen in countries adjacent to Kenya, for example. So collaboration was key. We did bring in an extra person to really just help focus on campaign speeches, for example, to make sure that they could monitor the different candidates. We did see a lot of. So this is a phenomenon that’s recurring in Africa, like fake headlines, screenshots of from pages from newspapers that were doctored or like TV stations tweets being doctored. You can really manufacture those things very, very easily. I could literally show you like in 2 seconds how to take a tweet from Joe Biden. I mean, it’s the easiest thing to do. So we see a lot of that. And I think when people, the better the more good news habits they have, the less likely they will be to take it for granted. But I think unfortunately, in a very fast moving media environment you get something sent on you what’s up? And you’ll be like oh wow, this is crazy. And you just forward it to the contacts in your book. And I think we saw a lot of that. We saw things kind of going viral very quickly on this messaging platform. But yeah, we were afraid also that there could be violence. Like Kenyan elections in previous years has been marked by deadly violence and luckily for us so far things have kind of gone mostly peaceful. But there was a moment where the commission, the electoral commission was a bit slow on releasing the result. And that certainly something that you would have witnessed in the US as well is when there is that space of tension, when results are very, very close, when we’re talking about 2%, there’s a lot of distrust that can fall into that space and disinformation then really can run rampant. And in Kenya we had several electoral monitors from overseas as well as on this international other societies who basically issued statements warning of, you know, not being transparent enough in the way you do your count, for example, not being fast enough and just warning specifically of the threat of online disinformation in election times when, yeah, when things kind of don’t go quite according to plan basically.

MATT ASHBURN

I think that’s an important event and it’s a great example of how fact checking can play a critical role in reporting. It really is an essential foundational element. Now, one thing that comes to mind is that media consumption varies greatly around the world, certainly by region, but also by culture. So for example, in areas that may be more developing in the world or areas where there’s a strong distrust of established institutions, people may not get their news from traditional media sources. They may not go to the newspaper or television station. They may get their news from a WhatsApp group or other types of messaging or word of mouth. And I’m thinking back to an instance at the beginning of the pandemic in March of 2020. I was in Puerto Rico and I was in a WhatsApp group that locals there used to communicate with one another and to share news. And there was, to me at least a very obviously false screenshot of a more established media institution. And the screenshot had a headline and a quick summary that essentially told people to be aware that the government is responding strongly to COVID-19 and to close all of your doors and windows and be in the house by 7 PM because the government is dispatching helicopters to spray some kind of antiviral gas or something like that to eradicate COVID-19. Story was completely false, but actually, I saw firsthand how that can instill a sense of panic in people and how it can affect people’s lives, even though the story was false. And to me it seemed a bit preposterous because of some of the distrust that people there have of established institutions. That story spread like wildfire. Can you speak a bit about how maybe the differences in media consumption can affect your work and sort of, in some cases, I guess, even compete with the work that you’re doing?

NINA LAMPARSKI

Yeah, I mean, that’s a really good example because when I took on that job, it was literally a month before the pandemic broke out. And it was wild because everyone just had to come together and find a way of fact checking medical claims that we weren’t even sure if we didn’t even know, like, what was actually happening. You know, things were very – directives were changing very quickly. We knew so little about how this virus would unfold. So that is definitely the ultimate space, the ultimate environment for stake needs to kind of just erupt. And I guess the way we approached it was one of the functions of the fact checking, of doing fact checking is that you want to educate it without patronizing your public. So we tell people exactly how we get from A to Z, hopefully to Z, sometimes not quite, but, you know, that’s the idea. What tools we’ve used, we have a page on our website that tells you our methodology. It links out to different online tools. So there was this aspect of, first of all, saying, okay, you can go and verify these things by yourself and just maybe next time. The main thing is to just tell people, breathe three or four times before hitting share. But it was quite difficult because speaking from the African point of view, there is a lot of distrust towards Western medicine, towards Western pharmaceutical companies. So we definitely saw all of the US sort of conspiracy theories found the fertile ground also in African countries. I think it’s a fight that you have to take on on various levels. Like, you are collaborating with other fact check organizations. So, unlike, if you just work for The Wire, you will not call up your competitor to tell them about the scoop that you’re having. But if you work for a fact on the fact checking side, there’s so much stuff out there that we know we have to create alliances. So that’s one thing that we did. We teamed up with other fact check organizations in Africa and with the WHO’s branch in Africa. And so we were able to just really stay on top of claims that involved COVID and sort of shared information with each other. So there’s that. And then I think it’s also just making people aware of, like, you have to be really disassembled in the way that you write a fact check, right? You cannot ever let your own beliefs let flow into that. You may have analysis pieces on the wire that will kind of indicate what side you’re on in terms of your media orientation, but for fact checkers, you can’t. So you have to just really verify every single thing that you write. There’s no bias, but it’s an ongoing process. It’s also talking to governments. It’s kind of helping governments to just be more transparent in what they tell people. And then I think the main sentence that we introduced into so many of our debunks at the time was to just say, this is what we know right now, so never to make it like, this is an absolute definite, that’s for sure, for the next 100 years. Currently, this is what research is telling us, but there’s still a lot of unknowns. So that’s kind of what we were trying to do. But it was so fast moving that we learned as we were going, and there’s a lot of conspiracy theorists out there in terms of COVID The one thing I’ve also learned is definitely, if somebody completely rejects the current sort of political system, mainstream media, like the government, they will think there’s this huge conspiracy going on. There’s really nothing you can say, because you are part of that system. So you just have to stick your facts. And I think don’t get involved in any argument. Like, just always be very polite and just say, these are the facts. These are the facts. These are the facts, which drives people nuts. But it kind of works. I haven’t gotten into a fight with anyone.

JEFF PHILLIPS

Well, that’s good.

NINA LAMPARSKI

I could have and I didn’t.

MATT ASHBURN

At least not physical fights, perhaps.

NINA LAMPARSKI

Certainly not a physical fight!

JEFF PHILLIPS

Well, if I continue with that, with the absolute volume of misinformation and questionable claims out there first of all, love that there’s an organization that’s about fact checking and that organizations such yourself have invest in that, but you certainly can’t fact check everything. What’s the process for deciding what to check and go ahead and focus on versus what to just let slide by?

NINA LAMPARSKI

So the key elements that we look for are the potential for harm, the potential for misleading people into not making important decisions or making them on the wrong or on the basis of wrong information. We do look at virality, but something being viral, I really don’t like that word. It’s like, essentially, it doesn’t mean anything. Being viral in Nigeria is very different to something being viral in Ethiopia, being different to being viral in Poland. Like, it just depends on your population. It depends on people’s sort of media consumption habits. And so on. But certainly the key element is to look at a piece of information, say, okay, is somebody here implying something or showing something that could potentially get the user who’s reading this so angry that they would go on the street and, for example, hunt down something? Or would you join a protest or would you not go voting? Because somebody’s posted a video of a polling station and said, look at this violence, like showing clashes. And so people are scared and they don’t go out. So that’s kind of something. That’s how you measure it. You can get very silly fact checks. You can get really people, like create doctored videos of, I don’t know, like a donkey on it, church people. It’s fine to fact check as if everyone believes in it, but that’s not really what we want to do. We’re not there to kill humor or stop people from making jokes or being creative. That’s not the point. The point is really to make sure that this information that can potentially harm people. For example, it’s okay for somebody to say this traditional medicine will treat you asthma. That’s absolutely fine. We’ll help alleviate pains of cancer, that’s absolutely fine. But we just want to tell people it’s not like you can do that, but don’t not go to see a doctor. Essentially, you kind of want to just make sure that somebody understands that there’s always more information than the one they have at hand. So that’s sort of yeah, I think these are the keys. Like sort of what’s the level of harm? And then how viral is it if somebody is saying something really outrageous, but it’s only being shared by 200 people in Nigeria? There’s no point in doing something like that because you’re going to attract more attention to it and you don’t want to push something out into the limelight when it’s currently still quite hidden.

MATT ASHBURN

Yeah, absolutely. I guess in closing, as we start to close out today, do you have any closing words of advice or other information you’d like to share with folks that may be in the journalism field or other investigative roles?

NINA LAMPARSKI

So, a couple of things. Maybe if I can do a little plug, but like, AFP has actually recently launched a ten module course on advanced – basic and advanced fact checking techniques. So it’s all digital investigation techniques. It’s free for anyone who’s a journalist or a journalism student. It was developed with the Google News initiative. It’s really solid. One of my colleagues on my team helped create it. And yeah, it’s really brilliant. It covers anything from basic sort of visual verification to more advanced searches on eyewitness accounts and that kind of stuff. And I would also say, yeah, just really the key element is to never share anything that you’re not sure of. Just take a breather. The stronger your bias, the stronger your emotion. That’s exactly when you should not share it. It’s like when you’re angry, don’t make that call. Go for a walk and then come back. Then speak up like it’s a similar thing. And then for journalists specifically, like, for researchers, I would say, set up the Google alert with your name, clear your cache, log out of all your social media accounts, and do searches for yourself and see what’s out there. What do people know about you? I think that’s really important in terms of your own safety and when you’re out there investigating, spying, instead of just navigating our field.

MATT ASHBURN

That’s fantastic advice, and that’s advice that we’ve given even here on the podcast ourselves many times over. So I appreciate you sharing that. Well, Nina, thank you so much for joining us today. That’s our guest, Nina Lamparski. She’s with AFP, and if you like what you heard, you can subscribe to our show wherever you get your podcasts. You can also watch episodes on YouTube and view transcripts and other information on our website at authentic8.com/needlestack. That’s authentic with the number eight dot com slash needlestack. And be sure to follow us on Twitter @needlestack_pod. And also join us next week. We’ll be back with more on fact checking and debunking. Lots of really good information. We’ll see you then.