Correcting Bad Information on Social Media-Craig Silvermen

DC Public Safety Radio

See the main site at http://media.csosa.gov

Listen to the radio show at http://media.csosa.gov/podcast/audio/2015/07/social-media-during-emergencies-craig-silverman-buzzfeed/

Leonard: From the nation’s capital, this is DC Public Safety. I’m your host, Leonard Sipes. Ladies and gentlemen, today’s topic: social media during emergencies. Does government have the ability to correct bad information?

I am honored today to have Craig Silverman. He is the founding editor of Buzzfeed Canada. He is at Twitter, @Craigsilverman. To say that this is an extraordinarily important topic is an understatement. Craig wrote a document; Lies, Damn Lies, and Viral Content. It was funded by the Tow Center for Digital Journalism and the Knight Foundation. Throughout this radio show of great importance, and immense complexity, I want us to focus on 3 things: A. How do we address misinformation? B. What can we do about misinformation, and C. If we had a dirty bomb that went off in your area, we practiced this all the time when I was with the Maryland emergency management, what are the implications for public safety and the surrounding area? Craig, welcome to DC Public Safety.

Craig: Thank you so much for having me.

Leonard: I am honored to have you. With the world economic forum in 2014 according to your report, top trends, viral misinformation. The world economic forum is saying viral misinformation is one of the top trends in the world, correct?

Craig: Yes, that’s right. It made their list of the people that they had asked to fill out a survey. People from around the world put that in the top 10.

Leonard: That’s amazing. Misinformation regardless as to it’s an emergency or not, is standard practice in today’s media world. Is that correct?

Craig: Absolutely. When you have a world of decentralized media, where people can easily publish instantly from a smart phone, can easily start to make something gather attention and vitality through social networks, we end up getting a lot of stuff that circulates that simply isn’t true. There have always been rumors. People have always traded information that had a questionable level of voracity, or was in that early emerging stage where you don’t know if it’s true or not. What happens today is these natural human tendencies that we have to pass along information, to share information, it can really go like wildfire because we live in a network society. There’s a huge amount of misinformation that gets out there. It can very easily be seen as true by a lot of people, and once it starts fooling influential folks, such as people in the press, or people in influential positions and government or other places, then it really starts to be seen as true. It’s a big priority, I think, in newsrooms to do a better job of understanding the dynamics of rumor and misinformation, and I think for the folks that you talk to, and in the roles that you’re in, it’s also critically important.

Leonard: It’s compounded during an emergency if we have a hard time correcting bad information on Facebook and Twitter, and the other social media platforms. If it gets into newsrooms, newsrooms end up repeating it. On a day to day basis, that’s a tough nut to crack. When there’s an emergency, if I’m driving in my car, when I was with Maryland Emergency Management, equipped lights and sirens. They say there’s a problem, I have to drive to the problem where we’re setting up a media briefing center, and while driving, and it’s going to take me an hour and a half to get there under the best of circumstances, the rumor goes off that there’s a dirty bomb. I don’t have a snowball’s chance in Hades of putting out that rumor. I mean, I’m driving to the scene. During an emergency, this stuff flows fast and furious.

Craig: Absolutely, it does. Obviously my area of expertise is not in emergency management, but the first thing that comes to mind to me as you’re describing that scenario is there are certainly some physical responsibilities, in person communication, going to scenes, evaluating what’s going on. There’s that piece of it, and I would hope that people involved in this world also think about setting up social media command centers to a certain point, to monitor the information that’s coming out of that specific area where a disaster might have happened, or there are reports of a disaster. Also looking at the larger networks and monitoring key words and other things to see what people are talking about. There’s a two-fold purpose in it. This has some overlap with newsrooms. On the one hand, there were people who are just simply closer to the problem, closer to what’s going on than you are, if you’re in your car driving over, what have you. They may have access to a smart phone, they may be on Twitter, they may be on Instagram. There’s a hug amount of information that could be coming out in real time, from the critical area, or from people who have a certain level of expertise and knowledge. They may be going to social networks to put that out there.

The other side of it, aside from gathering information, you want to use these channels to push out quality information. I think that through both of those things, the monitoring of information, you should be looking at rumors and claims that are out there, and trying to triangulate the information, and compare it to the information that you have coming in from your other sources. On the other side in terms of communicating information, you want to think about, we know this rumor is circulating. We know that it’s true, we know that it’s false, or maybe we don’t know whether it’s true, what can we put out to give some context and to help in the process of people understanding and making sense of the situation. I think in the scenario you described, you’re in the car driving as fast as you can to get there, I would hope that there are colleagues who are really monitoring social media. Not only to gather additional information to provide to you as you give you’re briefings, or what have you, but also to really see if there are things that are starting to take flight that may not be true, or that you really need to look more into to see whether they’re true or not.

Leonard: Even when it’s not during an emergency, I would contend that most of us in emergency management throughout the United States, most of us within the criminal justice system throughout the United States, we do not have these special I-teams to do that analysis. The question becomes whether or not we are all sophisticated enough and do we have our own personal social media accounts, and do we have enough of them? It may be Twitter, it may be Facebook, but it could be an endless number of others that are putting out this misinformation. It would take somebody savvy, it would take somebody who already has social media accounts, and it would take an almost instantaneous network to begin to compare notes, before we could put out these fires.

Craig: Yeah. This is one of the things that it goes to a point that’s really, really important. I worked on a project with the European Journalism Center, something called the verification handbook, where it was a document oriented towards newsrooms, but also humanitarian workers, helping them verify information in real time in emergency scenarios. One of the things that became very clear is as we talked to journalists with the Australian broadcasting Corporation, she talked about their experience with wildfires in Australia. I’m sure that this probably isn’t news to you, but one of the things that became very clear from talking to her is what you do before the emergency happens dictates how well you’re going to handle it and cover it in the moment.

In these scenarios, rather than trying to figure out who on the team has got social media accounts or what have you, obviously it has to become part of the planning process. What are the channels we’re going to use? What are the ones that are most important to use in terms of getting news out in real time? I would say that Twitter is a very important one, because that’s where people tend to look for real time information.

Leonard: Yeah.

Craig: Facebook is obviously where the most amount of people are. Facebook is doing a lot of work to be seen as more real time, and more hospitable to real time, so a Facebook page is also really important. I think setting this stuff up ahead of time of, where are we going to communicate this information, who is going to own these channels, and what are we going to use these channels for? Those are really, really important things to think about much sooner than before you get the reports that a dirty bomb has gone off. Figuring out who on the team has expertise in this area, figuring out what accounts exist or need to be set up, and then how you’re going to use those. Those are just as important as all the other pieces of preparation that are going to be done, because not only is it important for, again, getting the information out, but if people know this is a trusted source, and a place where good information is coming from, then they may actually bring intelligence and information to you, and it becomes a place where people can start to contribute. It again gets that kind of two-way thing going, which is really, really important and useful.

Leonard: I think you would agree that there’s no piece of technology out there that that’s going to solve this. I’ve been also looking at magazines and different companies are offering emergency social media analysis hardware/software. I don’t think this is a matter of hardware/software. I think this is a matter of, as you said before, preparation, having trained people in place, and the ability to instantaneously sit at a computer and analyze social media and instantaneously contact each other. This is something that’s not going to involve one or two or three people, this is something that’s going to involve in some cases, up to 20 or 30 people who can instantaneously drop what they’re doing, go to the computer, start searching hash tags, start sharing information with each other. That involves a pretty high degree of sophistication and preparation and technology that they have to have with them practically at all times.

Craig: Yeah. You need an internet connection, and things like say Tweet Deck or Hoot Suite, or things like that in terms of the tools, but there is a lot of it that goes down to training and expertise. Just a lot of it that is the human factor. You need to have access and basic knowledge of the tools, but you also need to have people who are well trained. I think actually a small group of people can achieve a lot. The larger you get, once you get beyond say five people, it becomes very hard to coordinate those folks. You might have people duplicating effort in the scenario of monitoring social media and analyzing it. Ideally, you have some specialists in this area who can get on that wan watch for it. I think there’s the technology piece, but what you bring up is kind of the human piece of it, and that’s really, really important when we’re talking about rumors and misinformation.

There are some really basic human needs that are filled by rumors. That’s why we have so many of them, particularly in emergency and disaster scenarios. When humans lack a certain amount of information, when it’s a very confusing scenario, and there’s lots of conflicting information, what we try to do, is we try to make sense of the world. It makes us very uncomfortable to not have information, especially in a critical scenario. It’s very tough for our brains to process conflicting information. What we naturally try to do is to make sense of this scenario, and that often causes us as we talk with other people to come up with, “well, maybe it’s because of this, or maybe it’s because of that.” We all put our pieces of information together, and that’s where we start to create and propagate rumors. It’s important for everyone to understand that there is this human need that rumors can often fill, especially in emergency scenarios, where there’s a real, imminent threat there. Understand that rumors aren’t necessarily people who are trying to put out false information, it’s people who mean well who are engaging in this process of sense-making, trying to figure it out.

Especially when it’s hurricanes or things like that, and there’s high anxiety. It actually is a coping mechanism in a lot of ways for us to fill in the gaps, the things that we don’t have, the information we don’t have, to put that out there. You’ll see this happening on Twitter and on social media, is people asking questions, interacting with each other, latching on to little scraps of information that come out that seem to make sense to them that they then propagate. I think that everyone should have a basic understanding of why rumor is such a basic human element, particularly in these scenarios. Whenever you have things like bombings, or hurricanes, natural disasters, they’re going to be there. It’s the human engine, emotions and brains and those kinds of things that are driving these rumors. Tools and technologies are important, but understanding human behavior is always really, really big in this kind of scenario.

Leonard: There’s also sources of purposeful misinformation. One of the things when I was learning how to do green screen television, is that I realized that I could buy readily available footage from elsewhere, I can do a green screen television shoot, and it will look exactly like any other news program. It will look and feel and smell and taste like a real news program, and I can purposely put out misinformation, and sites purposefully putting out misinformation. I refer to the photos of sharks swimming in the streets of Sandy Hook after Hurricane Sandy. We have that level of a complexity to deal with as well.

Craig: Yes, absolutely. One of the things in the research that I focused on were what I called fake news websites. These are websites where somebody’s taken a basic WordPress template or web template and it looks like a real news website. The articles are written with a newsy voice and tone to them, but everything on the website is fake. What I saw in my research was that they could have articles that could get hundreds of thousands of shares, driving a significant amount of traffic. What their strategy basically is, they’re trying to monetize on gullibility. They come up with fake articles about celebrities, or about what I saw when Ebola was a real threat in the United States, they did a lot of fake articles. One of them reported that an entire Texas town had been quarrantined because a family had contracted Ebola. Completely fake, of course. It got a huge amount of shares. Of course, that sends people to their website, they have ads on their pages, and they earn money that way.

There are certainly people who are conscientiously trying to spread misinformation, whether it’s fake news websites doing it to earn money, or perhaps there are people who have a malicious intent and other ways. That’s certainly something that’s going to emerge and come out there. Sometimes though, people are spreading fake information, but it’s not necessarily with malicious intent. This is a really hard thing particularly for journalists to understand, because why would somebody put out something that was fake? The answer that a lot of these hoaxers give, is that again, it’s a stress relief for them to just put a joking image. Unintentionally they put it out thinking everyone will know it’s a joke and have a laugh, but people start to take it and treat it as real.

Again, there’s this release valve that people need in these very anxiety inducing scenarios where they’ll often put something out like that. It is something to be aware of. It’s important to be aware that rumors are absolutely going to emerge in these scenarios. A hundred percent guaranteed, in an emergency response scenario, natural disaster scenario, what have you, rumors will abound. No question about it. There will also be people who intentionally or otherwise put out misinformation.

One is the monitoring aspect of this. The second piece where journalists also really need to raise their skill level, is the verification piece of it. You see a tweet, and somebody’s made a claim, how do you figure out if that’s true or not? This is a skill area where the more people who can know how to use some tools and some basic approaches to figure out whether it’s true or false, the better off we’re all going to be as a society.

Leonard: There’s a fascinating part of your report where basically you’re saying that there’s an economic model for anything that delivers clicks to a website and that the incentives are all wrong, which is one of the reasons why we’re having this problem to begin with.

We’re more than halfway through an extraordinarily entertaining and informative program. Ladies and gentlemen, Craig Silverman is by our microphones. Founding editor of Buzzffed Canada. You can reach him a Twitter @craigsilverman. The report itself, Lies, Damn Lies and Viral Content, again funded by the Tow Center for Digital Journalism and the Knight Foundation. Extraordinarily fascinating read. Find it on the internet if you are interested in rumor control, if you’re interested in emergency management, you must go to the website and get Lies, Damn Lies and Viral Content, and read it.

Craig, we have a 30% reduction in reporters over the last 10 years. Where we’ve relied upon media in the past to assess rumors and correct information internally, their ranks have been depleted tremendously. The reporters who I work with are doing more than ever before with less than ever before. The media that we counted on in the past to verify and to figure out what is correct information and what is incorrect information with our help from people who are directly at the scene. Because of their fewer numbers, it’s very hard for them to complete this task. It’s very hard for them to be the gatekeeper.

Craig: It is. The reason I do things like the verification handbook and the Lies, Damn Lies report, is because I see that there’s a lack of these skills in the newsroom. Of the journalists that are left, they absolutely are overloaded with different things, and they haven’t been given training in verification of social media content and things like that. I think you do have to in these scenarios, look at newsrooms as an important channel and partner in these things. I also do think that you have to look at the reality that you just articulated, and say we are going to be a source as well. We’re not just going to feed things through the media. We have to be able to be a credible, reliable, consistent source ourselves. I think that’s a really important thing to realize and to think about what that’s going to take in terms of resources and training on your own.

All that being said, a piece of advice we give to newsrooms in terms of preparing for disaster coverage is you need to think about all of the critically important government agencies, first responders, experts in your area and you need to create a line of communication with them ahead of time. I think it goes both ways in that, absolutely you should be thinking about who are the newsrooms, AP being one that’s all over the country, so how do we get a direct line into AP to make sure we’re communicating with them well. Thinking about locally, who are the critical local sources as well, and setting up those lines of communication to say when we have something important, we’re going to get it to you in this form. For the journalists, to tell them and say when you have an important question, when you know something, here is how you get that to us. I think when we’re talking about these kinds of scenarios of emergency situations, some of the mutual suspicion or distrust that tends to be there, it recedes a little bit because everybody’s in it together just trying to get the best possible information out to the public as possible. To tell the people who need to evacuate that they need to evacuate, and to make sure that it’s not a false order. And so on.

I think that you’ll find in these scenarios, newsrooms do want to be a very good partner. Figuring out the method of communication, helping them understand what kind of information they can rely on from your particular office, and which other places they might need to go for other information is very important. Then of course thinking about how you are going to get your information, not just out through the media, but through like we talked about; a Twitter account, a Facebook page, other means to make sure that it’s getting out there as much as possible.

Leonard: Bottom line is that we have to have a core of digital specialists who are extremely sophisticated about social media. Extremely sophisticated in terms of following the media, who already have these accounts set up, who are ready to go at a moment’s notice. What you’re saying is, is that we can do it with a small number. Let’s just say 1-10, we have to be digital specialists. We have to really know social media. We have to know its implications, we have to know who’s out there, and we have to be on the various platforms. I agree with you with Facebook and Twitter, but Instagram is rising in popularity more than anybody ever anticipated, and others. Periscope. There’s all sorts of things out there that are just exploding and we have to be knowledgeable of them all. We in the media, and we within government, and we within emergency management must become digital media experts.

Craig: Absolutely. You have to think like a newsroom in some ways, particularly when there’s an active scenario going on. You have to think about how you’re communicating the information. One of the other pieces that’s really important and this is talked about in the report, is deciding at what point you’re ready to communicate a piece of information.

Newsrooms, what I was talking about in the sense of there’s so much information that’s circulating online and circulating on social networks, and it may have news value and it may be of interest, and it may get them clicks, but what newsrooms have to figure out is, what’s your bar for when you’ll actually cover something? Do you need to have it 100% nailed down? Or will you just take anything that’s circulating and put some [hedging 00:21:58] language in, and saying, “Well, this is popular on Reddit, we don’t know if it’s true, but have a look at this photo.” I think it’s important for government agencies and communicators as well to think about that. Okay, so if we’re going to be a source of information in this scenario, what level of voracity, what level of conformation do we need to put something out? Are we only going to put out stuff that is 100% nailed down, or are we actually going to engage and say, “There’s a rumor circulating that this is happening in this area of the city, as of right now we have no information to confirm that.”

Thinking about how you’re going to engage on that level is I think a really important thing. Overall, yeah. I think that there’s no way to do this kind of work today without having people on your team who are social media savvy, who are good at monitoring, who are good at assessing, who are good at verifying, and who are good at communicating. These are absolutely core skills. There are definitely tools and things that can help you, but a lot of it comes down to human decision making and figuring out what you’re processes are and what you’re standards are, just like newsrooms have to do.

Leonard: Even in your report, when you mentioned Larry King, he criticized CNN his own network in terms of Flight 370, the Indonesian airliner that went down. He criticized their coverage of that as absurd. We say in the report that there is an economic model, that anything that drives clicks to a website, the incentives, fiscal incentives, financial incentives are all wrong. This is a challenge. This is a challenge for media, it’s a challenge for us to put it together, and guess what you’re saying is, is that not only must we train amongst ourselves, we’ve got to get together with media and figure all this out ahead of time, and have a protocols in place so bad information doesn’t get out information that’s going to harm literally thousands if not hundreds of thousands of people.

Craig: Absolutely. Yeah. That preparation piece. We keep coming back to it because it’s so essential. For the folks that this particular conversation is targeted at, I don’t know that their incentives are as misaligned as they are in a lot of newsrooms in the sense that a lot of digital newsrooms value the number of views and clicks you get, and if you take the extra time to nail something down and find out that it’s not real, then you get no clicks for that, because you haven’t gotten the story. Other people may get it first. In newsrooms, what they have to re-calibrate is what do we value? Do we value getting the most clicks on a story, or do we value being the ones who’ve said, “No, no, no. This story is fake,” and putting that out there and showing that we’re trusted and building that up over time.

I work at BuzzFeed now. I Joined BuzzFeed after I did the fellowship that led to the report we’re talking about. For me. BuzzFeed obviously is a huge organization, drives a huge amount of traffic, covers a huge amount of stuff that’s viral. What I saw before I joined, is a shift that had been happening culturally internally at BuzzFeed over the last couple of years, whereby rather than just finding anything and being first to it and getting it up first, the culture had shifted towards asking questions, reaching out, doing verification, and not having fake stuff on the site. A more journalistic culture has taken hold, and BuzzFeed as they’ve been hiring people and news.

I don’t know exactly what the incentives are in the world of the folks who are here, but I assume that there’s probably not as much pressure around clicks, and far more to be lost in terms of reputation if a government agency or a specialized company doing emergency response is putting out fake information. That is a game changer for them, whereas a single journalist, if they make 1 mistake they can move on. For a government agency putting out false information in an emergency, that’s almost game over, because of the amount of credibility it has.

Leonard: In the final minutes of the program, and there’s so much I wanted to get to but I’m not going to be able to get to it today. The more we care about a rumor, the more we have a stake, the more we participate in that rumor, the more we care about a topic, the more like we are to spread rumors, and the more likely we are to believe it’s true. We cherry pick information that we hear, that we come into contact with, and if it fits our preconceived notion of the world, there’s a good possibility that we will spread that rumor. We will believe it and spread it. Part of this is the psychology of the people who are reading and assessing this for themselves, and what you’re warning is that you’ve got to be very careful in debunking the rumor. You cant go after the person, you have to to go after the facts. There is a strong set of psychological principles that apply.

Craig: Yes. There’s a lot going on here in terms of why somebody would choose to propagate a rumor. If it aligns with existing knowledge and existing beliefs that they have, they are more likely to believe it and to propagate it. If it fits with suspicions they have, if it fits with their worldview, again they’re more likely to put it out there. They’re more likely to believe it. Another thing that people should note also about rumors, there is often a connection between repetition and believability. The more that someone is exposed to rumor, the more that any skepticism they might have about it starts to erode.

There have been studies about the number of times somebody was exposed to a rumor they start to believe it even more.

Leonard: That’s amazing.

Craig: It is. It just shows how important it is that so many of us news organizations, government communicators and other communicators, how we understand that we have to get out there early. When we see things, we have to talk about it and engage and warn people off of this stuff.

Then that gets us into the realm of debunking that you brought up. Which is, how do you do this effectively without in effect, repeating the rumor so much that people still ignore your debunking. One, as you mentioned, is people who pushed out a false rumor, you don’t want to attack them, you don’t want to personalize it. You want to debunk the idea and not the person is what people often say. That’s very important. Don’t go around shaming people. Make it easy for them to let go of this thing that they put out there and that they believed. It’s also really important to try to minimize the amount of times you’re repeating the false information. You want to express the truth in a more positive way.

A small example of this, rather than saying Barack Obama is not a Muslim, you would want to say Barack Obama is a Christian. That’s the more positive reinforcement of the correct information. It’s also important of course to get out there early. It’s important to do it in a positive way, to not attack people. It’s important to think about how you can connect with other trusted sources to get this information out there. What I mean by that is, if somebody is propagating a rumor because it aligns with say a political belief, a personal belief, some kind of orientation. If you can get other organizations that person might be perceived to be as aligned with their beliefs, and have that organization help you with pushing out the debunking, then you’ve got a much better chance of getting it out there. If they have suspicion about the government, then you should talk to other local organizations that aren’t seen as government organizations to get them to push it out there as well, whether that’s a red cross, or a local chamber of commerce. Think about all these different channels, how people who might be skeptical of government, might actually listen to one of these other channels.

Leonard: All right. Craig, you’ve got the final word. There’s so much here to discuss. Ladies and gentlemen, if you’re interested in emergency management, if you’re interested in relations with the media, if you’re interested in rumor control, please go to Lies, Damn Lies and Viral Content, easily found on the internet. It is one of the most fascinating reads that I’ve had within my 40 years within the criminal justice system. We’ve talked to Craig Silverman today. Twitter @craigsilverman, founding editor at BuzzFeed Canada. Ladies and gentlemen, this is DC Public Safety. We appreciate your comments. We even appreciate your criticisms, and we want everybody to have themselves a very pleasant day.

Share

Speak Your Mind

*

%d bloggers like this: