By Daniel Funke, Facebook’s little brother has a fake news problem — and nobody really knows how to solve it.
WhatsApp, the popular messaging platform that hit 1 billion daily users this summer, has been a breeding ground for misinformation in recent months, fact-checkers say. From fake news about last month’s election in Germany and this week’s Catalan independence vote, to rumors about storms in sub-Saharan Africa and fake kidnapping schemes in Brazil, hoaxes are spreading like wildfire in private groups on the platform.
That poses a big problem for both fact-checkers and WhatsApp itself.
“With WhatsApp, you have no idea how many people are reading what you’re putting in there. It’s like a black box,” said Juan Esteban Lewin, a journalist at La Silla Vacía in Colombia — one of the first fact-checking organizations to start debunking hoaxes on WhatsApp. “I’m not so sure that we’re going to be able to stop the flow (of fake news).”
While Facebook and Twitter are highly public spaces where news, photos and videos generally flow freely, WhatsApp is more compartmentalized. Groups are limited to 256 people, making it hard for fact-checkers to see when and where fake news goes viral, and all messages are automatically encrypted. Those problems are compounded by the fact that there’s currently no analytics system that journalists can use to monitor activity on the platform.
Fact-checkers aren’t the only ones having a hard time finding misinformation on WhatsApp, which Facebook purchased for $19 billion in spring 2014. The platform’s own staff has no real way of determining the content of messages that go viral on the platform.
“WhatsApp was designed to keep people’s information secure and private, so no one is able to access the contents of people’s messages,” said Carl Woog, the policy communications lead for WhatsApp, in an email to Poynter.
“We recognize that there is a false news challenge and we’re thinking through ways we can continue to keep WhatsApp safe.”
Woog said one of the ways WhatsApp is trying to combat fake news is by promoting digital literacy on the platform, an initiative that Facebook is helping with (it’s still unclear to what extent the parent company’s data and engineering powerhouses will be used to counter misinformation).
Aside from a simple handout showing users how to detect online hoaxes, what’s striking about WhatsApp’s current efforts to combat fake news is that they’re similar to what fact-checkers are doing: enlisting users.
Since fact-checkers can’t discover and debunk falsehoods on WhatsApp as easily as they can on Facebook or Twitter, they’ve turned to their readers for help. Over the past few months, fact-checking organizations like Chequeado, Africa Check and Boatos.org have bolstered their efforts to combat misinformation on the platform. Several have created institutional WhatsApp accounts to receive and respond to messages about fake news, photos and memes, which users send if they see something suspicious on the platform. In turn, fact-checkers ask readers to share the fact check in the same group as the hoax so as to stem the spread of misinformation.
At least, that’s the goal.
“People are afraid bad things will happen if they say something wrong on social media. I think WhatsApp is a kind of safe place for people and they can discuss more conspiracy things,” said Gülin Çavuş, a fact-checker with Teyit.org in Turkey (Çavuş is a 2017 International Fact-Checking Network fellow).
“We have to find a way to get suspicious news from all echo chambers.”
Laura Zommer, executive director of Chequeado, agreed. While the Argentina-based fact-checking outfit is relatively new to debunking WhatsApp messages, a process that it developed with La Silla Vacía’s guidance, Zommer said it has already run into the challenge of uncovering viral hoaxes and disseminating them in a way that resonates with users.
“It’s more difficult to have access to the source, because they are not sources that we as journalists are used to contacting,” she said. “Journalists must be talking about (what) people are talking about and not necessarily the issues that we think people should talk about.”
Considering the massive popularity of WhatsApp around the world, the problem of fighting fake news is apparent. The messaging platform is fairly ubiquitousoutside the United States — especially in countries where people have little or expensive access to mobile data.
In Brazil, where about 100 million people use WhatsApp, Boatos sends fact checks directly to users. Edgard Matsuki, a journalist at Boatos, told Poynter in an email that he receives about 500 messages per day on his two smartphones — which slows down the platform’s performance. He also has to save contacts manually.
“(One advantage) of WhatsApp is the possibility to have a direct contact with my followers and not depend of algorithms to send my messages,” he said. “I’m still studying some ways to improve my service.”
In Colombia, Lewin said the app is especially popular among middle and lower-class individuals, who came to use it as their primary newsgathering platform after last year’s referendum on a peace deal with the Revolutionary Armed Forces of Colombia (FARC).
“Instead of reading hard news, they were reading whatever they got on WhatsApp, which is huge in Colombia,” he said. “These messaging apps that are not so heavy on the data you’re using are becoming very, very popular.”
Lewin said most of the fake news he’s seen on WhatsApp looks like it was picked up online and reshared on the app, rather than created on the platform. Kate Wilkinson, senior researcher at Africa Check, said most of it isn’t hard to understand, but elicits emotional reactions. However, while WhatsApp-spread misinformation has become an increasing concern across the world — along with efforts to debunk it — the content of hoaxes does vary from region to region.
In Argentina and Colombia, messages are often political, containing misinformation about local and national elections. Last month, Chequeado debunked a meme found on WhatsApp claiming voters could write in votes against animal abuse on the primary election ballot, when in fact that would nullify their vote. In Colombia, Lewin said La Silla Vacía is doing at least one WhatsApp-based fact check per week and has found that the two biggest topics are the FARC and next year’s congressional and presidential elections.
Meanwhile, in sub-Saharan Africa, Wilkinson said most fake news she’s seen isn’t political at all.
“The viral misinformation that we see on WhatsApp is largely messages about some impending danger,” she said. “It’s mainly people passing around messages about crime, violence and severe weather.”
In February, Africa Check debunked a viral storm warning on WhatsApp that claimed a storm off the coast of South Africa was going to hit Johannesburg. In fact, it wasn’t going anywhere near Gauteng, the province in which Johannesburg is located.
The content of messages like those constitute a departure from extreme partisanship, which has become a main focus of fake news and online misinformation over the past few years. Wilkinson said many of the messages users send to her fact-checking organization seem to be shared out of a desire to help and warn people of danger — not to stoke fear or anger.
“WhatsApp is a very intimate form of communication, and if you receive information that leads you to believe you’re at danger … you probably balance the pros and cons of sharing it in your head,” Wilkinson said. “If there’s a chance someone could be hurt, I should probably pass it on.”
But that doesn’t make debunking misinformation on a predominantly private platform any easier. It’s unclear whether or not fact-checkers’ efforts to find and debunk hoaxes on WhatsApp are working, in part because it’s currently impossible to measure.
The fact-checkers interviewed for this article said they hadn’t been in contact with WhatsApp about improving their debunking process. And when they do surface fake news, it’s sometimes discouraging to see what went viral.
“I got a piece of news this week that had already been debunked a year and a half ago,” Lewin said.
However, it’s not all doom and gloom. Lewin said he’s also seen some groups change the way they talk about facts and the news since La Silla Vacía’s presence began on the platform.
“The conversation has changed [a] little bit in some groups, because once you have someone in the group who can say, ‘Okay, let’s stop a moment and check the facts … the level of debate can change,” he said. “There’s no easy way at least to promote our debunking in the platform. We have to rely on this crowdsourcing.”
Until WhatsApp develops a way to monitor messages — which it’s unlikely to do, given its end-to-end encryption — fact-checkers will probably forge ahead with solicited user-submitted occurrences of fake news in order to debunk them. It’s an imperfect solution to a problem of unknown scope on one of the world’s most popular messaging platforms.
“We don’t know how to stop the people from sharing fakes,” Zommer said. “We think, perhaps, we have to try every tool that we can.” – Poynter.org