A Facebook fact-checker speaks
I’ve never been a Facebook user. I have a page where I’ll share my posts sometimes but other than that it’s largely ignored. It’s just not my thing. (Twitter is my social media jones, so I’m not saying that I’m immune from the social media hellscape.)
Anyway, I know that Facebook is a huge resource for tens of millions of people across the world. And this should scare the hell out of us in terms of how propaganda is disseminated on this platform:
Facebook has always struggled to comprehend the scale of its fake news and propaganda problem. Now, it’s struggling to retain the fact-checkers it paid to try and deal with the crisis. Last week both Snopes and the Associated Press ended their partnerships with the social network, after a tense couple of years trying, without success, to tackle the epidemic.
But those partnerships should never have existed in the first place, and I say this as the former managing editor of Snopes, who Facebook first made contact with in 2016. When they first emailed me about a potential partnership, I knew it would bring much more attention to the work of our small newsroom — and much more scrutiny.
But what I didn’t realize was that we were entering a full-blown crisis, not just of “fake news,” but of journalism, democracy, and the nature of reality itself — one we’re all still trying to sort out in 2019, and which had more twists and turns than I’d ever thought possible. Looking back, my overwhelming impression of the years since 2016 is how surreal everything became.
It turned out that trying to fact-check a social media service that is used by a huge chunk of the world’s population is no easy task. We tried to make it easier by showing where disinformation would originate, but there were just too many stories. Trying to stem the tsunami of hoaxes, scams, and outright fake stories was like playing the world’s most doomed game of whack-a-mole, or like battling the Hydra of Greek myth. Every time we cut off a virtual head, two more would grow in its place. My excellent but exhausted and overworked team did as much as we could, but soon felt like we were floating around in a beat-up old skiff, trying to bail out the ocean with a leaky bucket.
Things soon got worse. Because of my own history reporting on refugee rights, I had contacts with groups all over the world working on migration and humanitarian crises. Since early 2015, I’d been hearing bits and pieces about Myanmar and the Rohingya Muslims, and how activists on the ground — exhausted, dispirited activists who were begging any reporter they could find to help spread the word — were saying the crisis had been fueled and spread by social media. The people of Myanmar had only experienced unfettered access to the internet since around 2012, and now Facebook, through its Internet.org program that provided free mobile internet access to its site, had quickly become the only source for news for a large portion of the population. Newsfeeds in Myanmar were pushing a narrative that helped justify ethnic cleansing and other human rights violations on a massive scale. I took it to my editorial team and we put out some stories, and then I took it to Facebook.
Nothing happened, and I came to see Myanmar as something of a model for the damage algorithms and disinformation could do to our world. That’s when the migraines started. I became obsessed with this connection — I dreamed about it at night, woke up thinking about it, and felt responsible for stopping a problem that few others even knew existed.
In case you’re curious, here’s what it was like to be an official Facebook fact-checker. We were given access to a tool that hooked into our personal Facebook accounts and was accessed that way (strike one, as far as I was concerned) and it spat out a long list of stories that had been flagged for checks. We were free to ignore the list, or mark stories as “true,” “false,” or “mixture.” (Facebook later added a “satire” category after what I like to call “the Babylon Bee incident“, where a satirical piece was incorrectly labeled false.)
It was clear from the start that that this list was generated via algorithm. It contained headlines and URLs, and a graph showing their popularity and how much time they had been on the site. There were puzzling aspects to it, though. We would often get the same story over and over again from different sites, which is to be expected to a certain degree because many of the most lingering stories have been recycled again and again. This is what Facebook likes to call “engagement.”
But no matter how many times we marked them “false,” stories would keep resurfacing with nothing more than a word or two changed. This happened often enough to make it clear that our efforts weren’t really helping, and that we were being directed toward a certain type of story — and, we presumed, away from others.
What were the algorithmic criteria that generated the lists of articles for us to check? We never knew, and no one ever told us.
Read the whole thing because it’s clear we have a major problem that nobody has yet figured out how to solve short of simply shutting down the platform — at which point something equally bad, or worse, will take its place.
This is one of the major challenges of our time and I don’t think anyone has a clue about what to do about it.