Stay on Top of Enterprise Technology Trends
Get updates impacting your industry from our GigaOm Research Community
If you’re driven bonkers by the fake viral news stories that proliferate on your Facebook feed, usually posted by that girl you went to high school or your excitable uncle in Vermont, it’s your lucky day. Instead of you having to be that jerk who comments with the Snopes retraction, Facebook will start weeding out fake posts itself. Kind of.
The company has altered its algorithm so that hoaxes and scams appear less in people’s feeds, which ultimately limits their spread.
Facebook is crowdsourcing the determination of what counts as a “fake” news story. It added a flagging button that allows users to report whether a story is fake. It’s algorithm will also collect information on which stories people are deleting after posting, since that’s an indicator of a hoax. In addition to lessening a scam’s appearance in people’s feeds, Facebook will occasionally stick a warning sentence on top of the post.
It was quick to say it wasn’t asserting any editorial control. “We are not removing stories people report as false and we are not reviewing content and making a determination on its accuracy,” a Facebook product manager wrote in the blog post announcing the change.
What’s the difference between editorial judgment and newsfeed algorithm judgment? In a day and age when we get our content increasingly from Facebook’s newsfeed instead of newspapers or website home pages, they’re arguably coalescing. Like it or not, the company has become the number one social director of traffic.
Although Facebook may just be aggregating content created by others, it’s acting in an editorial role by making the decision that fake news isn’t as valuable as real news. It’s just putting its algorithm — and users — in charge of the editorial execution. Facebook, like Medium, is playing with the gray area between platform and publisher.
It claims that satire websites will be spared Facebook’s newsfeed wrath because people are less likely to report them as hoaxes or delete their articles after sharing them. But as Mike Isaac mentioned on Twitter, the people Facebook is relying on for this judgment are the same population whose favorite passwords are “password” and “123456.”