Twitter says it’s working on improving how it labels tweets with problematic 5G or coronavirus content, after users reported their tweets were being mislabeled with a COVID-19 fact-check.
“In the last few weeks, you may have seen Tweets with labels linking to additional info about COVID-19,” Twitter Support tweeted. “Not all of those Tweets had potentially misleading content associating COVID-19 and 5G.”
Twitter began fact-checking tweets that linked 5G and the COVID-19 pandemic earlier this month, by adding the label that reads “get the facts about COVID-19” which links to a Twitter moment with “No, 5G isn’t causing coronavirus” as its title. Part of a conspiracy theory that has been widely debunked suggested that the spread of the coronavirus was somehow linked to the installation of new 5G mobile networks.
The fact-check label is part of the social media company’s wider effort to attach warning labels to provide context for tweets with misleading COVID-19 information. In April, the company went so far as to remove misleading COVID-19-related tweets that it viewed as inciting people to engage in “harmful activity.”
But the system that determines which tweets get flagged is apparently a little over-eager. It seems tweets including the words “oxygen” and “frequency” were being tagged with the fact-check label. The Week posits that “oxygen” and “frequency” may be keywords that trigger the label, since part of the conspiracy theory suggests that the 5G “frequency” is harmful to the point that it “sucks the oxygen out of the atmosphere.”
“Labeling Tweets that may contain misleading information continues to be an iterative process,” a Twitter spokesperson said in an email to The Verge. “Given the global spread of misinformation and disputed claims around 5G and COVID-19, we prioritized algorithmically labeling Tweets with that information. As we improve this process to be more precise, our goal is to show fewer labels on unrelated Tweets.”
Twitter Support says it’s “building new automated capabilities to apply these labels to Tweets we think could be relevant,” but it wasn’t clear when the fixes would be applied.
UPDATE June 28th 8:43AM ET: Added statement from Twitter spokesperson