Fake social media accounts controlled by automated software may be the most effective contributors to the rapid proliferation of COVID-19 misinformation, according to research published June 7 in JAMA Internal Medicine.
The research team identified Facebook groups that were heavily influenced by bots by measuring how quickly the same links were shared in a sample of 299,925 posts made to groups that shared 251,655 links. When links are repeatedly shared by multiple accounts within seconds of one another, it indicates that the posts are being made by bots.
Researchers monitored posts that shared a link to a Danish study published in the Annals of Internal Medicine showing face masks' effectiveness at controlling the spread of COVID-19. Thirty-nine percent of all posts sharing the study were made to Facebook groups the researchers determined were most influenced by bots, whereas 9 percent of posts were made to groups determined to be least influenced by bots.
Posts sharing the study that were made in groups most influenced by bots were more likely to misrepresent the data and make conspiratorial claims. Twenty percent of posts sharing the study that were made in groups most influenced by bots claimed that mask wearing harmed the wearer.
"Bots also appear to be undermining critical public health institutions," Brian Chu, one of the study's authors, said in a news release. "In our case study, bots mischaracterized a prominent publication from a prestigious medical journal to spread misinformation. This suggests that no content is safe from the dangers of weaponized misinformation."