Efforts to remove anti-vaccine content are not yielding results

Efforts by the social network Facebook to eliminate anti-vaccine content from its platform during the worst months of the covid-19 pandemic did not reduce its users’ exposure to this type of publication, according to a study published this Friday in the journal Science Advances.

Researchers at George Washington University in Washington, DC, downloaded public data from the platform for different dates (before the vaccine disinformation policy began, in December 2020, and after) and saw that the level of user interaction with this type of post not only did not decrease, but in some cases increased.

In a statement, one of the researchers, Lorien Abroms, explained that their results show how difficult it is to eliminate misinformation about health problems from public spaces.

Their main hypothesis was that Facebook’s own architecture, specifically its systems designed to promote group creation and information sharing, caused this increase in interactions with anti-vaccine content, despite the platform’s efforts to eliminate it.

“Individuals who are highly motivated to find and share anti-vaccine content are simply using the system as designed,” David Broniatowski, the study’s lead author, said in a statement.

Their conclusion is that removing content or changing algorithms is useless if the main goal of the platform is not modified, which is to connect different people who share common interests, in this case the fear of vaccines.

The researchers emphasized that this is the first scientific study to examine the effectiveness of Facebook’s efforts to eliminate misinformation from its platform.

According to the university, the results suggest that social media companies can mitigate these harmful effects by working together to create “building codes” for their apps, based on science, in the same way that architects need to ensure the safety of their buildings including ventilation, emergency exits and other measures.